Large language models (LLMs) have shown promising abilities as cost-effective and reference-free evaluators for assessing language generation quality. In particular, pairwise LLM evaluators, which compare two generated texts and determine the preferred one, have been employed in a wide range of applications. However, LLMs exhibit preference biases and worrying sensitivity to prompt designs. In this work, we first reveal that the predictive preference of LLMs can be highly brittle and skewed, even with semantically equivalent instructions. We find that fairer predictive preferences from LLMs consistently lead to judgments that are better aligned with humans. Motivated by this phenomenon, we propose an automatic Zero-shot Evaluation-oriented Prompt Optimization framework, ZEPO, which aims to produce fairer preference decisions and improve the alignment of LLM evaluators with human judgments. To this end, we propose a zero-shot learning objective based on the preference decision fairness. ZEPO demonstrates substantial performance improvements over state-of-the-art LLM evaluators, without requiring labeled data, on representative meta-evaluation benchmarks. Our findings underscore the critical correlation between preference fairness and human alignment, positioning ZEPO as an efficient prompt optimizer for bridging the gap between LLM evaluators and human judgments.

Top-view perspective denotes a typical way in which humans read and reason over different types of maps, and it is vital for localization and navigation of humans as well as of ‘non-human’ agents, such as the ones backed by large Vision-Language Models (VLMs). Nonetheless, spatial reasoning capabilities of modern VLMs in this setup remain unattested and underexplored. In this work, we study their capability to understand and reason over spatial relations from the top view. The focus on top view also enables controlled evaluations at different granularity of spatial reasoning; we clearly disentangle different abilities (e.g., recognizing particular objects versus understanding their relative positions). We introduce the TopViewRS (Top-View Reasoning in Space) dataset, consisting of 11,384 multiple-choice questions with either realistic or semantic top-view map as visual input. We then use it to study and evaluate VLMs across 4 perception and reasoning tasks with different levels of complexity. Evaluation of 10 representative open- and closed-source VLMs reveals the gap of more than 50% compared to average human performance, and it is even lower than the random baseline in some cases. Although additional experiments show that Chain-of-Thought reasoning can boost model capabilities by 5.82% on average, the overall performance of VLMs remains limited. Our findings underscore the critical need for enhanced model capability in top-view spatial reasoning and set a foundation for further research towards human-level proficiency of VLMs in real-world multimodal tasks.

Large pretrained language models are widely used in downstream NLP tasks via task- specific fine-tuning, but such procedures can be costly. Recently, Parameter-Efficient Fine-Tuning (PEFT) methods have achieved strong task performance while updating much fewer parameters than full model fine-tuning (FFT). However, it is non-trivial to make informed design choices on the PEFT configurations, such as their architecture, the number of tunable parameters, and even the layers in which the PEFT modules are inserted. Consequently, it is highly likely that the current, manually designed configurations are suboptimal in terms of their performance-efficiency trade-off. Inspired by advances in neural architecture search, we propose AutoPEFT for automatic PEFT configuration selection: We first design an expressive configuration search space with multiple representative PEFT modules as building blocks. Using multi-objective Bayesian optimization in a low-cost setup, we then discover a Pareto-optimal set of configurations with strong performance-cost trade-offs across different numbers of parameters that are also highly transferable across different tasks. Empirically, on GLUE and SuperGLUE tasks, we show that AutoPEFT-discovered configurations significantly outperform existing PEFT methods and are on par or better than FFT without incurring substantial training efficiency costs.

Traditional benchmarking in NLP typically involves using static, held-out test sets and calculating aggregated statistics based on diverse examples. However, this approach often results in an overestimation of performance and lacks the ability to offer comprehensive, interpretable, and dynamic assessments of NLP models. Recently, works like DynaBench and Checklist have addressed these limitations through behavioral testing of NLP models with test types generated by a multi-step human-annotated pipeline. Unfortunately, manually creating a variety of test types requires significant human labor, thus weakening efficiency. In this work, we propose SynthEval, a hybrid behavioral testing framework that leverages large language models (LLMs) to generate a wide range of test types for a comprehensive evaluation of NLP models. The SynthEval framework first generates sentences via LLMs using controlled generation, and then identifies challenging examples by comparing the predictions made by LLMs with task-specific NLP models. In the last stage, human experts investigate the challenging examples, manually design templates, and identify the types of failures the task-specific models consistently exhibit. We apply SynthEval to two classification tasks and show that our framework is effective in identifying weaknesses of strong models on these tasks.

Multiple choice question answering tasks evaluate the reasoning, comprehension, and mathematical abilities of Large Language Models (LLMs). While existing benchmarks employ automatic translation for multilingual evaluation, this approach is error-prone and potentially introduces culturally biased questions, especially in social sciences. We introduce the first multitask, multiple-choice Turkish QA benchmark, TurkishMMLU, to evaluate LLMs’ understanding of the Turkish language. TurkishMMLU includes over 10,000 questions, covering 9 different subjects from Turkish high-school education curricula. These questions are written by curriculum experts, suitable for the high-school curricula in Turkey, covering subjects ranging from natural sciences and math questions to more culturally representative topics such as Turkish Literature and the history of the Turkish Republic. We evaluate over 20 LLMs, including multilingual open-source (e.g., Gemma, Llama, MT5), closed-source (GPT 4o, Claude, Gemini), and Turkish-adapted (e.g., Trendyol) models. We provide an extensive evaluation, including zero-shot and few-shot evaluation of LLMs, chain-of-thought reasoning, and question difficulty analysis along with model performance. We provide an in-depth analysis of the Turkish capabilities and limitations of current LLMs to provide insights for future LLMs for the Turkish language. We publicly release our code for the dataset and evaluation: https://github.com/ArdaYueksel/TurkishMMLU

Instruction tuning enables language models to more effectively generalize and better follow user intent. However, obtaining instruction data is costly and challenging. Prior work employs methods such as expensive human annotation, crowd-sourced datasets with alignment issues, and generating noisy examples via LLMs. We introduce the LongForm-C dataset, which is created by reverse instructions. We generate instructions via LLMs for human-written corpus examples using reverse instructions. First we select a diverse set of human-written documents from corpora such as C4 and Wikipedia; then we generate instructions for these documents via LLMs. This approach provides a cheaper and cleaner instruction-tuning dataset with natural output and one suitable for long text generation. Our models outperform 10x larger language models without instruction tuning on tasks such as story/recipe generation and long-form question answering. Moreover, LongForm models outperform prior instruction-tuned models such as FLAN-T5 and Alpaca by a large margin, and improve language understanding capabilities further. We publicly release our data and models: [Anonymized-URL].

Human label variation (HLV) is a valuable source of information that arises when multiple human annotators provide different labels for valid reasons. In Natural Language Inference (NLI) earlier approaches to capturing HLV involve either collecting annotations from many crowd workers to represent human judgment distribution (HJD) or use expert linguists to provide detailed explanations for their chosen labels. While the former method provides denser HJD information, obtaining it is resource-intensive. In contrast, the latter offers richer textual information but it is challenging to scale up to many human judges. Besides, large language models (LLMs) are increasingly used as evaluators (“LLM judges”) but with mixed results, and few works aim to study HJDs. This study proposes to exploit LLMs to approximate HJDs using a small number of expert labels and explanations. Our experiments show that a few explanations significantly improve LLMs’ ability to approximate HJDs with and without explicit labels, thereby providing a solution to scale up annotations for HJD. However, fine-tuning smaller soft-label aware models with the LLM-generated model judgment distributions (MJDs) presents partially inconsistent results: while similar in distance, their resulting fine-tuned models and visualized distributions differ substantially. We show the importance of complementing instance-level distance measures with a global-level shape metric and visualization to more effectively evaluate MJDs against human judgment distributions.

Are Large Language Models (LLMs) temporally grounded? Since LLMs cannot perceive and interact with the environment, it is impossible to answer this question directly. Instead, we provide LLMs with textual narratives and probe them with respect to their common-sense knowledge of the structure and duration of events, their ability to order events along a timeline, and self-consistency within their temporal model (e.g., temporal relations such as after and before are mutually exclusive for any pair of events). We evaluate state-of-the-art LLMs (such as LLaMA 2 and GPT-4) on three tasks reflecting these abilities. Generally, we find that LLMs lag significantly behind both human performance as well as small-scale, specialised LMs. In-context learning, instruction tuning, and chain-of-thought prompting reduce this gap only to a limited degree. Crucially, LLMs struggle the most with self-consistency, displaying incoherent behaviour in at least 27.23% of their predictions. Contrary to expectations, we also find that scaling the model size does not guarantee positive gains in performance. To explain these results, we study the sources from which LLMs may gather temporal information: we find that sentence ordering in unlabelled texts, available during pre-training, is only weakly correlated with event ordering. Moreover, public instruction tuning mixtures contain few temporal tasks. Hence, we conclude that current LLMs lack a consistent temporal model of textual narratives.

Task-oriented dialogue (TOD) systems help users execute well-defined tasks across a variety of domains (e.g., flight booking or food ordering), with their Natural Language Understanding (NLU) components being dedicated to the analysis of user utterances, predicting users’ intents (Intent Detection, ID) and extracting values for informational slots (Value Extraction, VE). In most domains, labelled NLU data is scarce, making sample-efficient learning – enabled with effective transfer paradigms – paramount. In this work, we introduce SQATIN, a new framework for dialog NLU based on (i) instruction tuning and (ii) question-answering-based formulation of ID and VE tasks. According to the evaluation on established NLU benchmarks, SQATIN sets the new state of the art in dialogue NLU, substantially surpassing the performance of current models based on standard fine-tuning objectives in both in-domain training and cross-domain transfer, and it also surpasses off-the-shelf large language models for the same task, both in terms of performance and inference efficiency. Furthermore, SQATIN yields particularly large performance gains in cross-domain transfer, owing to the fact that our QA-based instruction tuning leverages similarities between natural language descriptions of classes (i.e., slots and intents) across domains.

We present DIALIGHT, a toolkit for developing and evaluating multilingual Task-Oriented Dialogue (ToD) systems which facilitates systematic evaluations and comparisons between ToD systems using fine-tuning of Pretrained Language Models (PLMs) and those utilising the zero-shot and in-context learning capabilities of Large Language Models (LLMs). In addition to automatic evaluation, this toolkit features (i) a secure, user-friendly web interface for fine-grained human evaluation at both local utterance level and global dialogue level, and (ii) a microservice-based backend, improving efficiency and scalability. Our evaluations reveal that while PLM fine-tuning leads to higher accuracy and coherence, LLM-based systems excel in producing diverse and likeable responses. However, we also identify significant challenges of LLMs in adherence to task-specific instructions and generating outputs in multiple languages, highlighting areas for future research. We hope this open-sourced toolkit will serve as a valuable resource for researchers aiming to develop and properly evaluate multilingual ToD systems and will lower, currently still high, entry barriers in the field.

Cross-lingual transfer has recently been tackled through modular, parameter-efficient fine-tuning methods which allow arbitrary combinations of language and task modules for transfer of any task to any language. Concurrently, task arithmetic has emerged as a powerful and modular tool for editing pretrained models using multiple full fine-tunings. In this work, we connect the paradigms of task arithmetic and cross-lingual transfer, demonstrating that modularity for cross-lingual transfer can be achieved even with full model fine-tuning. Our approach displays strong performance on a range of multilingual benchmarks encompassing both high-resource and low-resource languages.

Recent work has shown that, while large language models (LLMs) demonstrate strong word translation or bilingual lexicon induction (BLI) capabilities in few-shot setups, they still cannot match the performance of ‘traditional’ mapping-based approaches in the unsupervised scenario where no seed translation pairs are available, especially for lower-resource languages. To address this challenge with LLMs, we propose self-augmented in-context learning (SAIL) for unsupervised BLI: starting from a zero-shot prompt, SAIL iteratively induces a set of high-confidence word translation pairs for in-context learning (ICL) from an LLM, which it then reapplies to the same LLM in the ICL fashion. Our method shows substantial gains over zero-shot prompting of LLMs on two established BLI benchmarks spanning a wide range of language pairs, also outperforming mapping-based baselines across the board. In addition to achieving state-of-the-art unsupervised BLI performance, we also conduct comprehensive analyses on SAIL and discuss its limitations.

Developing imaging models capable of detecting pathologies from chest X-rays can be cost and time-prohibitive for large datasets as it requires supervision to attain state-of-the-art performance. Instead, labels extracted from radiology reports may serve as distant supervision since these are routinely generated as part of clinical practice. Despite their widespread use, current rule-based methods for label extraction rely on extensive rule sets that are limited in their robustness to syntactic variability. To alleviate these limitations, we introduce RadPert, a rule-based system that integrates an uncertainty-aware information schema with a streamlined set of rules, enhancing performance. Additionally, we have developed RadPrompt, a multi-turn prompting strategy that leverages RadPert to bolster the zero-shot predictive capabilities of large language models, achieving a statistically significant improvement in weighted average F1 score over GPT-4 Turbo. Most notably, RadPrompt surpasses both its underlying models, showcasing the synergistic potential of LLMs with rule-based models. We have evaluated our methods on two English Corpora: the MIMIC-CXR gold-standard test set and a gold-standard dataset collected from the Cambridge University Hospitals.

Language-based biomarkers have shown promising results in differentiating those with Alzheimer’s disease (AD) diagnosis from healthy individuals, but the earliest changes in language are thought to start years or even decades before the diagnosis. Detecting these changes is critical to allow early interventions, but research into the earliest signs is challenging, as it requires large longitudinal datasets that are time-consuming and expensive to collect. There is a need for alternative methods for tracking longitudinal language change, including Natural Language Processing (NLP) and speech recognition technologies. We present a novel corpus that can enable this: a corpus of transcripts of public interviews with 20 famous figures, half of whom will eventually be diagnosed with AD, recorded over several decades. We evaluate the corpus by validating patterns of vocabulary richness changes known from literature, such as decline in noun frequency, word length, and several other features. We show that public data could be used to collect longitudinal datasets without causing extra stress for the participant, and that these data can adequately reflect longitudinal AD-related changes in vocabulary richness. Our corpus can provide a valuable starting point for the development of early detection tools and enhance our understanding of how AD affects language over time.

End-to-end task-oriented dialogue systems are prone to fall into the so-called ‘likelihood trap’, resulting in generated responses which are dull, repetitive, and often inconsistent with dialogue history. Comparing ranked lists of multiple generated responses against the ‘gold response’ reveals a wide diversity in quality, with many good responses placed lower in the ranked list. The main challenge addressed in this work is how to reach beyond greedily generated system responses, that is, how to obtain and select high-quality responses from the list of overgenerated responses at inference without the availability of the gold response. To this end, we propose a simple yet effective reranking method to select high-quality items from the lists of initially overgenerated responses. The idea is to use any sequence-level scoring function to divide the semantic space of responses into high-scoring versus low-scoring partitions. At training, the high-scoring partition comprises all generated responses whose similarity to the gold response is higher than the similarity of the greedy response to the gold response. At inference, the aim is to estimate the probability that each overgenerated response belongs to the high-scoring partition. We evaluate our proposed method on the standard MultiWOZ dataset, the BiTOD dataset, and with human evaluation.

Research on text-to-image generation (TTI) still predominantly focuses on the English language due to the lack of annotated image-caption data in other languages; in the long run, this might widen inequitable access to TTI technology. In this work, we thus investigate multilingual TTI (termed mTTI) and the current potential of neural machine translation (NMT) to bootstrap mTTI systems. We provide two key contributions. 1) Relying on a multilingual multi-modal encoder, we provide a systematic empirical study of standard methods used in cross-lingual NLP when applied to mTTI: Translate Train, Translate Test, and Zero-Shot Transfer. 2) We propose Ensemble Adapter (EnsAd), a novel parameter-efficient approach that learns to weigh and consolidate the multilingual text knowledge within the mTTI framework, mitigating the language gap and thus improving mTTI performance. Our evaluations on standard mTTI datasets COCO-CN, Multi30K Task2, and LAION-5B demonstrate the potential of translation-enhanced mTTI systems and also validate the benefits of the proposed EnsAd which derives consistent gains across all datasets. Further investigations on model variants, ablation studies, and qualitative analyses provide additional insights on the inner workings of the proposed mTTI approaches.

Visual question answering (VQA) is one of the crucial vision-and-language tasks. Yet, existing VQA research has mostly focused on the English language, due to a lack of suitable evaluation resources. Previous work on cross-lingual VQA has reported poor zero-shot transfer performance of current multilingual multimodal Transformers with large gaps to monolingual performance, without any deeper analysis. In this work, we delve deeper into the different aspects of cross-lingual VQA, aiming to understand the impact of 1) modeling methods and choices, including architecture, inductive bias, fine-tuning; 2) learning biases: including question types and modality biases in cross-lingual setups. The key results of our analysis are: 1. We show that simple modifications to the standard training setup can substantially reduce the transfer gap to monolingual English performance, yielding +10 accuracy points over existing methods. 2. We analyze cross-lingual VQA across different question types of varying complexity for different multilingual multimodal Transformers, and identify question types that are the most difficult to improve on. 3. We provide an analysis of modality biases present in training data and models, revealing why zero-shot performance gaps remain for certain question types and languages.

Adapters have emerged as a modular and parameter-efficient approach to (zero-shot) cross-lingual transfer. The established MAD-X framework employs separate language and task adapters which can be arbitrarily combined to perform the transfer of any task to any target language. Subsequently, BAD-X, an extension of the MAD-X framework, achieves improved transfer at the cost of MAD-X’s modularity by creating ‘bilingual’ adapters specific to the source-target language pair. In this work, we aim to take the best of both worlds by (i) fine-tuning *task* adapters adapted to the target language(s) (so-called *‘target language-ready’ (TLR)* adapters) to maintain high transfer performance, but (ii) without sacrificing the highly modular design of MAD-X. The main idea of ‘target language-ready’ adapters is to resolve the training-vs-inference discrepancy of MAD-X: the task adapter ‘sees’ the target language adapter for the very first time during inference, and thus might not be fully compatible with it. We address this mismatch by exposing the task adapter to the target language adapter during training, and empirically validate several variants of the idea: in the simplest form, we alternate between using the source and target language adapters during task adapter training, which can be generalized to cycling over any set of language adapters. We evaluate different TLR-based transfer configurations with varying degrees of generality across a suite of standard cross-lingual benchmarks, and find that the most general (and thus most modular) configuration consistently outperforms MAD-X and BAD-X on most tasks and languages.

Task-oriented dialogue (ToD) systems have been widely deployed in many industries as they deliver more efficient customer support. These systems are typically constructed for a single domain or language and do not generalise well beyond this. To support work on Natural Language Understanding (NLU) in ToD across multiple languages and domains simultaneously, we constructed Multi3NLU++, a multilingual, multi-intent, multi-domain dataset. Multi3NLU++ extends the English-only NLU++ dataset to include manual translations into a range of high, medium, and low resource languages (Spanish, Marathi, Turkish and Amharic), in two domains (banking and hotels). Because of its multi-intent property, Multi3NLU++ represents complex and natural user goals, and therefore allows us to measure the realistic performance of ToD systems in a varied set of the world’s languages. We use Multi3NLU++ to benchmark state-of-the-art multilingual models for the NLU tasks of intent detection and slot labeling for ToD systems in the multilingual setting. The results demonstrate the challenging nature of the dataset, particularly in the low-resource language setting, offering ample room for future experimentation in multi-domain multilingual ToD setups.

Massively multilingual Transformers (MMTs), such as mBERT and XLM-R, are widely used for cross-lingual transfer learning. While these are pretrained to represent hundreds of languages, end users of NLP systems are often interested only in individual languages. For such purposes, the MMTs’ language coverage makes them unnecessarily expensive to deploy in terms of model size, inference time, energy, and hardware cost. We thus propose to extract compressed, language-specific models from MMTs which retain the capacity of the original MMTs for cross-lingual transfer. This is achieved by distilling the MMT *bilingually*, i.e., using data from only the source and target language of interest. Specifically, we use a two-phase distillation approach, termed BiStil: (i) the first phase distils a general bilingual model from the MMT, while (ii) the second, task-specific phase sparsely fine-tunes the bilingual “student” model using a task-tuned variant of the original MMT as its “teacher”. We evaluate this distillation technique in zero-shot cross-lingual transfer across a number of standard cross-lingual benchmarks. The key results indicate that the distilled models exhibit minimal degradation in target language performance relative to the base MMT despite being significantly smaller and faster. Furthermore, we find that they outperform multilingually distilled models such as DistilmBERT and MiniLMv2 while having a very modest training budget in comparison, even on a per-language basis. We also show that bilingual models distilled from MMTs greatly outperform bilingual models trained from scratch.

Historically, researchers and consumers have noticed a decrease in quality when applying NLP tools to minority variants of languages (i.e. Puerto Rican Spanish or Swiss German), but studies exploring this have been limited to a select few languages. Additionally, past studies have mainly been conducted in a monolingual context, so cross-linguistic trends have not been identified and tied to external factors. In this work, we conduct a comprehensive evaluation of the most influential, state-of-the-art large language models (LLMs) across two high-use applications, machine translation and automatic speech recognition, to assess their functionality on the regional dialects of several high- and low-resource languages. Additionally, we analyze how the regional dialect gap is correlated with economic, social, and linguistic factors. The impact of training data, including related factors like dataset size and its construction procedure, is shown to be significant but not consistent across models or languages, meaning a one-size-fits-all approach cannot be taken in solving the dialect gap. This work will lay the foundation for furthering the field of dialectal NLP by laying out evident disparities and identifying possible pathways for addressing them through mindful data collection.

Pretrained language models (PLMs) are key components in NLP, but they contain strong social biases. Quantifying these biases is challenging because current methods focusing on fill-the-mask objectives are sensitive to slight changes in input. To address this, we propose a bias probing technique called LABDet, for evaluating social bias in PLMs with a robust and language-agnostic method. For nationality as a case study, we show that LABDet “surfaces” nationality bias by training a classifier on top of a frozen PLM on non-nationality sentiment detection. We find consistent patterns of nationality bias across monolingual PLMs in six languages that align with historical and political context. We also show for English BERT that bias surfaced by LABDet correlates well with bias in the pretraining data; thus, our work is one of the few studies that directly links pretraining data to PLM behavior. Finally, we verify LABDet’s reliability and applicability to different templates and languages through an extensive set of robustness checks. We publicly share our code and dataset in https://github.com/akoksal/LABDet.

Prompt-based learning has been an effective paradigm for large pretrained language models (LLM), enabling few-shot or even zero-shot learning. Black-box prompt search has received growing interest recently for its distinctive properties of gradient-free optimization, proven particularly useful and powerful for model-as-a-service usage. However, the discrete nature and the complexity of combinatorial optimization hinder the efficiency of modern black-box approaches. Despite extensive research on search algorithms, the crucial aspect of search space design and optimization has been largely overlooked. In this paper, we first conduct a sensitivity analysis by prompting LLM, revealing that only a small number of tokens exert a disproportionate amount of influence on LLM predictions. Leveraging this insight, we propose the Clustering and Pruning for Efficient Black-box Prompt Search (ClaPS), a simple black-box search method that first clusters and prunes the search space to focus exclusively on influential prompt tokens. By employing even simple search methods within the pruned search space, ClaPS achieves state-of-the-art performance across various tasks and LLMs, surpassing the performance of complex approaches while significantly reducing search costs. Our findings underscore the critical role of search space design and optimization in enhancing both the usefulness and the efficiency of black-box prompt-based learning.

The scarcity of data in many of the world’s languages necessitates the transfer of knowledge from other, resource-rich languages. However, the level of scarcity varies significantly across multiple dimensions, including: i) the amount of task-specific data available in the source and target languages; ii) the amount of monolingual and parallel data available for both languages; and iii) the extent to which they are supported by pretrained multilingual and translation models. Prior work has largely treated these dimensions and the various techniques for dealing with them separately; in this paper, we offer a more integrated view by exploring how to deploy the arsenal of cross-lingual transfer tools across a range of scenarios, especially the most challenging, low-resource ones. To this end, we run experiments on the AmericasNLI and NusaX benchmarks over 20 languages, simulating a range of few-shot settings. The best configuration in our experiments employed parameter-efficient language and task adaptation of massively multilingual Transformers, trained simultaneously on source language data and both machine-translated and natural data for multiple target languages. In addition, we show that pre-trained translation models can be easily adapted to unseen languages, thus extending the range of our hybrid technique and translation-based transfer more broadly. Beyond new insights into the mechanisms of cross-lingual transfer, we hope our work will provide practitioners with a toolbox to integrate multiple techniques for different real-world scenarios. Our code is available at https://github.com/parovicm/unified-xlt.

Slot labeling (SL) is a core component of task-oriented dialogue (TOD) systems, where slots and corresponding values are usually language-, task- and domain-specific. Therefore, extending the system to any new language-domain-task configuration requires (re)running an expensive and resource-intensive data annotation process. To mitigate the inherent data scarcity issue, current research on multilingual ToD assumes that sufficient English-language annotated data are always available for particular tasks and domains, and thus operates in a standard cross-lingual transfer setup. In this work, we depart from this often unrealistic assumption. We examine challenging scenarios where such transfer-enabling English annotated data cannot be guaranteed, and focus on bootstrapping multilingual data-efficient slot labelers in transfer-free scenarios directly in the target languages without any English-ready data. We propose a two-stage slot labeling approach (termed TWOSL) which transforms standard multilingual sentence encoders into effective slot labelers. In Stage 1, relying on SL-adapted contrastive learning with only a handful of SL-annotated examples, we turn sentence encoders into task-specific span encoders. In Stage 2, we recast SL from a token classification into a simpler, less data-intensive span classification task. Our results on two standard multilingual TOD datasets and across diverse languages confirm the effectiveness and robustness of TWOSL. It is especially effective for the most challenging transfer-free few-shot setups, paving the way for quick and data-efficient bootstrapping of multilingual slot labelers for TOD.

Achieving robust language technologies that can perform well across the world’s many languages is a central goal of multilingual NLP. In this work, we take stock of and empirically analyse task performance disparities that exist between multilingual task-oriented dialogue (ToD) systems. We first define new quantitative measures of absolute and relative equivalence in system performance, capturing disparities across languages and within individual languages. Through a series of controlled experiments, we demonstrate that performance disparities depend on a number of factors: the nature of the ToD task at hand, the underlying pretrained language model, the target language, and the amount of ToD annotated data. We empirically prove the existence of the adaptation and intrinsic biases in current ToD systems: e.g., ToD systems trained for Arabic or Turkish using annotated ToD data fully parallel to English ToD data still exhibit diminished ToD task performance. Beyond providing a series of insights into the performance disparities of ToD systems in different languages, our analyses offer practical tips on how to approach ToD data collection and system development for new languages.

Hallucinations pose a significant challenge to the reliability of neural models for abstractive summarisation. While automatically generated summaries may be fluent, they often lack faithfulness to the original document. This issue becomes even more pronounced in low-resource languages, where summarisation requires cross-lingual transfer. With the existing faithful metrics focusing on English, even measuring the extent of this phenomenon in cross-lingual settings is hard. To address this, we first develop a novel metric, mFACT, evaluating the faithfulness of non-English summaries, leveraging translation-based transfer from multiple English faithfulness metrics. Through extensive experiments in multiple languages, we demonstrate that mFACT is best suited to detect hallucinations compared to alternative metrics. With mFACT, we assess a broad range of multilingual large language models, and find that they all tend to hallucinate often in languages different from English. We then propose a simple but effective method to reduce hallucinations in cross-lingual transfer, which weighs the loss of each training example by its faithfulness score. This method drastically increases both performance and faithfulness according to both automatic and human evaluation when compared to strong baselines for cross-lingual transfer such as MAD-X. Our code and dataset are available at https://github.com/yfqiu-nlp/mfact-summ.

Bilingual Lexicon Induction (BLI) is a core task in multilingual NLP that still, to a large extent, relies on calculating cross-lingual word representations. Inspired by the global paradigm shift in NLP towards Large Language Models (LLMs), we examine the potential of the latest generation of LLMs for the development of bilingual lexicons. We ask the following research question: Is it possible to prompt and fine-tune multilingual LLMs (mLLMs) for BLI, and how does this approach compare against and complement current BLI approaches? To this end, we systematically study 1) zero-shot prompting for unsupervised BLI and 2) few-shot in-context prompting with a set of seed translation pairs, both without any LLM fine-tuning, as well as 3) standard BLI-oriented fine-tuning of smaller LLMs. We experiment with 18 open-source text-to-text mLLMs of different sizes (from 0.3B to 13B parameters) on two standard BLI benchmarks covering a range of typologically diverse languages. Our work is the first to demonstrate strong BLI capabilities of text-to-text mLLMs. The results reveal that few-shot prompting with in-context examples from nearest neighbours achieves the best performance, establishing new state-of-the-art BLI scores for many language pairs. We also conduct a series of in-depth analyses and ablation studies, providing more insights on BLI with (m)LLMs, also along with their limitations.

Acquiring factual knowledge with Pretrained Language Models (PLMs) has attracted increasing attention, showing promising performance in many knowledge-intensive tasks. Their good performance has led the community to believe that the models do possess a modicum of reasoning competence rather than merely memorising the knowledge. In this paper, we conduct a comprehensive evaluation of the learnable deductive (also known as explicit) reasoning capability of PLMs. Through a series of controlled experiments, we posit two main findings. 1) PLMs inadequately generalise learned logic rules and perform inconsistently against simple adversarial surface form edits. 2) While the deductive reasoning fine-tuning of PLMs does improve their performance on reasoning over unseen knowledge facts, it results in catastrophically forgetting the previously learnt knowledge. Our main results suggest that PLMs cannot yet perform reliable deductive reasoning, demonstrating the importance of controlled examinations and probing of PLMs’ deductive reasoning abilities; we reach beyond (misleading) task performance, revealing that PLMs are still far from robust reasoning capabilities, even for simple deductive tasks.

Pretrained multilingual language models (LMs) can be successfully transformed into multilingual sentence encoders (SEs; e.g., LaBSE, xMPNet) via additional fine-tuning or model distillation with parallel data. However, it remains unclear how to best leverage them to represent sub-sentence lexical items (i.e., words and phrases) in cross-lingual lexical tasks. In this work, we probe SEs for the amount of cross-lingual lexical knowledge stored in their parameters, and compare them against the original multilingual LMs. We also devise a simple yet efficient method for exposing the cross-lingual lexical knowledge by means of additional fine-tuning through inexpensive contrastive learning that requires only a small amount of word translation pairs. Using bilingual lexical induction (BLI), cross-lingual lexical semantic similarity, and cross-lingual entity linking as lexical probing tasks, we report substantial gains on standard benchmarks (e.g., +10 Precision@1 points in BLI). The results indicate that the SEs such as LaBSE can be ‘rewired’ into effective cross-lingual lexical encoders via the contrastive learning procedure, and that it is possible to expose more cross-lingual lexical knowledge compared to using them as off-the-shelf SEs. This way, we also provide an effective tool for harnessing ‘covert’ multilingual lexical knowledge hidden in multilingual sentence encoders.

Multilingual task-oriented dialogue (ToD) facilitates access to services and information for many (communities of) speakers. Nevertheless, its potential is not fully realized, as current multilingual ToD datasets—both for modular and end-to-end modeling—suffer from severe limitations. 1) When created from scratch, they are usually small in scale and fail to cover many possible dialogue flows. 2) Translation-based ToD datasets might lack naturalness and cultural specificity in the target language. In this work, to tackle these limitations we propose a novel outline-based annotation process for multilingual ToD datasets, where domain-specific abstract schemata of dialogue are mapped into natural language outlines. These in turn guide the target language annotators in writing dialogues by providing instructions about each turn’s intents and slots. Through this process we annotate a new large-scale dataset for evaluation of multilingual and cross-lingual ToD systems. Our Cross-lingual Outline-based Dialogue dataset (cod) enables natural language understanding, dialogue state tracking, and end-to-end dialogue evaluation in 4 diverse languages: Arabic, Indonesian, Russian, and Kiswahili. Qualitative and quantitative analyses of cod versus an equivalent translation-based dataset demonstrate improvements in data quality, unlocked by the outline-based approach. Finally, we benchmark a series of state-of-the-art systems for cross-lingual ToD, setting reference scores for future work and demonstrating that cod prevents over-inflated performance, typically met with prior translation-based ToD datasets.

Creating high-quality annotated data for task-oriented dialog (ToD) is known to be notoriously difficult, and the challenges are amplified when the goal is to create equitable, culturally adapted, and large-scale ToD datasets for multiple languages. Therefore, the current datasets are still very scarce and suffer from limitations such as translation-based non-native dialogs with translation artefacts, small scale, or lack of cultural adaptation, among others. In this work, we first take stock of the current landscape of multilingual ToD datasets, offering a systematic overview of their properties and limitations. Aiming to reduce all the detected limitations, we then introduce Multi3WOZ, a novel multilingual, multi-domain, multi-parallel ToD dataset. It is large-scale and offers culturally adapted dialogs in 4 languages to enable training and evaluation of multilingual and cross-lingual ToD systems. We describe a complex bottom–up data collection process that yielded the final dataset, and offer the first sets of baseline scores across different ToD-related tasks for future reference, also highlighting its challenging nature.

Adapter modules enable modular and efficient zero-shot cross-lingual transfer, where current state-of-the-art adapter-based approaches learn specialized language adapters (LAs) for individual languages. In this work, we show that it is more effective to learn bilingual language pair adapters (BAs) when the goal is to optimize performance for a particular source-target transfer direction. Our novel BAD-X adapter framework trades off some modularity of dedicated LAs for improved transfer performance: we demonstrate consistent gains in three standard downstream tasks, and for the majority of evaluated low-resource languages.

Fine-tuning the entire set of parameters of a large pretrained model has become the mainstream approach for transfer learning. To increase its efficiency and prevent catastrophic forgetting and interference, techniques like adapters and sparse fine-tuning have been developed. Adapters are modular, as they can be combined to adapt a model towards different facets of knowledge (e.g., dedicated language and/or task adapters). Sparse fine-tuning is expressive, as it controls the behavior of all model components. In this work, we introduce a new fine-tuning method with both these desirable properties. In particular, we learn sparse, real-valued masks based on a simple variant of the Lottery Ticket Hypothesis. Task-specific masks are obtained from annotated data in a source language, and language-specific masks from masked language modeling in a target language. Both these masks can then be composed with the pretrained model. Unlike adapter-based fine-tuning, this method neither increases the number of parameters at inference time nor alters the original model architecture. Most importantly, it outperforms adapters in zero-shot cross-lingual transfer by a large margin in a series of multilingual benchmarks, including Universal Dependencies, MasakhaNER, and AmericasNLI. Based on an in-depth analysis, we additionally find that sparsity is crucial to prevent both 1) interference between the fine-tunings to be composed and 2) overfitting. We release the code and models at https://github.com/cambridgeltl/composable-sft.

Word translation or bilingual lexicon induction (BLI) is a key cross-lingual task, aiming to bridge the lexical gap between different languages. In this work, we propose a robust and effective two-stage contrastive learning framework for the BLI task. At Stage C1, we propose to refine standard cross-lingual linear maps between static word embeddings (WEs) via a contrastive learning objective; we also show how to integrate it into the self-learning procedure for even more refined cross-lingual maps. In Stage C2, we conduct BLI-oriented contrastive fine-tuning of mBERT, unlocking its word translation capability. We also show that static WEs induced from the ‘C2-tuned’ mBERT complement static WEs from Stage C1. Comprehensive experiments on standard BLI datasets for diverse languages and different experimental setups demonstrate substantial gains achieved by our framework. While the BLI method from Stage C1 already yields substantial gains over all state-of-the-art BLI methods in our comparison, even stronger improvements are met with the full two-stage framework: e.g., we report gains for 112/112 BLI setups, spanning 28 language pairs.

State-of-the-art pretrained contextualized models (PCM) eg. BERT use tasks such as WiC and WSD to evaluate their word-in-context representations. This inherently assumes that performance in these tasks reflect how well a model represents the coupled word and context semantics. We question this assumption by presenting the first quantitative analysis on the context-word interaction being tested in major contextual lexical semantic tasks. To achieve this, we run probing baselines on masked input, and propose measures to calculate and visualize the degree of context or word biases in existing datasets. The analysis was performed on both models and humans. Our findings demonstrate that models are usually not being tested for word-in-context semantics in the same way as humans are in these tasks, which helps us better understand the model-human gap. Specifically, to PCMs, most existing datasets fall into the extreme ends (the retrieval-based tasks exhibit strong target word bias while WiC-style tasks and WSD show strong context bias); In comparison, humans are less biased and achieve much better performance when both word and context are available than with masked input. We recommend our framework for understanding and controlling these biases for model interpretation and future task design.

Scaling dialogue systems to a multitude of domains, tasks and languages relies on costly and time-consuming data annotation for different domain-task-language configurations. The annotation efforts might be substantially reduced by the methods that generalise well in zero- and few-shot scenarios, and also effectively leverage external unannotated data sources (e.g., Web-scale corpora). We propose two methods to this aim, offering improved dialogue natural language understanding (NLU) across multiple languages: 1) Multi-SentAugment, and 2) LayerAgg. Multi-SentAugment is a self-training method which augments available (typically few-shot) training data with similar (automatically labelled) in-domain sentences from large monolingual Web-scale corpora. LayerAgg learns to select and combine useful semantic information scattered across different layers of a Transformer model (e.g., mBERT); it is especially suited for zero-shot scenarios as semantically richer representations should strengthen the model’s cross-lingual capabilities. Applying the two methods with state-of-the-art NLU models obtains consistent improvements across two standard multilingual NLU datasets covering 16 diverse languages. The gains are observed in zero-shot, few-shot, and even in full-data scenarios. The results also suggest that the two methods achieve a synergistic effect: the best overall performance in few-shot setups is attained when the methods are used together.

Bilingual lexicon induction (BLI) with limited bilingual supervision is a crucial yet challenging task in multilingual NLP. Current state-of-the-art BLI methods rely on the induction of cross-lingual word embeddings (CLWEs) to capture cross-lingual word similarities; such CLWEs are obtained <b>1)</b> via traditional static models (e.g., VecMap), or <b>2)</b> by extracting type-level CLWEs from multilingual pretrained language models (mPLMs), or <b>3)</b> through combining the former two options. In this work, we propose a novel semi-supervised <i>post-hoc</i> reranking method termed <b>BLICEr</b> (<b>BLI</b> with <b>C</b>ross-<b>E</b>ncoder <b>R</b>eranking), applicable to any precalculated CLWE space, which improves their BLI capability. The key idea is to ‘extract’ cross-lingual lexical knowledge from mPLMs, and then combine it with the original CLWEs. This crucial step is done via <b>1)</b> creating a word similarity dataset, comprising positive word pairs (i.e., true translations) and hard negative pairs induced from the original CLWE space, and then <b>2)</b> fine-tuning an mPLM (e.g., mBERT or XLM-R) in a cross-encoder manner to predict the similarity scores. At inference, we <b>3)</b> combine the similarity score from the original CLWE space with the score from the BLI-tuned cross-encoder. BLICEr establishes new state-of-the-art results on two standard BLI benchmarks spanning a wide spectrum of diverse languages: it substantially outperforms a series of strong baselines across the board. We also validate the robustness of BLICEr with different CLWEs.

Transformer-based language models (LMs) pretrained on large text collections implicitly store a wealth of lexical semantic knowledge, but it is non-trivial to extract that knowledge effectively from their parameters. Inspired by prior work on semantic specialization of static word embedding (WE) models, we show that it is possible to expose and enrich lexical knowledge from the LMs, that is, to specialize them to serve as effective and universal “decontextualized” word encoders even when fed input words “in isolation” (i.e., without any context). Their transformation into such word encoders is achieved through a simple and efficient lexical fine-tuning procedure (termed LexFit) based on dual-encoder network structures. Further, we show that LexFit can yield effective word encoders even with limited lexical supervision and, via cross-lingual transfer, in different languages without any readily available external knowledge. Our evaluation over four established, structurally different lexical-level tasks in 8 languages indicates the superiority of LexFit-based WEs over standard static WEs (e.g., fastText) and WEs from vanilla LMs. Other extensive experiments and ablation studies further profile the LexFit framework, and indicate best practices and performance variations across LexFit variants, languages, and lexical tasks, also directly questioning the usefulness of traditional WE models in the era of large neural models.

Few-shot crosslingual transfer has been shown to outperform its zero-shot counterpart with pretrained encoders like multilingual BERT. Despite its growing popularity, little to no attention has been paid to standardizing and analyzing the design of few-shot experiments. In this work, we highlight a fundamental risk posed by this shortcoming, illustrating that the model exhibits a high degree of sensitivity to the selection of few shots. We conduct a large-scale experimental study on 40 sets of sampled few shots for six diverse NLP tasks across up to 40 languages. We provide an analysis of success and failure cases of few-shot transfer, which highlights the role of lexical features. Additionally, we show that a straightforward full model finetuning approach is quite effective for few-shot transfer, outperforming several state-of-the-art few-shot approaches. As a step towards standardizing few-shot crosslingual experimental designs, we make our sampled few shots publicly available.

Linguistic probing of pretrained Transformer-based language models (LMs) revealed that they encode a range of syntactic and semantic properties of a language. However, they are still prone to fall back on superficial cues and simple heuristics to solve downstream tasks, rather than leverage deeper linguistic information. In this paper, we target a specific facet of linguistic knowledge, the interplay between verb meaning and argument structure. We investigate whether injecting explicit information on verbs’ semantic-syntactic behaviour improves the performance of pretrained LMs in event extraction tasks, where accurate verb processing is paramount. Concretely, we impart the verb knowledge from curated lexical resources into dedicated adapter modules (verb adapters), allowing it to complement, in downstream tasks, the language knowledge obtained during LM-pretraining. We first demonstrate that injecting verb knowledge leads to performance gains in English event extraction. We then explore the utility of verb adapters for event extraction in other languages: we investigate 1) zero-shot language transfer with multilingual Transformers and 2) transfer via (noisy automatic) translation of English verb-based lexical knowledge. Our results show that the benefits of verb knowledge injection indeed extend to other languages, even when relying on noisily translated lexical knowledge.

Injecting external domain-specific knowledge (e.g., UMLS) into pretrained language models (LMs) advances their capability to handle specialised in-domain tasks such as biomedical entity linking (BEL). However, such abundant expert knowledge is available only for a handful of languages (e.g., English). In this work, by proposing a novel cross-lingual biomedical entity linking task (XL-BEL) and establishing a new XL-BEL benchmark spanning 10 typologically diverse languages, we first investigate the ability of standard knowledge-agnostic as well as knowledge-enhanced monolingual and multilingual LMs beyond the standard monolingual English BEL task. The scores indicate large gaps to English performance. We then address the challenge of transferring domain-specific knowledge in resource-rich languages to resource-poor ones. To this end, we propose and evaluate a series of cross-lingual transfer methods for the XL-BEL task, and demonstrate that general-domain bitext helps propagate the available English knowledge to languages with little to no in-domain data. Remarkably, we show that our proposed domain-specific transfer methods yield consistent gains across all target languages, sometimes up to 20 Precision@1 points, without any in-domain knowledge in the target language, and without any in-domain parallel data.

Most combinations of NLP tasks and language varieties lack in-domain examples for supervised training because of the paucity of annotated data. How can neural models make sample-efficient generalizations from task–language combinations with available data to low-resource ones? In this work, we propose a Bayesian generative model for the space of neural parameters. We assume that this space can be factorized into latent variables for each language and each task. We infer the posteriors over such latent variables based on data from seen task–language combinations through variational inference. This enables zero-shot classification on unseen combinations at prediction time. For instance, given training data for named entity recognition (NER) in Vietnamese and for part-of-speech (POS) tagging in Wolof, our model can perform accurate predictions for NER in Wolof. In particular, we experiment with a typologically diverse sample of 33 languages from 4 continents and 11 families, and show that our model yields comparable or better results than state-of-the-art, zero-shot cross-lingual transfer methods. Our code is available at github.com/cambridgeltl/parameter-factorization.

Recent work indicated that pretrained language models (PLMs) such as BERT and RoBERTa can be transformed into effective sentence and word encoders even via simple self-supervised techniques. Inspired by this line of work, in this paper we propose a fully unsupervised approach to improving word-in-context (WiC) representations in PLMs, achieved via a simple and efficient WiC-targeted fine-tuning procedure: MirrorWiC. The proposed method leverages only raw texts sampled from Wikipedia, assuming no sense-annotated data, and learns context-aware word representations within a standard contrastive learning setup. We experiment with a series of standard and comprehensive WiC benchmarks across multiple languages. Our proposed fully unsupervised MirrorWiC models obtain substantial gains over off-the-shelf PLMs across all monolingual, multilingual and cross-lingual setups. Moreover, on some standard WiC benchmarks, MirrorWiC is even on-par with supervised models fine-tuned with in-task data and sense labels.

Research into representation learning models of lexical semantics usually utilizes some form of intrinsic evaluation to ensure that the learned representations reflect human semantic judgments. Lexical semantic similarity estimation is a widely used evaluation method, but efforts have typically focused on pairwise judgments of words in isolation, or are limited to specific contexts and lexical stimuli. There are limitations with these approaches that either do not provide any context for judgments, and thereby ignore ambiguity, or provide very specific sentential contexts that cannot then be used to generate a larger lexical resource. Furthermore, similarity between more than two items is not considered. We provide a full description and analysis of our recently proposed methodology for large-scale data set construction that produces a semantic classification of a large sample of verbs in the first phase, as well as multi-way similarity judgments made within the resultant semantic classes in the second phase. The methodology uses a spatial multi-arrangement approach proposed in the field of cognitive neuroscience for capturing multi-way similarity judgments of visual stimuli. We have adapted this method to handle polysemous linguistic stimuli and much larger samples than previous work. We specifically target verbs, but the method can equally be applied to other parts of speech. We perform cluster analysis on the data from the first phase and demonstrate how this might be useful in the construction of a comprehensive verb resource. We also analyze the semantic information captured by the second phase and discuss the potential of the spatially induced similarity judgments to better reflect human notions of word similarity. We demonstrate how the resultant data set can be used for fine-grained analyses and evaluation of representation learning models on the intrinsic tasks of semantic clustering and semantic similarity. In particular, we find that stronger static word embedding methods still outperform lexical representations emerging from more recent pre-training methods, both on word-level similarity and clustering. Moreover, thanks to the data set’s vast coverage, we are able to compare the benefits of specializing vector representations for a particular type of external knowledge by evaluating FrameNet- and VerbNet-retrofitted models on specific semantic domains such as “Heat” or “Motion.”

Semi-supervised learning through deep generative models and multi-lingual pretraining techniques have orchestrated tremendous success across different areas of NLP. Nonetheless, their development has happened in isolation, while the combination of both could potentially be effective for tackling task-specific labelled data shortage. To bridge this gap, we combine semi-supervised deep generative models and multi-lingual pretraining to form a pipeline for document classification task. Compared to strong supervised learning baselines, our semi-supervised classification framework is highly competitive and outperforms the state-of-the-art counterparts in low-resource settings across several languages.

The performance of NMT systems has improved drastically in the past few years but the translation of multi-sense words still poses a challenge. Since word senses are not represented uniformly in the parallel corpora used for training, there is an excessive use of the most frequent sense in MT output. In this work, we propose CmBT (Contextually-mined Back-Translation), an approach for improving multi-sense word translation leveraging pre-trained cross-lingual contextual word representations (CCWRs). Because of their contextual sensitivity and their large pre-training data, CCWRs can easily capture word senses that are missing or very rare in parallel corpora used to train MT. Specifically, CmBT applies bilingual lexicon induction on CCWRs to mine sense-specific target sentences from a monolingual dataset, and then back-translates these sentences to generate a pseudo parallel corpus as additional training data for an MT system. We test the translation quality of ambiguous words on the MuCoW test suite, which was built to test the word sense disambiguation effectiveness of MT systems. We show that our system improves on the translation of difficult unseen and low frequency word senses.

Adapter modules have emerged as a general parameter-efficient means to specialize a pretrained encoder to new domains. Massively multilingual transformers (MMTs) have particularly benefited from additional training of language-specific adapters. However, this approach is not viable for the vast majority of languages, due to limitations in their corpus size or compute budgets. In this work, we propose MAD-G (Multilingual ADapter Generation), which contextually generates language adapters from language representations based on typological features. In contrast to prior work, our time- and space-efficient MAD-G approach enables (1) sharing of linguistic knowledge across languages and (2) zero-shot inference by generating language adapters for unseen languages. We thoroughly evaluate MAD-G in zero-shot cross-lingual transfer on part-of-speech tagging, dependency parsing, and named entity recognition. While offering (1) improved fine-tuning efficiency (by a factor of around 50 in our experiments), (2) a smaller parameter budget, and (3) increased language coverage, MAD-G remains competitive with more expensive methods for language-specific adapter training across the board. Moreover, it offers substantial benefits for low-resource languages, particularly on the NER task in low-resource African languages. Finally, we demonstrate that MAD-G’s transfer performance can be further improved via: (i) multi-source training, i.e., by generating and combining adapters of multiple languages with available task-specific training data; and (ii) by further fine-tuning generated MAD-G adapters for languages with monolingual data.

Previous work has indicated that pretrained Masked Language Models (MLMs) are not effective as universal lexical and sentence encoders off-the-shelf, i.e., without further task-specific fine-tuning on NLI, sentence similarity, or paraphrasing tasks using annotated task data. In this work, we demonstrate that it is possible to turn MLMs into effective lexical and sentence encoders even without any additional data, relying simply on self-supervision. We propose an extremely simple, fast, and effective contrastive learning technique, termed Mirror-BERT, which converts MLMs (e.g., BERT and RoBERTa) into such encoders in 20-30 seconds with no access to additional external knowledge. Mirror-BERT relies on identical and slightly modified string pairs as positive (i.e., synonymous) fine-tuning examples, and aims to maximise their similarity during “identity fine-tuning”. We report huge gains over off-the-shelf MLMs with Mirror-BERT both in lexical-level and in sentence-level tasks, across different domains and different languages. Notably, in sentence similarity (STS) and question-answer entailment (QNLI) tasks, our self-supervised Mirror-BERT model even matches the performance of the Sentence-BERT models from prior work which rely on annotated task data. Finally, we delve deeper into the inner workings of MLMs, and suggest some evidence on why this simple Mirror-BERT fine-tuning approach can yield effective universal lexical and sentence encoders.

Capturing word meaning in context and distinguishing between correspondences and variations across languages is key to building successful multilingual and cross-lingual text representation models. However, existing multilingual evaluation datasets that evaluate lexical semantics “in-context” have various limitations. In particular, 1) their language coverage is restricted to high-resource languages and skewed in favor of only a few language families and areas, 2) a design that makes the task solvable via superficial cues, which results in artificially inflated (and sometimes super-human) performances of pretrained encoders, and 3) no support for cross-lingual evaluation. In order to address these gaps, we present AM2iCo (Adversarial and Multilingual Meaning in Context), a wide-coverage cross-lingual and multilingual evaluation set; it aims to faithfully assess the ability of state-of-the-art (SotA) representation models to understand the identity of word meaning in cross-lingual contexts for 14 language pairs. We conduct a series of experiments in a wide range of setups and demonstrate the challenging nature of AM2iCo. The results reveal that current SotA pretrained encoders substantially lag behind human performance, and the largest gaps are observed for low-resource languages and languages dissimilar to English.

Unsupervised pretraining models have been shown to facilitate a wide range of downstream NLP applications. These models, however, retain some of the limitations of traditional static word embeddings. In particular, they encode only the distributional knowledge available in raw text corpora, incorporated through language modeling objectives. In this work, we complement such distributional knowledge with external lexical knowledge, that is, we integrate the discrete knowledge on word-level semantic similarity into pretraining. To this end, we generalize the standard BERT model to a multi-task learning setting where we couple BERT’s masked language modeling and next sentence prediction objectives with an auxiliary task of binary word relation classification. Our experiments suggest that our “Lexically Informed” BERT (LIBERT), specialized for the word-level semantic similarity, yields better performance than the lexically blind “vanilla” BERT on several language understanding tasks. Concretely, LIBERT outperforms BERT in 9 out of 10 tasks of the GLUE benchmark and is on a par with BERT in the remaining one. Moreover, we show consistent gains on 3 benchmarks for lexical simplification, a task where knowledge about word-level semantic similarity is paramount, as well as large gains on lexical reasoning probes.

While state-of-the-art models that rely upon massively multilingual pretrained encoders achieve sample efficiency in downstream applications, they still require abundant amounts of unlabelled text. Nevertheless, most of the world’s languages lack such resources. Hence, we investigate a more radical form of unsupervised knowledge transfer in the absence of linguistic data. In particular, for the first time we pretrain neural networks via emergent communication from referential games. Our key assumption is that grounding communication on images—as a crude approximation of real-world environments—inductively biases the model towards learning natural languages. On the one hand, we show that this substantially benefits machine translation in few-shot settings. On the other hand, this also provides an extrinsic evaluation protocol to probe the properties of emergent languages ex vitro. Intuitively, the closer they are to natural languages, the higher the gains from pretraining on them should be. For instance, in this work we measure the influence of communication success and maximum sequence length on downstream performances. Finally, we introduce a customised adapter layer and annealing strategies for the regulariser of maximum-a-posteriori inference during fine-tuning. These turn out to be crucial to facilitate knowledge transfer and prevent catastrophic forgetting. Compared to a recurrent baseline, our method yields gains of 59.0% 147.6% in BLEU score with only 500 NMT training instances and 65.1% 196.7% with 1,000 NMT training instances across four language pairs. These proof-of-concept results reveal the potential of emergent communication pretraining for both natural language processing tasks in resource-poor settings and extrinsic evaluation of artificial languages.

We present the first evaluation of the applicability of a spatial arrangement method (SpAM) to a typologically diverse language sample, and its potential to produce semantic evaluation resources to support multilingual NLP, with a focus on verb semantics. We demonstrate SpAM’s utility in allowing for quick bottom-up creation of large-scale evaluation datasets that balance cross-lingual alignment with language specificity. Starting from a shared sample of 825 English verbs, translated into Chinese, Japanese, Finnish, Polish, and Italian, we apply a two-phase annotation process which produces (i) semantic verb classes and (ii) fine-grained similarity scores for nearly 130 thousand verb pairs. We use the two types of verb data to (a) examine cross-lingual similarities and variation, and (b) evaluate the capacity of static and contextualised representation models to accurately reflect verb semantics, contrasting the performance of large language specific pretraining models with their multilingual equivalent on semantic clustering and lexical similarity, across different domains of verb meaning. We release the data from both phases as a large-scale multilingual resource, comprising 85 verb classes and nearly 130k pairwise similarity scores, offering a wealth of possibilities for further evaluation and research on multilingual verb semantics.

Work on projection-based induction of cross-lingual word embedding spaces (CLWEs) predominantly focuses on the improvement of the projection (i.e., mapping) mechanisms. In this work, in contrast, we show that a simple method for post-processing monolingual embedding spaces facilitates learning of the cross-lingual alignment and, in turn, substantially improves bilingual lexicon induction (BLI). The post-processing method we examine is grounded in the generalisation of first- and second-order monolingual similarities to the nth-order similarity. By post-processing monolingual spaces before the cross-lingual alignment, the method can be coupled with any projection-based method for inducing CLWE spaces. We demonstrate the effectiveness of this simple monolingual post-processing across a set of 15 typologically diverse languages (i.e., 15*14 BLI setups), and in combination with two different projection methods.

Lexical entailment (LE) is a fundamental asymmetric lexico-semantic relation, supporting the hierarchies in lexical resources (e.g., WordNet, ConceptNet) and applications like natural language inference and taxonomy induction. Multilingual and cross-lingual NLP applications warrant models for LE detection that go beyond language boundaries. As part of SemEval 2020, we carried out a shared task (Task 2) on multilingual and cross-lingual LE. The shared task spans three dimensions: (1) monolingual vs. cross-lingual LE, (2) binary vs. graded LE, and (3) a set of 6 diverse languages (and 15 corresponding language pairs). We offered two different evaluation tracks: (a) Dist: for unsupervised, fully distributional models that capture LE solely on the basis of unannotated corpora, and (b) Any: for externally informed models, allowed to leverage any resources, including lexico-semantic networks (e.g., WordNet or BabelNet). In the Any track, we recieved runs that push state-of-the-art across all languages and language pairs, for both binary LE detection and graded LE prediction.

We present a novel methodology for fast bottom-up creation of large-scale semantic similarity resources to support development and evaluation of NLP systems. Our work targets verb similarity, but the methodology is equally applicable to other parts of speech. Our approach circumvents the bottleneck of slow and expensive manual development of lexical resources by leveraging semantic intuitions of native speakers and adapting a spatial multi-arrangement approach from cognitive neuroscience, used before only with visual stimuli, to lexical stimuli. Our approach critically obtains judgments of word similarity in the context of a set of related words, rather than of word pairs in isolation. We also handle lexical ambiguity as a natural consequence of a two-phase process where verbs are placed in broad semantic classes prior to the fine-grained spatial similarity judgments. Our proposed design produces a large-scale verb resource comprising 17 relatedness-based classes and a verb similarity dataset containing similarity scores for 29,721 unique verb pairs and 825 target verbs, which we release with this paper.

We present a neural framework for learning associations between interrelated groups of words such as the ones found in Subject-Verb-Object (SVO) structures. Our model induces a joint function-specific word vector space, where vectors of e.g. plausible SVO compositions lie close together. The model retains information about word group membership even in the joint space, and can thereby effectively be applied to a number of tasks reasoning over the SVO structure. We show the robustness and versatility of the proposed framework by reporting state-of-the-art results on the tasks of estimating selectional preference and event similarity. The results indicate that the combinations of representations learned with our task-independent model outperform task-specific architectures from prior work, while reducing the number of parameters by up to 95%.

This paper presents an investigation on the distribution of word vectors belonging to a certain word class in a pre-trained word vector space. To this end, we made several assumptions about the distribution, modeled the distribution accordingly, and validated each assumption by comparing the goodness of each model. Specifically, we considered two types of word classes – the semantic class of direct objects of a verb and the semantic class in a thesaurus – and tried to build models that properly estimate how likely it is that a word in the vector space is a member of a given word class. Our results on selectional preference and WordNet datasets show that the centroid-based model will fail to achieve good enough performance, the geometry of the distribution and the existence of subgroups will have limited impact, and also the negative instances need to be considered for adequate modeling of the distribution. We further investigated the relationship between the scores calculated by each model and the degree of membership and found that discriminative learning-based models are best in finding the boundaries of a class, while models based on the offset between positive and negative instances perform best in determining the degree of membership.

Effective projection-based cross-lingual word embedding (CLWE) induction critically relies on the iterative self-learning procedure. It gradually expands the initial small seed dictionary to learn improved cross-lingual mappings. In this work, we present ClassyMap, a classification-based approach to self-learning, yielding a more robust and a more effective induction of projection-based CLWEs. Unlike prior self-learning methods, our approach allows for integration of diverse features into the iterative process. We show the benefits of ClassyMap for bilingual lexicon induction: we report consistent improvements in a weakly supervised setup (500 seed translation pairs) on a benchmark with 28 language pairs.

We introduce Multi-SimLex, a large-scale lexical resource and evaluation benchmark covering data sets for 12 typologically diverse languages, including major languages (e.g., Mandarin Chinese, Spanish, Russian) as well as less-resourced ones (e.g., Welsh, Kiswahili). Each language data set is annotated for the lexical relation of semantic similarity and contains 1,888 semantically aligned concept pairs, providing a representative coverage of word classes (nouns, verbs, adjectives, adverbs), frequency ranks, similarity intervals, lexical fields, and concreteness levels. Additionally, owing to the alignment of concepts across languages, we provide a suite of 66 crosslingual semantic similarity data sets. Because of its extensive size and language coverage, Multi-SimLex provides entirely novel opportunities for experimental evaluation and analysis. On its monolingual and crosslingual benchmarks, we evaluate and analyze a wide array of recent state-of-the-art monolingual and crosslingual representation models, including static and contextualized word embeddings (such as fastText, monolingual and multilingual BERT, XLM), externally informed lexical representations, as well as fully unsupervised and (weakly) supervised crosslingual word embeddings. We also present a step-by-step data set creation protocol for creating consistent, Multi-Simlex–style resources for additional languages. We make these contributions—the public release of Multi-SimLex data sets, their creation protocol, strong baseline results, and in-depth analyses which can be helpful in guiding future developments in multilingual lexical semantics and representation learning—available via a Web site that will encourage community effort in further expansion of Multi-Simlex to many more languages. Such a large-scale semantic resource could inspire significant further advances in NLP across languages.

In order to simulate human language capacity, natural language processing systems must be able to reason about the dynamics of everyday situations, including their possible causes and effects. Moreover, they should be able to generalise the acquired world knowledge to new languages, modulo cultural differences. Advances in machine reasoning and cross-lingual transfer depend on the availability of challenging evaluation benchmarks. Motivated by both demands, we introduce Cross-lingual Choice of Plausible Alternatives (XCOPA), a typologically diverse multilingual dataset for causal commonsense reasoning in 11 languages, which includes resource-poor languages like Eastern Apurímac Quechua and Haitian Creole. We evaluate a range of state-of-the-art models on this novel dataset, revealing that the performance of current methods based on multilingual pretraining and zero-shot fine-tuning falls short compared to translation-based transfer. Finally, we propose strategies to adapt multilingual models to out-of-sample resource-lean languages where only a small corpus or a bilingual dictionary is available, and report substantial improvements over the random baseline. The XCOPA dataset is freely available at github.com/cambridgeltl/xcopa.

Performance in cross-lingual NLP tasks is impacted by the (dis)similarity of languages at hand: e.g., previous work has suggested there is a connection between the expected success of bilingual lexicon induction (BLI) and the assumption of (approximate) isomorphism between monolingual embedding spaces. In this work we present a large-scale study focused on the correlations between monolingual embedding space similarity and task performance, covering thousands of language pairs and four different tasks: BLI, parsing, POS tagging and MT. We hypothesize that statistics of the spectrum of each monolingual embedding space indicate how well they can be aligned. We then introduce several isomorphism measures between two embedding spaces, based on the relevant statistics of their individual spectra. We empirically show that (1) language similarity scores derived from such spectral isomorphism measures are strongly associated with performance observed in different cross-lingual tasks, and (2) our spectral-based measures consistently outperform previous standard isomorphism measures, while being computationally more tractable and easier to interpret. Finally, our measures capture complementary information to typologically driven language distance measures, and the combination of measures from the two families yields even higher task performance correlations.

One of the most powerful features of contextualized models is their dynamic embeddings for words in context, leading to state-of-the-art representations for context-aware lexical semantics. In this paper, we present a post-processing technique that enhances these representations by learning a transformation through static anchors. Our method requires only another pre-trained model and no labeled data is needed. We show consistent improvement in a range of benchmark tasks that test contextual variations of meaning both across different usages of a word and across different words as they are used in context. We demonstrate that while the original contextual representations can be improved by another embedding space from both contextualized and static models, the static embeddings, which have lower computational requirements, provide the most gains.

The success of large pretrained language models (LMs) such as BERT and RoBERTa has sparked interest in probing their representations, in order to unveil what types of knowledge they implicitly capture. While prior research focused on morphosyntactic, semantic, and world knowledge, it remains unclear to which extent LMs also derive lexical type-level knowledge from words in context. In this work, we present a systematic empirical analysis across six typologically diverse languages and five different lexical tasks, addressing the following questions: 1) How do different lexical knowledge extraction strategies (monolingual versus multilingual source LM, out-of-context versus in-context encoding, inclusion of special tokens, and layer-wise averaging) impact performance? How consistent are the observed effects across tasks and languages? 2) Is lexical knowledge stored in few parameters, or is it scattered throughout the network? 3) How do these representations fare against traditional static word vectors in lexical tasks 4) Does the lexical information emerging from independently trained monolingual LMs display latent similarities? Our main results indicate patterns and best practices that hold universally, but also point to prominent variations across languages and tasks. Moreover, we validate the claim that lower Transformer layers carry more type-level lexical knowledge, but also show that this knowledge is distributed across multiple layers.

There is a growing awareness of the need to handle rare and unseen words in word representation modelling. In this paper, we focus on few-shot learning of emerging concepts that fully exploits only a few available contexts. We introduce a substitute-based context representation technique that can be applied on an existing word embedding space. Previous context-based approaches to modelling unseen words only consider bag-of-word first-order contexts, whereas our method aggregates contexts as second-order substitutes that are produced by a sequence-aware sentence completion model. We experimented with three tasks that aim to test the modelling of emerging concepts. We found that these tasks show different emphasis on first and second order contexts, and our substitute-based method achieves superior performance on naturally-occurring contexts from corpora.

In this paper, we present a thorough investigation on methods that align pre-trained contextualized embeddings into shared cross-lingual context-aware embedding space, providing strong reference benchmarks for future context-aware crosslingual models. We propose a novel and challenging task, Bilingual Token-level Sense Retrieval (BTSR). It specifically evaluates the accurate alignment of words with the same meaning in cross-lingual non-parallel contexts, currently not evaluated by existing tasks such as Bilingual Contextual Word Similarity and Sentence Retrieval. We show how the proposed BTSR task highlights the merits of different alignment methods. In particular, we find that using context average type-level alignment is effective in transferring monolingual contextualized embeddings cross-lingually especially in non-parallel contexts, and at the same time improves the monolingual space. Furthermore, aligning independently trained models yields better performance than aligning multilingual embeddings with shared vocabulary.

Recent work has validated the importance of subword information for word representation learning. Since subwords increase parameter sharing ability in neural models, their value should be even more pronounced in low-data regimes. In this work, we therefore provide a comprehensive analysis focused on the usefulness of subwords for word representation learning in truly low-resource scenarios and for three representative morphological tasks: fine-grained entity typing, morphological tagging, and named entity recognition. We conduct a systematic study that spans several dimensions of comparison: 1) type of data scarcity which can stem from the lack of task-specific training data, or even from the lack of unannotated data required to train word embeddings, or both; 2) language type by working with a sample of 16 typologically diverse languages including some truly low-resource ones (e.g. Rusyn, Buryat, and Zulu); 3) the choice of the subword-informed word representation method. Our main results show that subword-informed models are universally useful across all language types, with large gains over subword-agnostic embeddings. They also suggest that the effective use of subwords largely depends on the language (type) and the task at hand, as well as on the amount of available data for training the embeddings and task-based models, where having sufficient in-task data is a more critical requirement.

The use of subword-level information (e.g., characters, character n-grams, morphemes) has become ubiquitous in modern word representation learning. Its importance is attested especially for morphologically rich languages which generate a large number of rare words. Despite a steadily increasing interest in such subword-informed word representations, their systematic comparative analysis across typologically diverse languages and different tasks is still missing. In this work, we deliver such a study focusing on the variation of two crucial components required for subword-level integration into word representation models: 1) segmentation of words into subword units, and 2) subword composition functions to obtain final word representations. We propose a general framework for learning subword-informed word representations that allows for easy experimentation with different segmentation and composition components, also including more advanced techniques based on position embeddings and self-attention. Using the unified framework, we run experiments over a large number of subword-informed word representation configurations (60 in total) on 3 tasks (general and rare word similarity, dependency parsing, fine-grained entity typing) for 5 languages representing 3 language types. Our main results clearly indicate that there is no “one-size-fits-all” configuration, as performance is both language- and task-dependent. We also show that configurations based on unsupervised segmentation (e.g., BPE, Morfessor) are sometimes comparable to or even outperform the ones based on supervised word segmentation.

While neural dependency parsers provide state-of-the-art accuracy for several languages, they still rely on large amounts of costly labeled training data. We demonstrate that in the small data regime, where uncertainty around parameter estimation and model prediction matters the most, Bayesian neural modeling is very effective. In order to overcome the computational and statistical costs of the approximate inference step in this framework, we utilize an efficient sampling procedure via stochastic gradient Langevin dynamics to generate samples from the approximated posterior. Moreover, we show that our Bayesian neural parser can be further improved when integrated into a multi-task parsing and POS tagging framework, designed to minimize task interference via an adversarial procedure. When trained and tested on 6 languages with less than 5k training instances, our parser consistently outperforms the strong bilstm baseline (Kiperwasser and Goldberg, 2016). Compared with the biaffine parser (Dozat et al., 2017) our model achieves an improvement of up to 3% for Vietnames and Irish, while our multi-task model achieves an improvement of up to 9% across five languages: Farsi, Russian, Turkish, Vietnamese, and Irish.

In recent years neural language models (LMs) have set the state-of-the-art performance for several benchmarking datasets. While the reasons for their success and their computational demand are well-documented, a comparison between neural models and more recent developments in n-gram models is neglected. In this paper, we examine the recent progress in n-gram literature, running experiments on 50 languages covering all morphological language families. Experimental results illustrate that a simple extension of Modified Kneser-Ney outperforms an lstm language model on 42 languages while a word-level Bayesian n-gram LM (Shareghi et al., 2017) outperforms the character-aware neural model (Kim et al., 2016) on average across all languages, and its extension which explicitly injects linguistic knowledge (Gerz et al., 2018) on 8 languages. Further experiments on larger Europarl datasets for 3 languages indicate that neural architectures are able to outperform computationally much cheaper n-gram models: n-gram training is up to 15,000x quicker. Our experiments illustrate that standalone n-gram models lend themselves as natural choices for resource-lean or morphologically rich languages, while the recent progress has significantly improved their accuracy.

Dialogue systems benefit greatly from optimizing on detailed annotations, such as transcribed utterances, internal dialogue state representations and dialogue act labels. However, collecting these annotations is expensive and time-consuming, holding back development in the area of dialogue modelling. In this paper, we investigate semi-supervised learning methods that are able to reduce the amount of required intermediate labelling. We find that by leveraging un-annotated data instead, the amount of turn-level annotations of dialogue state can be significantly reduced when building a neural dialogue system. Our analysis on the MultiWOZ corpus, covering a range of domains and topics, finds that annotations can be reduced by up to 30% while maintaining equivalent system performance. We also describe and evaluate the first end-to-end dialogue model created for the MultiWOZ corpus.

Semantic specialization integrates structured linguistic knowledge from external resources (such as lexical relations in WordNet) into pretrained distributional vectors in the form of constraints. However, this technique cannot be leveraged in many languages, because their structured external resources are typically incomplete or non-existent. To bridge this gap, we propose a novel method that transfers specialization from a resource-rich source language (English) to virtually any target language. Our specialization transfer comprises two crucial steps: 1) Inducing noisy constraints in the target language through automatic word translation; and 2) Filtering the noisy constraints via a state-of-the-art relation prediction model trained on the source language constraints. This allows us to specialize any set of distributional vectors in the target language with the refined constraints. We prove the effectiveness of our method through intrinsic word similarity evaluation in 8 languages, and with 3 downstream tasks in 5 languages: lexical simplification, dialog state tracking, and semantic textual similarity. The gains over the previous state-of-art specialization methods are substantial and consistent across languages. Our results also suggest that the transfer method is effective even for lexically distant source-target language pairs. Finally, as a by-product, our method produces lists of WordNet-style lexical relations in resource-poor languages.

Can we construct a neural language model which is inductively biased towards learning human language? Motivated by this question, we aim at constructing an informative prior for held-out languages on the task of character-level, open-vocabulary language modelling. We obtain this prior as the posterior over network weights conditioned on the data from a sample of training languages, which is approximated through Laplace’s method. Based on a large and diverse sample of languages, the use of our prior outperforms baseline models with an uninformative prior in both zero-shot and few-shot settings, showing that the prior is imbued with universal linguistic knowledge. Moreover, we harness broad language-specific information available for most languages of the world, i.e., features from typological databases, as distant supervision for held-out languages. We explore several language modelling conditioning techniques, including concatenation and meta-networks for parameter generation. They appear beneficial in the few-shot setting, but ineffective in the zero-shot setting. Since the paucity of even plain digital text affects the majority of the world’s languages, we hope that these insights will broaden the scope of applications for language technology.

Recent efforts in cross-lingual word embedding (CLWE) learning have predominantly focused on fully unsupervised approaches that project monolingual embeddings into a shared cross-lingual space without any cross-lingual signal. The lack of any supervision makes such approaches conceptually attractive. Yet, their only core difference from (weakly) supervised projection-based CLWE methods is in the way they obtain a seed dictionary used to initialize an iterative self-learning procedure. The fully unsupervised methods have arguably become more robust, and their primary use case is CLWE induction for pairs of resource-poor and distant languages. In this paper, we question the ability of even the most robust unsupervised CLWE approaches to induce meaningful CLWEs in these more challenging settings. A series of bilingual lexicon induction (BLI) experiments with 15 diverse languages (210 language pairs) show that fully unsupervised CLWE methods still fail for a large number of language pairs (e.g., they yield zero BLI performance for 87/210 pairs). Even when they succeed, they never surpass the performance of weakly supervised methods (seeded with 500-1,000 translation pairs) using the same self-learning procedure in any BLI setup, and the gaps are often substantial. These findings call for revisiting the main motivations behind fully unsupervised CLWE methods.

Linguistic typology aims to capture structural and semantic variation across the world’s languages. A large-scale typology could provide excellent guidance for multilingual Natural Language Processing (NLP), particularly for languages that suffer from the lack of human labeled resources. We present an extensive literature survey on the use of typological information in the development of NLP techniques. Our survey demonstrates that to date, the use of information in existing typological databases has resulted in consistent but modest improvements in system performance. We show that this is due to both intrinsic limitations of databases (in terms of coverage and feature granularity) and under-utilization of the typological features included in them. We advocate for a new approach that adapts the broad and discrete nature of typological categories to the contextual and continuous nature of machine learning algorithms used in contemporary NLP. In particular, we suggest that such an approach could be facilitated by recent developments in data-driven induction of typological knowledge.

Verbs play a fundamental role in many biomed-ical tasks and applications such as relation and event extraction. We hypothesize that performance on many downstream tasks can be improved by aligning the input pretrained embeddings according to semantic verb classes. In this work, we show that by using semantic clusters for verbs, a large lexicon of verbclasses derived from biomedical literature, weare able to improve the performance of common pretrained embeddings in downstream tasks by retrofitting them to verb classes. We present a simple and computationally efficient approach using a widely-available “off-the-shelf” retrofitting algorithm to align pretrained embeddings according to semantic verb clusters. We achieve state-of-the-art results on text classification and relation extraction tasks.

Word vector specialisation (also known as retrofitting) is a portable, light-weight approach to fine-tuning arbitrary distributional word vector spaces by injecting external knowledge from rich lexical resources such as WordNet. By design, these post-processing methods only update the vectors of words occurring in external lexicons, leaving the representations of all unseen words intact. In this paper, we show that constraint-driven vector space specialisation can be extended to unseen words. We propose a novel post-specialisation method that: a) preserves the useful linguistic knowledge for seen words; while b) propagating this external signal to unseen words in order to improve their vector representations as well. Our post-specialisation approach explicits a non-linear specialisation function in the form of a deep neural network by learning to predict specialised vectors from their original distributional counterparts. The learned function is then used to specialise vectors of unseen words. This approach, applicable to any post-processing model, yields considerable gains over the initial specialisation models both in intrinsic word similarity tasks, and in two downstream tasks: dialogue state tracking and lexical text simplification. The positive effects persist across three languages, demonstrating the importance of specialising the full vocabulary of distributional word vector spaces.

The transfer or share of knowledge between languages is a potential solution to resource scarcity in NLP. However, the effectiveness of cross-lingual transfer can be challenged by variation in syntactic structures. Frameworks such as Universal Dependencies (UD) are designed to be cross-lingually consistent, but even in carefully designed resources trees representing equivalent sentences may not always overlap. In this paper, we measure cross-lingual syntactic variation, or anisomorphism, in the UD treebank collection, considering both morphological and structural properties. We show that reducing the level of anisomorphism yields consistent gains in cross-lingual transfer tasks. We introduce a source language selection procedure that facilitates effective cross-lingual parser transfer, and propose a typologically driven method for syntactic tree processing which reduces anisomorphism. Our results show the effectiveness of this method for both machine translation and cross-lingual sentence similarity, demonstrating the importance of syntactic structure compatibility for boosting cross-lingual transfer in NLP.

Neural architectures are prominent in the construction of language models (LMs). However, word-level prediction is typically agnostic of subword-level information (characters and character sequences) and operates over a closed vocabulary, consisting of a limited word set. Indeed, while subword-aware models boost performance across a variety of NLP tasks, previous work did not evaluate the ability of these models to assist next-word prediction in language modeling tasks. Such subword-level informed models should be particularly effective for morphologically-rich languages (MRLs) that exhibit high type-to-token ratios. In this work, we present a large-scale LM study on 50 typologically diverse languages covering a wide variety of morphological systems, and offer new LM benchmarks to the community, while considering subword-level information. The main technical contribution of our work is a novel method for injecting subword-level information into semantic word vectors, integrated into the neural language modeling training, to facilitate word-level prediction. We conduct experiments in the LM setting where the number of infrequent words is large, and demonstrate strong perplexity gains across our 50 languages, especially for morphologically-rich languages. Our code and data sets are publicly available.

Semantic specialization is a process of fine-tuning pre-trained distributional word vectors using external lexical knowledge (e.g., WordNet) to accentuate a particular semantic relation in the specialized vector space. While post-processing specialization methods are applicable to arbitrary distributional vectors, they are limited to updating only the vectors of words occurring in external lexicons (i.e., seen words), leaving the vectors of all other words unchanged. We propose a novel approach to specializing the full distributional vocabulary. Our adversarial post-specialization method propagates the external lexical knowledge to the full distributional space. We exploit words seen in the resources as training examples for learning a global specialization function. This function is learned by combining a standard L2-distance loss with a adversarial loss: the adversarial component produces more realistic output vectors. We show the effectiveness and robustness of the proposed method across three languages and on three tasks: word similarity, dialog state tracking, and lexical simplification. We report consistent improvements over distributional word vectors and vectors specialized by other state-of-the-art specialization frameworks. Finally, we also propose a cross-lingual transfer method for zero-shot specialization which successfully specializes a full target distributional space without any lexical knowledge in the target language and without any bilingual data.

A key challenge in cross-lingual NLP is developing general language-independent architectures that are equally applicable to any language. However, this ambition is largely hampered by the variation in structural and semantic properties, i.e. the typological profiles of the world’s languages. In this work, we analyse the implications of this variation on the language modeling (LM) task. We present a large-scale study of state-of-the art n-gram based and neural language models on 50 typologically diverse languages covering a wide variety of morphological systems. Operating in the full vocabulary LM setup focused on word-level prediction, we demonstrate that a coarse typology of morphological systems is predictive of absolute LM performance. Moreover, fine-grained typological features such as exponence, flexivity, fusion, and inflectional synthesis are borne out to be responsible for the proliferation of low-frequency phenomena which are organically difficult to model by statistical architectures, or for the meaning ambiguity of character n-grams. Our study strongly suggests that these features have to be taken into consideration during the construction of next-level language-agnostic LM architectures, capable of handling morphologically complex languages such as Tamil or Korean.

This paper is concerned with identifying contexts useful for training word representation models for different word classes such as adjectives (A), verbs (V), and nouns (N). We introduce a simple yet effective framework for an automatic selection of class-specific context configurations. We construct a context configuration space based on universal dependency relations between words, and efficiently search this space with an adapted beam search algorithm. In word similarity tasks for each word class, we show that our framework is both effective and efficient. Particularly, it improves the Spearman’s rho correlation with human scores on SimLex-999 over the best previously proposed class-specific contexts by 6 (A), 6 (V) and 5 (N) rho points. With our selected context configurations, we train on only 14% (A), 26.2% (V), and 33.6% (N) of all dependency-based contexts, resulting in a reduced training time. Our results generalise: we show that the configurations our algorithm learns for one English training setup outperform previously proposed context types in another training setup for English. Moreover, basing the configuration space on universal dependencies, it is possible to transfer the learned configurations to German and Italian. We also demonstrate improved per-class results over other context types in these two languages..

We introduce HyperLex—a data set and evaluation resource that quantifies the extent of the semantic category membership, that is, type-of relation, also known as hyponymy–hypernymy or lexical entailment (LE) relation between 2,616 concept pairs. Cognitive psychology research has established that typicality and category/class membership are computed in human semantic memory as a gradual rather than binary relation. Nevertheless, most NLP research and existing large-scale inventories of concept category membership (WordNet, DBPedia, etc.) treat category membership and LE as binary. To address this, we asked hundreds of native English speakers to indicate typicality and strength of category membership between a diverse range of concept pairs on a crowdsourcing platform. Our results confirm that category membership and LE are indeed more gradual than binary. We then compare these human judgments with the predictions of automatic systems, which reveals a huge gap between human performance and state-of-the-art LE, distributional and representation learning models, and substantial differences between the models themselves. We discuss a pathway for improving semantic models to overcome this discrepancy, and indicate future application areas for improved graded LE systems.

Distributed representations of sentences have been developed recently to represent their meaning as real-valued vectors. However, it is not clear how much information such representations retain about the polarity of sentences. To study this question, we decode sentiment from unsupervised sentence representations learned with different architectures (sensitive to the order of words, the order of sentences, or none) in 9 typologically diverse languages. Sentiment results from the (recursive) composition of lexical items and grammatical strategies such as negation and concession. The results are manifold: we show that there is no ‘one-size-fits-all’ representation architecture outperforming the others across the board. Rather, the top-ranking architectures depend on the language at hand. Moreover, we find that in several cases the additive composition model based on skip-gram word vectors may surpass supervised state-of-art architectures such as bi-directional LSTMs. Finally, we provide a possible explanation of the observed variation based on the type of negative constructions in each language.

Existing approaches to automatic VerbNet-style verb classification are heavily dependent on feature engineering and therefore limited to languages with mature NLP pipelines. In this work, we propose a novel cross-lingual transfer method for inducing VerbNets for multiple languages. To the best of our knowledge, this is the first study which demonstrates how the architectures for learning word embeddings can be applied to this challenging syntactic-semantic task. Our method uses cross-lingual translation pairs to tie each of the six target languages into a bilingual vector space with English, jointly specialising the representations to encode the relational information from English VerbNet. A standard clustering algorithm is then run on top of the VerbNet-specialised representations, using vector dimensions as features for learning verb classes. Our results show that the proposed cross-lingual transfer approach sets new state-of-the-art verb classification performance across all six target languages explored in this work.

Recent work on evaluating representation learning architectures in NLP has established a need for evaluation protocols based on subconscious cognitive measures rather than manually tailored intrinsic similarity and relatedness tasks. In this work, we propose a novel evaluation framework that enables large-scale evaluation of such architectures in the free word association (WA) task, which is firmly grounded in cognitive theories of human semantic representation. This evaluation is facilitated by the existence of large manually constructed repositories of word association data. In this paper, we (1) present a detailed analysis of the new quantitative WA evaluation protocol, (2) suggest new evaluation metrics for the WA task inspired by its direct analogy with information retrieval problems, (3) evaluate various state-of-the-art representation models on this task, and (4) discuss the relationship between WA and prior evaluations of semantic representation with well-known similarity and relatedness evaluation sets. We have made the WA evaluation toolkit publicly available.

We present Attract-Repel, an algorithm for improving the semantic quality of word vectors by injecting constraints extracted from lexical resources. Attract-Repel facilitates the use of constraints from mono- and cross-lingual resources, yielding semantically specialized cross-lingual vector spaces. Our evaluation shows that the method can make use of existing cross-lingual lexicons to construct high-quality vector spaces for a plethora of different languages, facilitating semantic transfer from high- to lower-resource ones. The effectiveness of our approach is demonstrated with state-of-the-art results on semantic similarity datasets in six languages. We next show that Attract-Repel-specialized vectors boost performance in the downstream task of dialogue state tracking (DST) across multiple languages. Finally, we show that cross-lingual vector spaces produced by our algorithm facilitate the training of multilingual DST models, which brings further performance improvements.

Causal relations play a key role in information extraction and reasoning. Most of the times, their expression is ambiguous or implicit, i.e. without signals in the text. This makes their identification challenging. We aim to improve their identification by implementing a Feedforward Neural Network with a novel set of features for this task. In particular, these are based on the position of event mentions and the semantics of events and participants. The resulting classifier outperforms strong baselines on two datasets (the Penn Discourse Treebank and the CSTNews corpus) annotated with different schemes and containing examples in two languages, English and Portuguese. This result demonstrates the importance of events for identifying discourse relations.

Many tasks in the biomedical domain require the assignment of one or more predefined labels to input text, where the labels are a part of a hierarchical structure (such as a taxonomy). The conventional approach is to use a one-vs.-rest (OVR) classification setup, where a binary classifier is trained for each label in the taxonomy or ontology where all instances not belonging to the class are considered negative examples. The main drawbacks to this approach are that dependencies between classes are not leveraged in the training and classification process, and the additional computational cost of training parallel classifiers. In this paper, we apply a new method for hierarchical multi-label text classification that initializes a neural network model final hidden layer such that it leverages label co-occurrence relations such as hypernymy. This approach elegantly lends itself to hierarchical classification. We evaluated this approach using two hierarchical multi-label text classification tasks in the biomedical domain using both sentence- and document-level classification. Our evaluation shows promising results for this approach.

Morphologically rich languages accentuate two properties of distributional vector space models: 1) the difficulty of inducing accurate representations for low-frequency word forms; and 2) insensitivity to distinct lexical relations that have similar distributional signatures. These effects are detrimental for language understanding systems, which may infer that ‘inexpensive’ is a rephrasing for ‘expensive’ or may not associate ‘acquire’ with ‘acquires’. In this work, we propose a novel morph-fitting procedure which moves past the use of curated semantic lexicons for improving distributional vector spaces. Instead, our method injects morphological constraints generated using simple language-specific rules, pulling inflectional forms of the same word close together and pushing derivational antonyms far apart. In intrinsic evaluation over four languages, we show that our approach: 1) improves low-frequency word estimates; and 2) boosts the semantic quality of the entire word vector collection. Finally, we show that morph-fitted vectors yield large gains in the downstream task of dialogue state tracking, highlighting the importance of morphology for tackling long-tail phenomena in language understanding tasks.

Distributional models that learn rich semantic word representations are a success story of recent NLP research. However, developing models that learn useful representations of phrases and sentences has proved far harder. We propose using the definitions found in everyday dictionaries as a means of bridging this gap between lexical and phrasal semantics. Neural language embedding models can be effectively trained to map dictionary definitions (phrases) to (lexical) representations of the words defined by those definitions. We present two applications of these architectures: reverse dictionaries that return the name of a concept given a definition or description and general-knowledge crossword question answerers. On both tasks, neural language embedding models trained on definitions from a handful of freely-available lexical resources perform as well or better than existing commercial systems that rely on significant task-specific engineering. The results highlight the effectiveness of both neural embedding architectures and definition-based training for developing models that understand phrases and sentences.

Methods based on deep learning approaches have recently achieved state-of-the-art performance in a range of machine learning tasks and are increasingly applied to natural language processing (NLP). Despite strong results in various established NLP tasks involving general domain texts, there is only limited work applying these models to biomedical NLP. In this paper, we consider a Convolutional Neural Network (CNN) approach to biomedical text classification. Evaluation using a recently introduced cancer domain dataset involving the categorization of documents according to the well-established hallmarks of cancer shows that a basic CNN model can achieve a level of performance competitive with a Support Vector Machine (SVM) trained using complex manually engineered features optimized to the task. We further show that simple modifications to the CNN hyperparameters, initialization, and training process allow the model to notably outperform the SVM, establishing a new state of the art result at this task. We make all of the resources and tools introduced in this study available under open licenses from https://cambridgeltl.github.io/cancer-hallmark-cnn/.

In recent years linguistic typologies, which classify the world’s languages according to their functional and structural properties, have been widely used to support multilingual NLP. While the growing importance of typologies in supporting multilingual tasks has been recognised, no systematic survey of existing typological resources and their use in NLP has been published. This paper provides such a survey as well as discussion which we hope will both inform and inspire future work in the area.

The conventional solution for handling sparsely labelled data is extensive feature engineering. This is time consuming and task and domain specific. We present a novel approach for learning embedded features that aims to alleviate this problem. Our approach jointly learns embeddings at different levels of granularity (word, sentence and document) along with the class labels. The intuition is that topic semantics represented by embeddings at multiple levels results in better classification. We evaluate this approach in unsupervised and semi-supervised settings on two sparsely labelled classification tasks, outperforming the handcrafted models and several embedding baselines.

Inferring the information structure of scientific documents is useful for many NLP applications. Existing approaches to this task require substantial human effort. We propose a framework for constraint learning that reduces human involvement considerably. Our model uses topic models to identify latent topics and their key linguistic features in input documents, induces constraints from this information and maps sentences to their dominant information structure categories through a constrained unsupervised model. When the induced constraints are combined with a fully unsupervised model, the resulting model challenges existing lightly supervised feature-based models as well as unsupervised models that use manually constructed declarative knowledge. Our results demonstrate that useful declarative knowledge can be learned from data with very limited human involvement.

Native Language Identification (NLI) is a task aimed at determining the native language (L1) of learners of second language (L2) on the basis of their written texts. To date, research on NLI has focused on relatively small corpora. We apply NLI to the recently released EFCamDat corpus which is not only multiple times larger than previous L2 corpora but also provides longitudinal data at several proficiency levels. Our investigation using accurate machine learning with a wide range of linguistic features reveals interesting patterns in the longitudinal data which are useful for both further development of NLI and its application to research on L2 acquisition.

Multi-modal models that learn semantic representations from both linguistic and perceptual input outperform language-only models on a range of evaluations, and better reflect human concept acquisition. Most perceptual input to such models corresponds to concrete noun concepts and the superiority of the multi-modal approach has only been established when evaluating on such concepts. We therefore investigate which concepts can be effectively learned by multi-modal models. We show that concreteness determines both which linguistic features are most informative and the impact of perceptual input in such models. We then introduce ridge regression as a means of propagating perceptual information from concrete nouns to more abstract concepts that is more robust than previous approaches. Finally, we present weighted gram matrix combination, a means of combining representations from distinct modalities that outperforms alternatives when both modalities are sufficiently rich.

The Enron Email Corpus provides ``Real World'' text in the business email domain, which is a target domain for many speech and language applications. We present a section of this corpus annotated with number senses - labelling each number as a date, time, year, telephone number etc. We show that sense categories and their frequencies are very different in this domain than in newswire text. The annotated corpus can provide valuable material for the development of number sense disambiguation techniques. We have released the annotations into the public domain, to allow other researchers to perform comparisons.

This paper presents LexSchem - the first large, fully automatically acquired subcategorization lexicon for French verbs. The lexicon includes subcategorization frame and frequency information for 3297 French verbs. When evaluated on a set of 20 test verbs against a gold standard dictionary, it shows 0.79 precision, 0.55 recall and 0.65 F-measure. We have made this resource freely available to the research community on the web.

Lexical classifications have proved useful in supporting various natural language processing (NLP) tasks. The largest verb classification for English is Levin's (1993) work which defined groupings of verbs based on syntactic properties. VerbNet - the largest computational verb lexicon currently available for English - provides detailed syntactic-semantic descriptions of Levin classes. While the classes included are extensive enough for some NLP use, they are not comprehensive. Korhonen and Briscoe (2004) have proposed a significant extension of Levin's classification which incorporates 57 novel classes for verbs not covered (comprehensively) by Levin. This paper describes the integration of these classes into VerbNet. The result is the most extensive Levin-style classification for English verbs which can be highly useful for practical applications.

We introduce a large computational subcategorizationlexicon which includes subcategorization frame (SCF) and frequencyinformation for 6,397 English verbs. This extensive lexicon was acquiredautomatically from five corpora and the Web using the current version of the comprehensive subcategorization acquisition system of Briscoe and Carroll (1997). The lexicon is provided freely for research use, along with a script which can be used to filter and build sub-lexicons suited for different natural languageprocessing (NLP) purposes. Documentation is also provided whichexplains each sub-lexicon option and evaluates its accuracy.