Large language models (LLMs) have achieved remarkable performance on a variety of natural language understanding tasks. However, existing benchmarks are inadequate in measuring the complex logical reasoning capabilities of a model. We present FOLIO, a human-annotated, logically complex and diverse dataset for reasoning in natural language (NL), equipped with first-order logic (FOL) annotations. FOLIO consists of 1,430 examples (unique conclusions), each paired with one of 487 sets of premises used to deductively reason for the validity of each conclusion. The logical correctness of the premises and conclusions is ensured by their FOL annotations, which are automatically verified by an FOL inference engine. In addition to the main NL reasoning task, NL-FOL pairs in FOLIO constitute a new NL-FOL translation dataset. Our experiments on FOLIO systematically evaluate the FOL reasoning ability of supervised fine-tuning on medium-sized language models. For both NL reasoning and NL-FOL translation, we benchmark multiple state-of-the-art language models. Our results show that a subset of FOLIO remains a challenge for one of the most capable Large Language Model (LLM) publicly available, GPT-4.

Generative retrieval (Wang et al., 2022; Tay et al., 2022) is a popular approach for end-to-end document retrieval that directly generates document identifiers given an input query. We introduce summarization-based document IDs, in which each document’s ID is composed of an extractive summary or abstractive keyphrases generated by a language model, rather than an integer ID sequence or bags of n-grams as proposed in past work. We find that abstractive, content-based IDs (ACID) and an ID based on the first 30 tokens are very effective in direct comparisons with previous approaches to ID creation. We show that using ACID improves top-10 and top-20 recall by 15.6% and 14.4% (relative) respectively versus the cluster-based integer ID baseline on the MSMARCO 100k retrieval task, and 9.8% and 9.9% respectively on the Wikipedia-based NQ 100k retrieval task. Our results demonstrate the effectiveness of human-readable, natural-language IDs created through summarization for generative retrieval. We also observed that extractive IDs outperformed abstractive IDs on Wikipedia articles in NQ but not the snippets in MSMARCO, which suggests that document characteristics affect generative retrieval performance.

Text generation with beam search has proven successful in a wide range of applications. We point out that, though largely overlooked in the literature, the commonly-used implementation of beam decoding (e.g., Hugging Face Transformers and fairseq) uses a first come, first served heuristic: it keeps a set of already completed sequences over time steps and stops when the size of this set reaches the beam size. Based on this finding, we introduce a patience factor, a simple modification to this beam decoding implementation, that generalizes the stopping criterion and provides flexibility to the depth of search. Empirical results demonstrate that adjusting this patience factor improves decoding performance of strong pretrained models on news text summarization and machine translation over diverse language pairs, with a negligible inference slowdown. Our approach only modifies one line of code and can be thus readily incorporated in any implementation. Further, we find that different versions of beam decoding result in large performance differences in summarization, demonstrating the need for clarity in specifying the beam search implementation in research work. Our code will be available upon publication.

The BLOOM model is a large publicly available multilingual language model, but its pretraining was limited to 46 languages. To extend the benefits of BLOOM to other languages without incurring prohibitively large costs, it is desirable to adapt BLOOM to new languages not seen during pretraining. In this work, we apply existing language adaptation strategies to BLOOM and benchmark its zero-shot prompting performance on eight new languages in a resource-constrained setting. We find language adaptation to be effective at improving zero-shot performance in new languages. Surprisingly, we find that adapter-based finetuning is more effective than continued pretraining for large models. In addition, we discover that prompting performance is not significantly affected by language specifics, such as the writing system. It is primarily determined by the size of the language adaptation data. We also add new languages to BLOOMZ, which is a multitask finetuned version of BLOOM capable of following task instructions zero-shot. We find including a new language in the multitask fine-tuning mixture to be the most effective method to teach BLOOMZ a new language. We conclude that with sufficient training data language adaptation can generalize well to diverse languages. Our code is available at https://github.com/bigscience-workshop/multilingual-modeling.

Large-scale language model pretraining is a very successful form of self-supervised learning in natural language processing, but it is increasingly expensive to perform as the models and pretraining corpora have become larger over time. We propose NarrowBERT, a modified transformer encoder that increases the throughput for masked language model pretraining by more than 2x. NarrowBERT sparsifies the transformer model such that the self-attention queries and feedforward layers only operate on the masked tokens of each sentence during pretraining, rather than all of the tokens as with the usual transformer encoder. We also show that NarrowBERT increases the throughput at inference time by as much as 3.5x with minimal (or no) performance degradation on sentence encoding tasks like MNLI. Finally, we examine the performance of NarrowBERT on the IMDB and Amazon reviews classification and CoNLL NER tasks and show that it is also comparable to standard BERT performance.

We introduce INSTRUCTOR, a new method for computing text embeddings given task instructions: every text input is embedded together with instructions explaining the use case (e.g., task and domain descriptions). Unlike encoders from prior work that are more specialized, INSTRUCTOR is a single embedder that can generate text embeddings tailored to different downstream tasks and domains, without any further training. We first annotate instructions for 330 diverse tasks and train INSTRUCTOR on this multitask mixture with a contrastive loss. We evaluate INSTRUCTOR on 70 embedding evaluation tasks (66 of which are unseen during training), ranging from classification and information retrieval to semantic textual similarity and text generation evaluation. INSTRUCTOR, while having an order of magnitude fewer parameters than the previous best model, achieves state-of-the-art performance, with an average improvement of 3.4% compared to the previous best results on the 70 diverse datasets. Our analysis suggests that INSTRUCTOR is robust to changes in instructions, and that instruction finetuning mitigates the challenge of training a single model on diverse datasets. Our model, code, and data are available at https://instructor-embedding.github.io.

Language models have graduated from being research prototypes to commercialized products offered as web APIs, and recent works have highlighted the multilingual capabilities of these products. The API vendors charge their users based on usage, more specifically on the number of “tokens” processed or generated by the underlying language models. What constitutes a token, however, is training data and model dependent with a large variance in the number of tokens required to convey the same information in different languages. In this work, we analyze the effect of this non-uniformity on the fairness of an API’s pricing policy across languages. We conduct a systematic analysis of the cost and utility of OpenAI’s language model API on multilingual benchmarks in 22 typologically diverse languages. We show evidence that speakers of a large number of the supported languages are overcharged while obtaining poorer results. These speakers tend to also come from regions where the APIs are less affordable, to begin with. Through these analyses, we aim to increase transparency around language model APIs’ pricing policies and encourage the vendors to make them more equitable.

Performing inference on large volumes of samples with large language models (LLMs) can be computationally and financially costly in industry and real-world use. We propose batch prompting, a simple yet effective prompting approach that enables the LLM to run inference in batches, instead of one sample at a time. Our method reduces both token and time costs while retaining downstream performance. We theoretically demonstrate that under a few-shot in-context learning setting, the inference costs decrease almost inverse linearly with the number of samples in each batch. We extensively validate the effectiveness of batch prompting on ten datasets across commonsense QA, arithmetic reasoning, and NLI/NLU: batch prompting significantly (up to 5× with six samples in batch) reduces the LLM (Codex) inference token and time costs while achieving better or comparable performance. For state-of-the-art Chat-based LLMs, e.g., GPT-3.5 and GPT-4, we show the benefits of batch prompting also hold. Further analysis shows that the number of samples in each batch and the complexity of tasks affect its performance. Moreover, batch prompting can be applied across different reasoning methods using LLMs. Our code is released at the site https://github.com/xlang-ai/batch-prompting.

The dominant paradigm for neural text generation is left-to-right decoding from autoregressive language models. Constrained or controllable generation under complex lexical constraints, however, requires foresight to plan ahead feasible future paths. Drawing inspiration from the A^* search algorithm, we propose NeuroLogic A*esque, a decoding algorithm that incorporates heuristic estimates of future cost. We develop lookahead heuristics that are efficient for large-scale language models, making our method a drop-in replacement for common techniques such as beam search and top-k sampling. To enable constrained generation, we build on NeuroLogic decoding (Lu et al., 2021), combining its flexibility in incorporating logical constraints with A*esque estimates of future constraint satisfaction. Our approach outperforms competitive baselines on five generation tasks, and achieves new state-of-the-art performance on table-to-text generation, constrained machine translation, and keyword-constrained generation. The improvements are particularly notable on tasks that require complex constraint satisfaction or in few-shot or zero-shot settings. NeuroLogic A*esque illustrates the power of decoding for improving and enabling new capabilities of large-scale language models.

We establish THumB, a rubric-based human evaluation protocol for image captioning models. Our scoring rubrics and their definitions are carefully developed based on machine- and human-generated captions on the MSCOCO dataset. Each caption is evaluated along two main dimensions in a tradeoff (precision and recall) as well as other aspects that measure the text quality (fluency, conciseness, and inclusive language). Our evaluations demonstrate several critical problems of the current evaluation practice. Human-generated captions show substantially higher quality than machine-generated ones, especially in coverage of salient information (i.e., recall), while most automatic metrics say the opposite. Our rubric-based results reveal that CLIPScore, a recent metric that uses image features, better correlates with human judgments than conventional text-only metrics because it is more sensitive to recall. We hope that this work will promote a more transparent evaluation protocol for image captioning and its automatic metrics.

Natural language processing researchers have identified limitations of evaluation methodology for generation tasks, with new questions raised about the validity of automatic metrics and of crowdworker judgments. Meanwhile, efforts to improve generation models tend to depend on simple n-gram overlap metrics (e.g., BLEU, ROUGE). We argue that new advances on models and metrics should each more directly benefit and inform the other. We therefore propose a generalization of leaderboards, bidimensional leaderboards (Billboards), that simultaneously tracks progress in language generation models and metrics for their evaluation. Unlike conventional unidimensional leaderboards that sort submitted systems by predetermined metrics, a Billboard accepts both generators and evaluation metrics as competing entries. A Billboard automatically creates an ensemble metric that selects and linearly combines a few metrics based on a global analysis across generators. Further, metrics are ranked based on their correlation with human judgments. We release four Billboards for machine translation, summarization, and image captioning. We demonstrate that a linear ensemble of a few diverse metrics sometimes substantially outperforms existing metrics in isolation. Our mixed-effects model analysis shows that most automatic metrics, especially the reference-based ones, overrate machine over human generation, demonstrating the importance of updating metrics as generation models become stronger (and perhaps more similar to humans) in the future.

Transformer architectures have achieved state- of-the-art results on a variety of natural language processing (NLP) tasks. However, their attention mechanism comes with a quadratic complexity in sequence lengths, making the computational overhead prohibitive, especially for long sequences. Attention context can be seen as a random-access memory with each token taking a slot. Under this perspective, the memory size grows linearly with the sequence length, and so does the overhead of reading from it. One way to improve the efficiency is to bound the memory size. We show that disparate approaches can be subsumed into one abstraction, attention with bounded-memory control (ABC), and they vary in their organization of the memory. ABC reveals new, unexplored possibilities. First, it connects several efficient attention variants that would otherwise seem apart. Second, this abstraction gives new insights—an established approach (Wang et al., 2020b) previously thought to not be applicable in causal attention, actually is. Last, we present a new instance of ABC, which draws inspiration from existing ABC approaches, but replaces their heuristic memory-organizing functions with a learned, contextualized one. Our experiments on language modeling, machine translation, and masked language model finetuning show that our approach outperforms previous efficient attention models; compared to the strong transformer baselines, it significantly improves the inference time and space efficiency with no or negligible accuracy loss.

Many language generation models are now available for a wide range of generation tasks, including machine translation and summarization. Combining such diverse models may lead to further progress, but ensembling generation models is challenging during inference: conventional ensembling methods (e.g., shallow fusion) require that the models share vocabulary/tokenization schemes. We introduce Twist decoding, a simple and general text generation algorithm that benefits from diverse models at inference time. Our method does not assume the vocabulary, tokenization or even generation order is shared. Our extensive evaluations on machine translation and scientific paper summarization demonstrate that Twist decoding substantially outperforms each model decoded in isolation over various scenarios, including cases where domain-specific and general-purpose models are both available. Twist decoding also consistently outperforms the popular reranking heuristic where output candidates from one model are rescored by another. We hope that our work will encourage researchers and practitioners to examine generation models collectively, not just independently, and to seek out models with complementary strengths to the currently available models.

While often assumed a gold standard, effective human evaluation of text generation remains an important, open area for research.We revisit this problem with a focus on producing consistent evaluations that are reproducible—over time and across different populations. We study this goal in different stages of the human evaluation pipeline. In particular, we consider design choices for the annotation interface used to elicit human judgments and their impact on reproducibility. Furthermore, we develop an automated mechanism for maintaining annotator quality via a probabilistic model that detects and excludes noisy annotators. Putting these lessons together, we introduce GENIE: a system for running standardized human evaluations across different generation tasks.We instantiate GENIE with datasets representing four core challenges in text generation: machine translation, summarization, commonsense reasoning, and machine comprehension.For each task, GENIE offers a leaderboard that automatically crowdsources annotations for submissions, evaluating them along axes such as correctness, conciseness, and fluency.We have made the GENIE leaderboards publicly available, and have already ranked 50 submissions from 10 different research groups. We hope GENIE encourages further progress toward effective, standardized evaluations for text generation.

The attention mechanism is considered the backbone of the widely-used Transformer architecture. It contextualizes the input by computing input-specific attention matrices. We find that this mechanism, while powerful and elegant, is not as important as typically thought for pretrained language models. We introduce PAPA, a new probing method that replaces the input-dependent attention matrices with constant ones—the average attention weights over multiple inputs. We use PAPA to analyze several established pretrained Transformers on six downstream tasks. We find that without any input-dependent attention, all models achieve competitive performance—an average relative drop of only 8% from the probing baseline. Further, little or no performance drop is observed when replacing half of the input-dependent attention matrices with constant (input-independent) ones. Interestingly, we show that better-performing models lose more from applying our method than weaker models, suggesting that the utilization of the input-dependent attention mechanism might be a factor in their success. Our results motivate research on simpler alternatives to input-dependent attention, as well as on methods for better utilization of this mechanism in the Transformer architecture.

We present the results of the Workshop on Multilingual Information Access (MIA) 2022 Shared Task, evaluating cross-lingual open-retrieval question answering (QA) systems in 16 typologically diverse languages. In this task, we adapted two large-scale cross-lingual open-retrieval QA datasets in 14 typologically diverse languages, and newly annotated open-retrieval QA data in 2 underrepresented languages: Tagalog and Tamil. Four teams submitted their systems. The best constrained system uses entity-aware contextualized representations for document retrieval, thereby achieving an average F1 score of 31.6, which is 4.1 F1 absolute higher than the challenging baseline. The best system obtains particularly significant improvements in Tamil (20.8 F1), whereas most of the other systems yield nearly zero scores. The best unconstrained system achieves 32.2 F1, outperforming our baseline by 4.5 points.

Multilingual question answering tasks typically assume that answers exist in the same language as the question. Yet in practice, many languages face both information scarcity—where languages have few reference articles—and information asymmetry—where questions reference concepts from other cultures. This work extends open-retrieval question answering to a cross-lingual setting enabling questions from one language to be answered via answer content from another language. We construct a large-scale dataset built on 40K information-seeking questions across 7 diverse non-English languages that TyDi QA could not find same-language answers for. Based on this dataset, we introduce a task framework, called Cross-lingual Open-Retrieval Question Answering (XOR QA), that consists of three new tasks involving cross-lingual document retrieval from multilingual and English resources. We establish baselines with state-of-the-art machine translation systems and cross-lingual pretrained models. Experimental results suggest that XOR QA is a challenging task that will facilitate the development of novel techniques for multilingual question answering. Our data and code are available at https://nlp.cs.washington.edu/xorqa/.

This paper presents the results of the newstranslation task, the multilingual low-resourcetranslation for Indo-European languages, thetriangular translation task, and the automaticpost-editing task organised as part of the Con-ference on Machine Translation (WMT) 2021.In the news task, participants were asked tobuild machine translation systems for any of10 language pairs, to be evaluated on test setsconsisting mainly of news stories. The taskwas also opened up to additional test suites toprobe specific aspects of translation.

Models of language trained on very large corpora have been demonstrated useful for natural language processing. As fixed artifacts, they have become the object of intense study, with many researchers “probing” the extent to which they acquire and readily demonstrate linguistic abstractions, factual and commonsense knowledge, and reasoning abilities. Recent work applied several probes to intermediate training stages to observe the developmental process of a large-scale model (Chiang et al., 2020). Following this effort, we systematically answer a question: for various types of knowledge a language model learns, when during (pre)training are they acquired? Using RoBERTa as a case study, we find: linguistic knowledge is acquired fast, stably, and robustly across domains. Facts and commonsense are slower and more domain-sensitive. Reasoning abilities are, in general, not stably acquired. As new datasets, pretraining protocols, and probes emerge, we believe that probing-across-time analyses can help researchers understand the complex, intermingled learning that these models undergo and guide us toward more efficient approaches that accomplish necessary learning faster.

Transformers have outperformed recurrent neural networks (RNNs) in natural language generation. But this comes with a signifi- cant computational cost, as the attention mechanism’s complexity scales quadratically with sequence length. Efficient transformer variants have received increasing interest in recent works. Among them, a linear-complexity recurrent variant has proven well suited for autoregressive generation. It approximates the softmax attention with randomized or heuristic feature maps, but can be difficult to train and may yield suboptimal accuracy. This work aims to convert a pretrained transformer into its efficient recurrent counterpart, improving efficiency while maintaining accuracy. Specifically, we propose a swap-then-finetune procedure: in an off-the-shelf pretrained transformer, we replace the softmax attention with its linear-complexity recurrent alternative and then finetune. With a learned feature map, our approach provides an improved tradeoff between efficiency and accuracy over the standard transformer and other recurrent variants. We also show that the finetuning process has lower training cost relative to training these recurrent variants from scratch. As many models for natural language tasks are increasingly dependent on large-scale pretrained transformers, this work presents a viable approach to improving inference efficiency without repeating the expensive pretraining process.

Entity resolution (ER) is the task of identifying different representations of the same real-world entities across databases. It is a key step for knowledge base creation and text mining. Recent adaptation of deep learning methods for ER mitigates the need for dataset-specific feature engineering by constructing distributed representations of entity records. While these methods achieve state-of-the-art performance over benchmark data, they require large amounts of labeled data, which are typically unavailable in realistic ER applications. In this paper, we develop a deep learning-based method that targets low-resource settings for ER through a novel combination of transfer learning and active learning. We design an architecture that allows us to learn a transferable model from a high-resource setting to a low-resource one. To further adapt to the target dataset, we incorporate active learning that carefully selects a few informative examples to fine-tune the transferred model. Empirical evaluation demonstrates that our method achieves comparable, if not better, performance compared to state-of-the-art learning-based methods while using an order of magnitude fewer labels.

Despite advances in dependency parsing, languages with small treebanks still present challenges. We assess recent approaches to multilingual contextual word representations (CWRs), and compare them for crosslingual transfer from a language with a large treebank to a language with a small or nonexistent treebank, by sharing parameters between languages in the parser itself. We experiment with a diverse selection of languages in both simulated and truly low-resource scenarios, and show that multilingual CWRs greatly facilitate low-resource dependency parsing even without crosslingual supervision such as dictionaries or parallel text. Furthermore, we examine the non-contextual part of the learned language models (which we call a “decontextual probe”) to demonstrate that polyglot language models better encode crosslingual lexical correspondence compared to aligned monolingual language models. This analysis provides further evidence that polyglot training is an effective approach to crosslingual transfer.

We introduce a new syntax-aware model for dependency-based semantic role labeling that outperforms syntax-agnostic models for English and Spanish. We use a BiLSTM to tag the text with supertags extracted from dependency parses, and we feed these supertags, along with words and parts of speech, into a deep highway BiLSTM for semantic role labeling. Our model combines the strengths of earlier models that performed SRL on the basis of a full dependency parse with more recent models that use no syntactic information at all. Our local and non-ensemble model achieves state-of-the-art performance on the CoNLL 09 English and Spanish datasets. SRL models benefit from syntactic information, and we show that supertagging is a simple, powerful, and robust way to incorporate syntax into a neural SRL system.

We introduce Rosita, a method to produce multilingual contextual word representations by training a single language model on text from multiple languages. Our method combines the advantages of contextual word representations with those of multilingual representation learning. We produce language models from dissimilar language pairs (English/Arabic and English/Chinese) and use them in dependency parsing, semantic role labeling, and named entity recognition, with comparisons to monolingual and non-contextual variants. Our results provide further evidence for the benefits of polyglot learning, in which representations are shared across multiple languages.

Pretrained deep contextual representations have advanced the state-of-the-art on various commonsense NLP tasks, but we lack a concrete understanding of the capability of these models. Thus, we investigate and challenge several aspects of BERT’s commonsense representation abilities. First, we probe BERT’s ability to classify various object attributes, demonstrating that BERT shows a strong ability in encoding various commonsense features in its embedding space, but is still deficient in many areas. Next, we show that, by augmenting BERT’s pretraining data with additional data related to the deficient attributes, we are able to improve performance on a downstream commonsense reasoning task while using a minimal amount of data. Finally, we develop a method of fine-tuning knowledge graphs embeddings alongside BERT and show the continued importance of explicit knowledge graphs.

Adversarial training (AT) is a powerful regularization method for neural networks, aiming to achieve robustness to input perturbations. Yet, the specific effects of the robustness obtained from AT are still unclear in the context of natural language processing. In this paper, we propose and analyze a neural POS tagging model that exploits AT. In our experiments on the Penn Treebank WSJ corpus and the Universal Dependencies (UD) dataset (27 languages), we find that AT not only improves the overall tagging accuracy, but also 1) prevents over-fitting well in low resource languages and 2) boosts tagging accuracy for rare / unseen words. We also demonstrate that 3) the improved tagging performance by AT contributes to the downstream task of dependency parsing, and that 4) AT helps the model to learn cleaner word representations. 5) The proposed AT model is generally effective in different sequence labeling tasks. These positive results motivate further use of AT for natural language tasks.

We present a graph-based Tree Adjoining Grammar (TAG) parser that uses BiLSTMs, highway connections, and character-level CNNs. Our best end-to-end parser, which jointly performs supertagging, POS tagging, and parsing, outperforms the previously reported best results by more than 2.2 LAS and UAS points. The graph-based parsing architecture allows for global inference and rich feature representations for TAG parsing, alleviating the fundamental trade-off between transition-based and graph-based parsing systems. We also demonstrate that the proposed parser achieves state-of-the-art performance in the downstream tasks of Parsing Evaluation using Textual Entailments (PETE) and Unbounded Dependency Recovery. This provides further support for the claim that TAG is a viable formalism for problems that require rich structural analysis of sentences.

We present supertagging-based models for Tree Adjoining Grammar parsing that use neural network architectures and dense vector representation of supertags (elementary trees) to achieve state-of-the-art performance in unlabeled and labeled attachment scores. The shift-reduce parsing model eschews lexical information entirely, and uses only the 1-best supertags to parse a sentence, providing further support for the claim that supertagging is “almost parsing.” We demonstrate that the embedding vector representations the parser induces for supertags possess linguistically interpretable structure, supporting analogies between grammatical structures like those familiar from recent work in distributional semantics. This dense representation of supertags overcomes the drawbacks for statistical models of TAG as compared to CCG parsing, raising the possibility that TAG is a viable alternative for NLP tasks that require the assignment of richer structural descriptions to sentences.