Large language model agents have exhibited exceptional performance across a range of complex interactive tasks. Recent approaches have utilized tuning with expert trajectories to enhance agent performance, yet they primarily concentrate on outcome rewards, which may lead to errors or suboptimal actions due to the absence of process supervision signals. In this paper, we introduce the **I**terative step-level **P**rocess **R**efinement **(IPR)** framework, which provides detailed step-by-step guidance to enhance agent training. Specifically, we adopt the Monte Carlo method to estimate step-level rewards. During each iteration, the agent explores along the expert trajectory and generates new actions. These actions are then evaluated against the corresponding step of expert trajectory using step-level rewards. Such comparison helps identify discrepancies, yielding contrastive action pairs that serve as training data for the agent. Our experiments on three complex agent tasks demonstrate that our framework outperforms a variety of strong baselines. Moreover, our analytical finds highlight the effectiveness of IPR in augmenting action efficiency and its applicability to diverse models.
Modern large language models (LLMs) have exhibited cooperative synergy on complex task-solving, and collective decision-making (CDM) is a pivotal component in LLM-based multi-agent collaboration frameworks. Our survey on 52 recent such systems uncovers a severe lack of diversity, with a heavy reliance on dictatorial and plurality voting for CDM. Through the lens of social choice theory, we scrutinize widely-adopted CDM methods and identify their limitations. To enrich current landscape of LLM-based CDM, we present GEDI, an electoral CDM module that incorporates various ordinal preferential voting mechanisms. Our empirical case study across three benchmarks shows that the integration of certain CDM methods can markedly improve the reasoning capabilities and robustness of some leading LLMs, all without requiring intricate system designs. Additionally, we find that some CDM mechanisms generate positive synergies even with as few as three agents. The voting-based methods also demonstrate robustness against single points of failure, as well as diversity in terms of hit-rate@k and subject-wise impacts.
Despite large language models’ (LLMs’) recent advancements, their bias and hallucination issues persist, and their ability to offer consistent and preferential rankings remains underexplored. This study investigates the capacity of LLMs to provide consistent ordinal preferences, a crucial aspect in scenarios lacking absolute answers. We introduce a formalization of consistency based on order theory, outlining criteria such as transitivity, asymmetry, reversibility, and independence from irrelevant alternatives. Our diagnostic experiments on selected state-of-the-art LLMs reveal their inability to meet these criteria, indicating a strong positional bias and poor transitivity, with preferences easily swayed by irrelevant alternatives. These findings highlight a significant inconsistency in LLM-generated preferential rankings, underscoring the need for further research to address these limitations.
A Large Language Model (LLM) tends to generate inconsistent and sometimes contradictory outputs when presented with a prompt that has equivalent semantics but is expressed differently from the original prompt. To achieve semantic consistency of an LLM, one of the key approaches is to finetune the model with prompt-output pairs with semantically equivalent meanings. Despite its effectiveness, a data-driven finetuning method incurs substantial computation costs in data preparation and model optimization. In this regime, an LLM is treated as a “black box”, restricting our ability to gain deeper insights into its internal mechanism. In this paper, we are motivated to enhance the semantic consistency of LLMs through a more interpretable method (i.e., model editing) to this end. We first identify the model components (i.e., attention heads) that have a key impact on the semantic consistency of an LLM. We subsequently inject biases into the output of these model components along the semantic-consistency activation direction. It is noteworthy that these modifications are cost-effective, without reliance on mass manipulations of the original model parameters. Through comprehensive experiments on the constructed NLU and open-source NLG datasets, our method demonstrates significant improvements in the semantic consistency and task performance of LLMs. Additionally, our method exhibits promising generalization capabilities by performing well on tasks beyond the primary tasks.
Fine-tuning on agent-environment interaction trajectory data holds significant promise for surfacing generalized agent capabilities in open-source large language models (LLMs). In this work, we introduce AgentBank, by far the largest trajectory tuning data collection featuring more than 50k diverse high-quality interaction trajectories which comprises 16 tasks covering five distinct agent skill dimensions. Leveraging a novel annotation pipeline, we are able to scale the annotated trajectories and generate a trajectory dataset with minimized difficulty bias. Furthermore, we fine-tune LLMs on AgentBank to get a series of agent models, Samoyed. Our comparative experiments demonstrate the effectiveness of scaling the interaction trajectory data to acquire generalized agent capabilities. Additional studies also reveal some key observations regarding trajectory tuning and agent skill generalization.
The factual knowledge of LLMs is typically evaluated using accuracy, yet this metric does not capture the vulnerability of LLMs to hallucination-inducing factors like prompt and context variability. How do we evaluate the capabilities of LLMs to consistently produce factually correct answers? In this paper, we propose MOdel kNowledge relIabiliTy scORe (MONITOR), a novel metric designed to directly measure LLMs’ factual reliability. MONITOR is designed to compute the distance between the probability distributions of a valid output and its counterparts produced by the same LLM probing the same fact using different styles of prompts and contexts. Experiments on a comprehensive range of 12 LLMs demonstrate the effectiveness of MONITOR in evaluating the factual reliability of LLMs while maintaining a low computational overhead. In addition, we release the FKTC (Factual Knowledge Test Corpus) to foster research along this line https://github.com/Vicky-Wil/MONITOR.
Contextual information, including the sentences in the same document and in other documents of the dataset, plays a crucial role in improving the accuracy of document-level ASR Error Correction (AEC), while most previous works ignore this. In this paper, we propose a context-aware method that utilizes a k-Nearest Neighbors (kNN) approach to enhance the AEC model by retrieving a datastore containing contextual information. We conduct experiments on two English and two Chinese datasets, and the results demonstrate that our proposed model can effectively utilize contextual information to improve document-level AEC. Furthermore, the context information from the whole dataset provides even better results.
Nowadays, the spread of misinformation is a prominent problem in society. Our research focuses on aiding the automatic identification of misinformation by analyzing the persuasive strategies employed in textual documents. We introduce a novel annotation scheme encompassing common persuasive writing tactics to achieve our objective. Additionally, we provide a dataset on health misinformation, thoroughly annotated by experts utilizing our proposed scheme. Our contribution includes proposing a new task of annotating pieces of text with their persuasive writing strategy types. We evaluate fine-tuning and prompt-engineering techniques with pre-trained language models of the BERT family and the generative large language models of the GPT family using persuasive strategies as an additional source of information. We evaluate the effects of employing persuasive strategies as intermediate labels in the context of misinformation detection. Our results show that those strategies enhance accuracy and improve the explainability of misinformation detection models. The persuasive strategies can serve as valuable insights and explanations, enabling other models or even humans to make more informed decisions regarding the trustworthiness of the information.
Most existing intent discovery methods leverage representation learning and clustering to transfer the prior knowledge of known intents to unknown ones. The learned representations are limited to the syntactic forms of sentences, therefore, fall short of recognizing adequate variations under the same meaning of unknown intents. This paper proposes an approach utilizing frame knowledge as conceptual semantic guidance to bridge the gap between known intents representation learning and unknown intents clustering. Specifically, we employ semantic regularization to minimize the bidirectional KL divergence between model predictions for frame-based and sentence-based samples. Moreover, we construct a frame-guided data augmenter to capture intent-friendly semantic information and implement contrastive clustering learning for unsupervised sentence embedding. Extensive experiments on two benchmark datasets show that our method achieves substantial improvements in accuracy (5%+) compared to solid baselines.
With the wide use of automatic speech recognition(ASR) systems, researchers pay more attention to the ASR error correction task to improve the quality of recognition results. In particular, ASR in bilingual or multilingual settings, namely code-switching ASR, has greater challenges and research value. In this paper, we first present code-switching ASR correction datasets obtained from solid ASR systems and automatic annotators. The datasets contain Chinese-English code-switching dialogues of bilingual speakers in Singapore, Malaysia, and Hong Kong. Based on this task, we propose a controllable iterative (CI) data augmentation method for improving the performance of mainstream ASR error correction systems. With a small amount of training data, our proposed method has the ability to iteratively produce abundant pseudo parallel data from the monolingual corpus for Chinese-English code-switching ASR correction. Results of experiments show that our method achieves the best performance compared with the rule-based, back-translation-based data augmentation methods and large language model ChatGPT.
Neural Machine Translation (NMT) systems encounter a significant challenge when translating a pro-drop (‘pronoun-dropping’) language (e.g., Chinese) to a non-pro-drop one (e.g., English), since the pro-drop phenomenon demands NMT systems to recover omitted pronouns. This unique and crucial task, however, lacks sufficient datasets for benchmarking. To bridge this gap, we introduce PROSE, a new benchmark featured in diverse pro-drop instances for document-level Chinese-English spoken language translation. Furthermore, we conduct an in-depth investigation of the pro-drop phenomenon in spoken Chinese on this dataset, reconfirming that pro-drop reduces the performance of NMT systems in Chinese-English translation. To alleviate the negative impact introduced by pro-drop, we propose Mention-Aware Semantic Augmentation, a novel approach that leverages the semantic embedding of dropped pronouns to augment training pairs. Results from the experiments on four Chinese-English translation corpora show that our proposed method outperforms existing methods regarding omitted pronoun retrieval and overall translation quality.
Topic segmentation aims to detect topic boundaries and split automatic speech recognition transcriptions (e.g., meeting transcripts) into segments that are bounded by thematic meanings. In this work, we propose M3Seg, a novel Maximum-Minimum Mutual information paradigm for linear topic segmentation without using any parallel data. Specifically, by employing sentence representations provided by pre-trained language models, M3Seg first learns a region-based segment encoder based on the maximization of mutual information between the global segment representation and the local contextual sentence representation. Secondly, an edge-based boundary detection module aims to segment the whole by topics based on minimizing the mutual information between different segments. Experiment results on two public datasets demonstrate the effectiveness of M3Seg, which outperform the state-of-the-art methods by a significant (18%–37% improvement) margin.
Dialogue agents have been receiving increasing attention for years, and this trend has been further boosted by the recent progress of large language models (LLMs). Stance detection and dialogue summarization are two core tasks of dialogue agents in application scenarios that involve argumentative dialogues. However, research on these tasks is limited by the insufficiency of public datasets, especially for non-English languages. To address this language resource gap in Chinese, we present ORCHID (Oral Chinese Debate), the first Chinese dataset for benchmarking target-independent stance detection and debate summarization. Our dataset consists of 1,218 real-world debates that were conducted in Chinese on 476 unique topics, containing 2,436 stance-specific summaries and 14,133 fully annotated utterances. Besides providing a versatile testbed for future research, we also conduct an empirical study on the dataset and propose an integrated task. The results show the challenging nature of the dataset and suggest a potential of incorporating stance detection in summarization for argumentative dialogue.
The task of response selection in multi-turn dialogue is to find the best option from all candidates. In order to improve the reasoning ability of the model, previous studies pay more attention to using explicit algorithms to model the dependencies between utterances, which are deterministic, limited and inflexible. In addition, few studies consider differences between the options before and after reasoning. In this paper, we propose an Implicit Relational Reasoning Graph Network to address these issues, which consists of the Utterance Relational Reasoner (URR) and the Option Dual Comparator (ODC). URR aims to implicitly extract dependencies between utterances, as well as utterances and options, and make reasoning with relational graph convolutional networks. ODC focuses on perceiving the difference between the options through dual comparison, which can eliminate the interference of the noise options. Experimental results on two multi-turn dialogue reasoning benchmark datasets MuTual and MuTualplus show that our method significantly improves the baseline of four pre-trained language models and achieves state-of-the-art performance. The model surpasses human performance for the first time on the MuTual dataset.
This paper describes Huawei Artificial Intelligence Application Research Center’s neural machine translation system (“BabelTar”). Our submission to the WMT22 biomedical translation shared task covers language directions between English and the other seven languages (French, German, Italian, Spanish, Portuguese, Russian, and Chinese). During the past four years, our participation in this domain-specific track has witnessed a paradigm shift of methodology from a purely data-driven focus to embracing diversified techniques, including pre-trained multilingual NMT models, homograph disambiguation, ensemble learning, and preprocessing methods. We illustrate practical insights and measured performance improvements relating to how we further improve our domain-specific NMT system.
Motivations, emotions, and actions are inter-related essential factors in human activities. While motivations and emotions have long been considered at the core of exploring how people take actions in human activities, there has been relatively little research supporting analyzing the relationship between human mental states and actions. We present the first study that investigates the viability of modeling motivations, emotions, and actions in language-based human activities, named COMMA (Cognitive Framework of Human Activities). Guided by COMMA, we define three natural language processing tasks (emotion understanding, motivation understanding and conditioned action generation), and build a challenging dataset Hail through automatically extracting samples from Story Commonsense. Experimental results on NLP applications prove the effectiveness of modeling the relationship. Furthermore, our models inspired by COMMA can better reveal the essential relationship among motivations, emotions and actions than existing methods.
Despite their progress in high-resource language settings, unsupervised bilingual lexicon induction (UBLI) models often fail on corpora with low-resource distant language pairs due to insufficient initialization. In this work, we propose a cross-lingual feature extraction (CFE) method to learn the cross-lingual features from monolingual corpora for low-resource UBLI, enabling representations of words with the same meaning leveraged by the initialization step. By integrating cross-lingual representations with pre-trained word embeddings in a fully unsupervised initialization on UBLI, the proposed method outperforms existing state-of-the-art methods on low-resource language pairs (EN-VI, EN-TH, EN-ZH, EN-JA). The ablation study also proves that the learned cross-lingual features can enhance the representational ability and robustness of the existing embedding model.
Controllable story generation is a challenging task in the field of NLP, which has attracted increasing research interest in recent years. However, most existing works generate a whole story conditioned on the appointed keywords or emotions, ignoring the psychological changes of the protagonist. Inspired by psychology theories, we introduce global psychological state chains, which include the needs and emotions of the protagonists, to help a story generation system create more controllable and well-planned stories. In this paper, we propose a Psychology-guided Controllable Story Generation System (PICS) to generate stories that adhere to the given leading context and desired psychological state chains for the protagonist. Specifically, psychological state trackers are employed to memorize the protagonist’s local psychological states to capture their inner temporal relationships. In addition, psychological state planners are adopted to gain the protagonist’s global psychological states for story planning. Eventually, a psychology controller is designed to integrate the local and global psychological states into the story context representation for composing psychology-guided stories. Automatic and manual evaluations demonstrate that PICS outperforms baselines, and each part of PICS shows effectiveness for writing stories with more consistent psychological changes.
This paper introduces our systems for all three subtasks of SemEval-2021 Task 4: Reading Comprehension of Abstract Meaning. To help our model better represent and understand abstract concepts in natural language, we well-design many simple and effective approaches adapted to the backbone model (RoBERTa). Specifically, we formalize the subtasks into the multiple-choice question answering format and add special tokens to abstract concepts, then, the final prediction of QA is considered as the result of subtasks. Additionally, we employ many finetuning tricks to improve the performance. Experimental results show that our approach gains significant performance compared with the baseline systems. Our system achieves eighth rank (87.51%) and tenth rank (89.64%) on the official blind test set of subtask 1 and subtask 2 respectively.
Most language understanding models in task-oriented dialog systems are trained on a small amount of annotated training data, and evaluated in a small set from the same distribution. However, these models can lead to system failure or undesirable output when being exposed to natural language perturbation or variation in practice. In this paper, we conduct comprehensive evaluation and analysis with respect to the robustness of natural language understanding models, and introduce three important aspects related to language understanding in real-world dialog systems, namely, language variety, speech characteristics, and noise perturbation. We propose a model-agnostic toolkit LAUG to approximate natural language perturbations for testing the robustness issues in task-oriented dialog. Four data augmentation approaches covering the three aspects are assembled in LAUG, which reveals critical robustness issues in state-of-the-art models. The augmented dataset through LAUG can be used to facilitate future research on the robustness testing of language understanding in task-oriented dialog.
Further pre-training language models on in-domain data (domain-adaptive pre-training, DAPT) or task-relevant data (task-adaptive pre-training, TAPT) before fine-tuning has been shown to improve downstream tasks’ performances. However, in task-oriented dialog modeling, we observe that further pre-training MLM does not always boost the performance on a downstream task. We find that DAPT is beneficial in the low-resource setting, but as the fine-tuning data size grows, DAPT becomes less beneficial or even useless, and scaling the size of DAPT data does not help. Through Representational Similarity Analysis, we conclude that more data for fine-tuning yields greater change of the model’s representations and thus reduces the influence of initialization.
This paper describes Huawei Artificial Intelligence Application Research Center’s neural machine translation systems and submissions to the WMT21 biomedical translation shared task. Four of the submissions achieve state-of-the-art BLEU scores based on the official-released automatic evaluation results (EN->FR, EN<->IT and ZH->EN). We perform experiments to unveil the practical insights of the involved domain adaptation techniques, including finetuning order, terminology dictionaries, and ensemble decoding. Issues associated with overfitting and under-translation are also discussed.
Neural Machine Translation (NMT) has shown a strong ability to utilize local context to disambiguate the meaning of words. However, it remains a challenge for NMT to leverage broader context information like topics. In this paper, we propose heterogeneous ways of embedding topic information at the sentence level into an NMT model to improve translation performance. Specifically, the topic information can be incorporated as pre-encoder topic embedding, post-encoder topic embedding, and decoder topic embedding to increase the likelihood of selecting target words from the same topic of the source sentence. Experimental results show that NMT models with the proposed topic knowledge embedding outperform the baselines on the English -> German and English -> French translation tasks.
We propose a novel Bi-directional Cognitive Knowledge Framework (BCKF) for reading comprehension from the perspective of complementary learning systems theory. It aims to simulate two ways of thinking in the brain to answer questions, including reverse thinking and inertial thinking. To validate the effectiveness of our framework, we design a corresponding Bi-directional Cognitive Thinking Network (BCTN) to encode the passage and generate a question (answer) given an answer (question) and decouple the bi-directional knowledge. The model has the ability to reverse reasoning questions which can assist inertial thinking to generate more accurate answers. Competitive improvement is observed in DuReader dataset, confirming our hypothesis that bi-directional knowledge helps the QA task. The novel framework shows an interesting perspective on machine reading comprehension and cognitive science.
Creating high-quality annotated dialogue corpora is challenging. It is essential to develop practical annotation tools to support humans in this time-consuming and error-prone task. We present metaCAT, which is an open-source web-based annotation tool designed specifically for developing task-oriented dialogue data. To the best of our knowledge, metaCAT is the first annotation tool that provides comprehensive metadata annotation coverage to the domain, intent, and span information. The data annotation quality is enhanced by a real-time annotation constraint-checking mechanism. An Automatic Speech Recognition (ASR) function is implemented to allow users to paraphrase and create more diversified annotated utterances. metaCAT is publicly available for the community.
This paper describes Huawei’s submissions to the WMT20 biomedical translation shared task. Apart from experimenting with finetuning on domain-specific bitexts, we explore effects of in-domain dictionaries on enhancing cross-domain neural machine translation performance. We utilize a transfer learning strategy through pre-trained machine translation models and extensive scope of engineering endeavors. Four of our ten submissions achieve state-of-the-art performance according to the official automatic evaluation results, namely translation directions on English<->French, English->German and English->Italian.
This paper illustrates Huawei’s submission to the WMT20 low-resource parallel corpus filtering shared task. Our approach focuses on developing a proxy task learner on top of a transformer-based multilingual pre-trained language model to boost the filtering capability for noisy parallel corpora. Such a supervised task also helps us to iterate much more quickly than using an existing neural machine translation system to perform the same task. After performing empirical analyses of the finetuning task, we benchmark our approach by comparing the results with past years’ state-of-theart records. This paper wraps up with a discussion of limitations and future work. The scripts for this study will be made publicly available.
This paper introduces our systems for the first two subtasks of SemEval Task4: Commonsense Validation and Explanation. To clarify the intention for judgment and inject contrastive information for selection, we propose the input reconstruction strategy with prompt templates. Specifically, we formalize the subtasks into the multiple-choice question answering format and construct the input with the prompt templates, then, the final prediction of question answering is considered as the result of subtasks. Experimental results show that our approaches achieve significant performance compared with the baseline systems. Our approaches secure the third rank on both official test sets of the first two subtasks with an accuracy of 96.4 and an accuracy of 94.3 respectively.
We introduce The Benchmark of Linguistic Minimal Pairs (BLiMP),1 a challenge set for evaluating the linguistic knowledge of language models (LMs) on major grammatical phenomena in English. BLiMP consists of 67 individual datasets, each containing 1,000 minimal pairs—that is, pairs of minimally different sentences that contrast in grammatical acceptability and isolate specific phenomenon in syntax, morphology, or semantics. We generate the data according to linguist-crafted grammar templates, and human aggregate agreement with the labels is 96.4%. We evaluate n-gram, LSTM, and Transformer (GPT-2 and Transformer-XL) LMs by observing whether they assign a higher probability to the acceptable sentence in each minimal pair. We find that state-of-the-art models identify morphological contrasts related to agreement reliably, but they struggle with some subtle semantic and syntactic phenomena, such as negative polarity items and extraction islands.
Though state-of-the-art sentence representation models can perform tasks requiring significant knowledge of grammar, it is an open question how best to evaluate their grammatical knowledge. We explore five experimental methods inspired by prior work evaluating pretrained sentence representation models. We use a single linguistic phenomenon, negative polarity item (NPI) licensing, as a case study for our experiments. NPIs like any are grammatical only if they appear in a licensing environment like negation (Sue doesn’t have any cats vs. *Sue has any cats). This phenomenon is challenging because of the variety of NPI licensing environments that exist. We introduce an artificially generated dataset that manipulates key features of NPI licensing for the experiments. We find that BERT has significant knowledge of these features, but its success varies widely across different experimental methods. We conclude that a variety of methods is necessary to reveal all relevant aspects of a model’s grammatical knowledge in a given domain.
This paper describes Huawei’s neural machine translation systems for the WMT 2019 biomedical translation shared task. We trained and fine-tuned our systems on a combination of out-of-domain and in-domain parallel corpora for six translation directions covering English–Chinese, English–French and English–German language pairs. Our submitted systems achieve the best BLEU scores on English–French and English–German language pairs according to the official evaluation results. In the English–Chinese translation task, our systems are in the second place. The enhanced performance is attributed to more in-domain training and more sophisticated models developed. Development of translation models and transfer learning (or domain adaptation) methods has significantly contributed to the progress of the task.