Search

Multimodal Large Language Models (MLLMs) still struggle with hallucinations despite their impressive capabilities. Recent studies have attempted to mitigate this by applying Direct Preference Optimization (DPO) to multimodal scenarios using preference pairs from text-based responses. However, our analysis of representation distributions reveals that multimodal DPO struggles to align image and text representations and to distinguish between hallucinated and non-hallucinated descriptions. To address these challenges, in this work, we propose a Cross-modal Hierarchical Direct Preference Optimization (CHiP) to address these limitations. We introduce a visual preference optimization module within the DPO framework, enabling MLLMs to learn from both textual and visual preferences simultaneously. Furthermore, we propose a hierarchical textual preference optimization module that allows the model to capture preferences at multiple granular levels, including response, segment, and token levels. We evaluate CHiP through both quantitative and qualitative analyses, with results across multiple benchmarks demonstrating its effectiveness in reducing hallucinations. On the Object HalBench dataset, CHiP outperforms DPO in hallucination reduction, achieving improvements of 52.7% and 55.5% relative points based on the base model Muffin and LLaVA models, respectively. We make all our datasets and code publicly available: https://github.com/LVUGAI/CHiP., Comment: Accepted by ICLR 2025

Instruction-following capabilities in large language models (LLMs) have significantly progressed, enabling more complex user interactions through detailed prompts. However, retrieval systems have not matched these advances, most of them still relies on traditional lexical and semantic matching techniques that fail to fully capture user intent. Recent efforts have introduced instruction-aware retrieval models, but these primarily focus on intrinsic content relevance, which neglects the importance of customized preferences for broader document-level attributes. This study evaluates the instruction-following capabilities of various retrieval models beyond content relevance, including LLM-based dense retrieval and reranking models. We develop InfoSearch, a novel retrieval evaluation benchmark spanning six document-level attributes: Audience, Keyword, Format, Language, Length, and Source, and introduce novel metrics -- Strict Instruction Compliance Ratio (SICR) and Weighted Instruction Sensitivity Evaluation (WISE) to accurately assess the models' responsiveness to instructions. Our findings reveal that while reranking models generally surpass retrieval models in instruction following, they still face challenges in handling certain attributes. Moreover, although instruction fine-tuning and increased model size lead to better performance, most models fall short of achieving comprehensive instruction compliance as assessed by our benchmark.

Large language models (LLMs) have excelled in various NLP tasks, including machine translation (MT), yet most studies focus on sentence-level translation. This work investigates the inherent capability of instruction-tuned LLMs for document-level translation (docMT). Unlike prior approaches that require specialized techniques, we evaluate LLMs by directly prompting them to translate entire documents in a single pass. Our results show that this method improves translation quality compared to translating sentences separately, even without document-level fine-tuning. However, this advantage is not reflected in BLEU scores, which often favor sentence-based translations. We propose using the LLM-as-a-judge paradigm for evaluation, where GPT-4 is used to assess document coherence, accuracy, and fluency in a more nuanced way than n-gram-based metrics. Overall, our work demonstrates that instruction-tuned LLMs can effectively leverage document context for translation. However, we caution against using BLEU scores for evaluating docMT, as they often provide misleading outcomes, failing to capture the quality of document-level translation. Code and data are available at https://github.com/EIT-NLP/BLEUless_DocMT

The Soft Actor-Critic (SAC) algorithm is known for its stability and high sample efficiency in deep reinforcement learning. However, the tanh transformation applied to sampled actions in SAC distorts the action distribution, hindering the selection of the most probable actions. This paper presents a novel action sampling method that directly identifies and selects the most probable actions within the transformed distribution, thereby addressing this issue. Extensive experiments on standard continuous control benchmarks demonstrate that the proposed method significantly enhances SAC's performance, resulting in faster convergence and higher cumulative rewards compared to the original algorithm.

Web agents have emerged as a promising direction to automate Web task completion based on user instructions, significantly enhancing user experience. Recently, Web agents have evolved from traditional agents to Large Language Models (LLMs)-based Web agents. Despite their success, existing LLM-based Web agents overlook the importance of personalized data (e.g., user profiles and historical Web behaviors) in assisting the understanding of users' personalized instructions and executing customized actions. To overcome the limitation, we first formulate the task of LLM-empowered personalized Web agents, which integrate personalized data and user instructions to personalize instruction comprehension and action execution. To address the absence of a comprehensive evaluation benchmark, we construct a Personalized Web Agent Benchmark (PersonalWAB), featuring user instructions, personalized user data, Web functions, and two evaluation paradigms across three personalized Web tasks. Moreover, we propose a Personalized User Memory-enhanced Alignment (PUMA) framework to adapt LLMs to the personalized Web agent task. PUMA utilizes a memory bank with a task-specific retrieval strategy to filter relevant historical Web behaviors. Based on the behaviors, PUMA then aligns LLMs for personalized action execution through fine-tuning and direct preference optimization. Extensive experiments validate the superiority of PUMA over existing Web agents on PersonalWAB., Comment: The code and data are available on the project website https://hongrucai.github.io/PersonalWAB/

In recent years, multimodal large language models (MLLMs) have garnered significant attention from both industry and academia. However, there is still considerable debate on constructing MLLM architectures, particularly regarding the selection of appropriate connectors for perception tasks of varying granularities. This paper systematically investigates the impact of connectors on MLLM performance. Specifically, we classify connectors into feature-preserving and feature-compressing types. Utilizing a unified classification standard, we categorize sub-tasks from three comprehensive benchmarks, MMBench, MME, and SEED-Bench, into three task types: coarse-grained perception, fine-grained perception, and reasoning, and evaluate the performance. Our findings reveal that feature-preserving connectors excel in \emph{fine-grained perception} tasks due to their ability to retain detailed visual information. In contrast, feature-compressing connectors, while less effective in fine-grained perception tasks, offer significant speed advantages and perform comparably in \emph{coarse-grained perception} and \emph{reasoning} tasks. These insights are crucial for guiding MLLM architecture design and advancing the optimization of MLLM architectures., Comment: Accepted to EMNLP 2024 Main Conference

Reinforcement Learning from Human Feedback significantly enhances Natural Language Processing by aligning language models with human expectations. A critical factor in this alignment is the strength of reward models used during training. This study explores whether stronger reward models invariably lead to better language models. In this paper, through experiments on relevance, factuality, and completeness tasks using the QA-FEEDBACK dataset and reward models based on Longformer, we uncover a surprising paradox: language models trained with moderately accurate reward models outperform those guided by highly accurate ones. This challenges the widely held belief that stronger reward models always lead to better language models, and opens up new avenues for future research into the key factors driving model performance and how to choose the most suitable reward models. Code and additional details are available at https://github.com/EIT-NLP/AccuracyParadox-RLHF., Comment: 10 pages, 27 figures (including 18 in the appendix), submitted to EMNLP 2024

Fine-tuning and in-context learning (ICL) are two prevalent methods in imbuing large language models with task-specific knowledge. It is commonly believed that fine-tuning can surpass ICL given sufficient training samples as it allows the model to adjust its internal parameters based on the data. However, this paper presents a counterintuitive finding: For tasks with implicit patterns, ICL captures these patterns significantly better than fine-tuning. We developed several datasets featuring implicit patterns, such as sequences determining answers through parity or identifying reducible terms in calculations. We then evaluated the models' understanding of these patterns under both fine-tuning and ICL across models ranging from 0.5B to 7B parameters. The results indicate that models employing ICL can quickly grasp deep patterns and significantly improve accuracy. In contrast, fine-tuning, despite utilizing thousands of times more training samples than ICL, achieved only limited improvements. We also proposed circuit shift theory from a mechanistic interpretability's view to explain why ICL wins., Comment: EMNLP'24 Findings

In recent advancements, multimodal large language models (MLLMs) have been fine-tuned on specific medical image datasets to address medical visual question answering (Med-VQA) tasks. However, this common approach of task-specific fine-tuning is costly and necessitates separate models for each downstream task, limiting the exploration of zero-shot capabilities. In this paper, we introduce MC-CoT, a modular cross-modal collaboration Chain-of-Thought (CoT) framework designed to enhance the zero-shot performance of MLLMs in Med-VQA by leveraging large language models (LLMs). MC-CoT improves reasoning and information extraction by integrating medical knowledge and task-specific guidance, where LLM provides various complex medical reasoning chains and MLLM provides various observations of medical images based on instructions of the LLM. Our experiments on datasets such as SLAKE, VQA-RAD, and PATH-VQA show that MC-CoT surpasses standalone MLLMs and various multimodality CoT frameworks in recall rate and accuracy. These findings highlight the importance of incorporating background information and detailed guidance in addressing complex zero-shot Med-VQA tasks., Comment: 21 pages, 14 figures, 6 tables

We provide a rigorous convergence proof demonstrating that the well-known semi-analytical Fourier cosine (COS) formula for the inverse Fourier transform of continuous probability distributions can be extended to discrete probability distributions, with the help of spectral filters. We establish general convergence rates for these filters and further show that several classical spectral filters achieve convergence rates one order faster than previously recognized in the literature on the Gibbs phenomenon. Our numerical experiments corroborate the theoretical convergence results. Additionally, we illustrate the computational speed and accuracy of the discrete COS method with applications in computational statistics and quantitative finance. The theoretical and numerical results highlight the method's potential for solving problems involving discrete distributions, particularly when the characteristic function is known, allowing the discrete Fourier transform (DFT) to be bypassed.

Humanoid robots offer significant versatility for performing a wide range of tasks, yet their basic ability to walk and run, especially at high velocities, remains a challenge. This letter presents a novel method that combines deep reinforcement learning with kinodynamic priors to achieve stable locomotion control (KSLC). KSLC promotes coordinated arm movements to counteract destabilizing forces, enhancing overall stability. Compared to the baseline method, KSLC provides more accurate tracking of commanded velocities and better generalization in velocity control. In simulation tests, the KSLC-enabled humanoid robot successfully tracked a target velocity of 3.5 m/s with reduced fluctuations. Sim-to-sim validation in a high-fidelity environment further confirmed its robust performance, highlighting its potential for real-world applications., Comment: This work has been submitted to the IEEE for possible publication

To reduce the need for human annotations, large language models (LLMs) have been proposed as judges of the quality of other candidate models. LLM judges are typically evaluated by measuring the correlation with human judgments on generation tasks such as summarization or machine translation. In contrast, we study LLM judges on mathematical reasoning tasks. These tasks require multi-step reasoning, and the correctness of their solutions is verifiable, enabling a more objective evaluation. We perform a detailed performance analysis and find that the used judges are mostly unable to improve task performance but are able to pick the better model. Our analysis uncovers a strong correlation between judgment performance and the candidate model task performance. We observe that judges tend to choose the model of higher quality even if its answer is incorrect. Further, we show that it is possible to use statistics, such as the task performances of the individual models, to predict judgment performance. In an ablation, we either swap or mask the candidate answers and observe that judges often keep the original judgment, providing evidence that judges incorporate writing style in their judgments. In summary, we find that regularities in the judgments are quantifiable using statistical measures and provide various angles on exploiting them.

Large language models (LLMs) exhibit remarkable in-context learning (ICL) capabilities. However, the underlying working mechanism of ICL remains poorly understood. Recent research presents two conflicting views on ICL: One emphasizes the impact of similar examples in the demonstrations, stressing the need for label correctness and more shots. The other attributes it to LLMs' inherent ability of task recognition, deeming label correctness and shot numbers of demonstrations as not crucial. In this work, we provide a Two-Dimensional Coordinate System that unifies both views into a systematic framework. The framework explains the behavior of ICL through two orthogonal variables: whether similar examples are presented in the demonstrations (perception) and whether LLMs can recognize the task (cognition). We propose the peak inverse rank metric to detect the task recognition ability of LLMs and study LLMs' reactions to different definitions of similarity. Based on these, we conduct extensive experiments to elucidate how ICL functions across each quadrant on multiple representative classification tasks. Finally, we extend our analyses to generation tasks, showing that our coordinate system can also be used to interpret ICL for generation tasks effectively.

The reward model has become increasingly important in alignment, assessment, and data construction for large language models (LLMs). Most existing researchers focus on enhancing reward models through data improvements, following the conventional training framework for reward models that directly optimizes the predicted rewards. In this paper, we propose a hybrid alignment framework HaF-RM for reward model training by introducing an additional constraint on token-level policy probabilities in addition to the reward score. It can simultaneously supervise the internal preference model at the token level and optimize the mapping layer of the reward model at the sequence level. Experiment results on five datasets sufficiently show the validity and effectiveness of our proposed hybrid framework for training a high-quality reward model. By decoupling the reward modeling procedure and incorporating hybrid supervision, our HaF-RM framework offers a principled and effective approach to enhancing the performance and alignment of reward models, a critical component in the responsible development of powerful language models. We release our code at https://haf-rm.github.io.

Retrieval-augmented generation has gained popularity as a framework to enhance large language models with external knowledge. However, its effectiveness hinges on the retrieval robustness of the model. If the model lacks retrieval robustness, its performance is constrained by the accuracy of the retriever, resulting in significant compromises when the retrieved context is irrelevant. In this paper, we evaluate the "implicit" retrieval robustness of various large language models, instructing them to directly output the final answer without explicitly judging the relevance of the retrieved context. Our findings reveal that fine-tuning on a mix of gold and distracting context significantly enhances the model's robustness to retrieval inaccuracies, while still maintaining its ability to extract correct answers when retrieval is accurate. This suggests that large language models can implicitly handle relevant or irrelevant retrieved context by learning solely from the supervision of the final answer in an end-to-end manner. Introducing an additional process for explicit relevance judgment can be unnecessary and disrupts the end-to-end approach.

While large language models (LLMs) have showcased impressive capabilities, they struggle with addressing legal queries due to the intricate complexities and specialized expertise required in the legal field. In this paper, we introduce InternLM-Law, a specialized LLM tailored for addressing diverse legal queries related to Chinese laws, spanning from responding to standard legal questions (e.g., legal exercises in textbooks) to analyzing complex real-world legal situations. We meticulously construct a dataset in the Chinese legal domain, encompassing over 1 million queries, and implement a data filtering and processing pipeline to ensure its diversity and quality. Our training approach involves a novel two-stage process: initially fine-tuning LLMs on both legal-specific and general-purpose content to equip the models with broad knowledge, followed by exclusive fine-tuning on high-quality legal data to enhance structured output generation. InternLM-Law achieves the highest average performance on LawBench, outperforming state-of-the-art models, including GPT-4, on 13 out of 20 subtasks. We make InternLM-Law and our dataset publicly available to facilitate future research in applying LLMs within the legal domain., Comment: Our dataset, code and models will be released at https://github.com/InternLM/InternLM-Law

Traditionally, success in multilingual machine translation can be attributed to three key factors in training data: large volume, diverse translation directions, and high quality. In the current practice of fine-tuning large language models (LLMs) for translation, we revisit the importance of these factors. We find that LLMs display strong translation capability after being fine-tuned on as few as 32 parallel sentences and that fine-tuning on a single translation direction enables translation in multiple directions. However, the choice of direction is critical: fine-tuning LLMs with only English on the target side can lead to task misinterpretation, which hinders translation into non-English languages. Problems also arise when noisy synthetic data is placed on the target side, especially when the target language is well-represented in LLM pre-training. Yet interestingly, synthesized data in an under-represented language has a less pronounced effect. Our findings suggest that when adapting LLMs to translation, the requirement on data quantity can be eased but careful considerations are still crucial to prevent an LLM from exploiting unintended data biases., Comment: EMNLP 2024 Main

Recent research has shown that large language models (LLMs) can achieve remarkable translation performance through supervised fine-tuning (SFT) using only a small amount of parallel data. However, SFT simply instructs the model to imitate the reference translations at the token level, making it vulnerable to the noise present in the references. Hence, the assistance from SFT often reaches a plateau once the LLMs have achieved a certain level of translation capability, and further increasing the size of parallel data does not provide additional benefits. To overcome this plateau associated with imitation-based SFT, we propose a preference-based approach built upon the Plackett-Luce model. The objective is to steer LLMs towards a more nuanced understanding of translation preferences from a holistic view, while also being more resilient in the absence of gold translations. We further build a dataset named MAPLE to verify the effectiveness of our approach, which includes multiple translations of varying quality for each source sentence. Extensive experiments demonstrate the superiority of our approach in "breaking the plateau" across diverse LLMs and test settings. Our in-depth analysis underscores the pivotal role of diverse translations and accurate preference scores in the success of our approach., Comment: Accepted to NAACL 2024 (long, main)

Scaling law principles indicate a power-law correlation between loss and variables such as model size, dataset size, and computational resources utilized during training. These principles play a vital role in optimizing various aspects of model pre-training, ultimately contributing to the success of large language models such as GPT-4, Llama and Gemini. However, the original scaling law paper by OpenAI did not disclose the complete details necessary to derive the precise scaling law formulas, and their conclusions are only based on models containing up to 1.5 billion parameters. Though some subsequent works attempt to unveil these details and scale to larger models, they often neglect the training dependency of important factors such as the learning rate, context length and batch size, leading to their failure to establish a reliable formula for predicting the test loss trajectory. In this technical report, we confirm that the scaling law formulations proposed in the original OpenAI paper remain valid when scaling the model size up to 33 billion, but the constant coefficients in these formulas vary significantly with the experiment setup. We meticulously identify influential factors and provide transparent, step-by-step instructions to estimate all constant terms in scaling-law formulas by training on models with only 1M~60M parameters. Using these estimated formulas, we showcase the capability to accurately predict various attributes for models with up to 33B parameters before their training, including (1) the minimum possible test loss; (2) the minimum required training steps and processed tokens to achieve a specific loss; (3) the critical batch size with an optimal time/computation trade-off at any loss value; and (4) the complete test loss trajectory with arbitrary batch size.

In-context learning is a popular inference strategy where large language models solve a task using only a few labeled demonstrations without needing any parameter updates. Although there have been extensive studies on English in-context learning, multilingual in-context learning remains under-explored, and we lack an in-depth understanding of the role of demonstrations in this context. To address this gap, we conduct a multidimensional analysis of multilingual in-context learning, experimenting with 5 models from different model families, 9 datasets covering classification and generation tasks, and 56 typologically diverse languages. Our results reveal that the effectiveness of demonstrations varies significantly across models, tasks, and languages. We also find that strong instruction-following models including Llama 2-Chat, GPT-3.5, and GPT-4 are largely insensitive to the quality of demonstrations. Instead, a carefully crafted template often eliminates the benefits of demonstrations for some tasks and languages altogether. These findings show that the importance of demonstrations might be overestimated. Our work highlights the need for granular evaluation across multiple axes towards a better understanding of in-context learning., Comment: ACL 2024 findings

The decoder-only Transformer architecture with causal masking and relative position encoding (RPE) has become the de facto choice in language modeling. Despite its exceptional performance across various tasks, we have identified two limitations: First, it requires all attention scores to be non-zero and sum up to 1, even if the current embedding has sufficient self-contained information. This compels the model to assign disproportional excessive attention to specific tokens. Second, RPE-based Transformers are not universal approximators due to their limited capacity at encoding absolute positional information, which limits their application in position-critical tasks. In this work, we propose StableMask: a parameter-free method to address both limitations by refining the causal mask. It introduces pseudo-attention values to balance attention distributions and encodes absolute positional information via a progressively decreasing mask ratio. StableMask's effectiveness is validated both theoretically and empirically, showing significant enhancements in language models with parameter sizes ranging from 71M to 1.4B across diverse datasets and encoding methods. We further show that it naturally supports (1) efficient extrapolation without special tricks such as StreamingLLM and (2) easy integration with existing attention optimization techniques., Comment: Preprint

Pre-trained models (PTMs) have achieved great success in various Software Engineering (SE) downstream tasks following the ``pre-train then fine-tune'' paradigm. As fully fine-tuning all parameters of PTMs can be computationally expensive, a widely used solution is parameter-efficient fine-tuning (PEFT), which freezes PTMs while introducing extra parameters. Though work has been done to test PEFT methods in the SE field, a comprehensive evaluation is still lacking. This paper aims to fill in this gap by evaluating the effectiveness of five PEFT methods on eight PTMs and four SE downstream tasks. For different tasks and PEFT methods, we seek answers to the following research questions: 1) Is it more effective to use PTMs trained specifically on source code, or is it sufficient to use PTMs trained on natural language text? 2) What is the impact of varying model sizes? 3) How does the model architecture affect the performance? Besides effectiveness, we also discuss the efficiency of PEFT methods, concerning the costs of required training time and GPU resource consumption. We hope that our findings can provide a deeper understanding of PEFT methods on various PTMs and SE downstream tasks. All the codes and data are available at \url{https://github.com/zwtnju/PEFT.git}.

This paper introduces a novel approach for computing netting--set level and counterparty level exposures, such as Potential Future Exposure (PFE) and Expected Exposure (EE), along with associated sensitivities. The method is essentially an extension of the Fourier-cosine series expansion (COS) method, originally proposed for option pricing. This method can accommodate a broad range of models where the joint distribution of involved risk factors is analytically or semi-analytically tractable. This inclusivity encompasses nearly all CCR models commonly employed in practice. A notable advantage of the COS method is its sustained efficiency, particularly when handling large portfolios. A theoretical error analysis is also provided to justify the method's theoretical stability and accuracy. Various numerical tests are conducted using real-sized portfolios, and the results underscore its potential as a significantly more efficient alternative to the Monte Carlo method for practical usage, particularly applicable to portfolios involving a relatively modest number of risk factors. Furthermore, the observed error convergence rates align closely with the theoretical error analysis., Comment: 26 Pages

Large language models (LLMs) have demonstrated strong capabilities in various aspects. However, when applying them to the highly specialized, safe-critical legal domain, it is unclear how much legal knowledge they possess and whether they can reliably perform legal-related tasks. To address this gap, we propose a comprehensive evaluation benchmark LawBench. LawBench has been meticulously crafted to have precise assessment of the LLMs' legal capabilities from three cognitive levels: (1) Legal knowledge memorization: whether LLMs can memorize needed legal concepts, articles and facts; (2) Legal knowledge understanding: whether LLMs can comprehend entities, events and relationships within legal text; (3) Legal knowledge applying: whether LLMs can properly utilize their legal knowledge and make necessary reasoning steps to solve realistic legal tasks. LawBench contains 20 diverse tasks covering 5 task types: single-label classification (SLC), multi-label classification (MLC), regression, extraction and generation. We perform extensive evaluations of 51 LLMs on LawBench, including 20 multilingual LLMs, 22 Chinese-oriented LLMs and 9 legal specific LLMs. The results show that GPT-4 remains the best-performing LLM in the legal domain, surpassing the others by a significant margin. While fine-tuning LLMs on legal specific text brings certain improvements, we are still a long way from obtaining usable and reliable LLMs in legal tasks. All data, model predictions and evaluation code are released in https://github.com/open-compass/LawBench/. We hope this benchmark provides in-depth understanding of the LLMs' domain-specified capabilities and speed up the development of LLMs in the legal domain.

Despite the progress we have recorded in the last few years in multilingual natural language processing, evaluation is typically limited to a small set of languages with available datasets which excludes a large number of low-resource languages. In this paper, we created SIB-200 -- a large-scale open-sourced benchmark dataset for topic classification in 200 languages and dialects to address the lack of evaluation dataset for Natural Language Understanding (NLU). For many of the languages covered in SIB-200, this is the first publicly available evaluation dataset for NLU. The dataset is based on Flores-200 machine translation corpus. We annotated the English portion of the dataset and extended the sentence-level annotation to the remaining 203 languages covered in the corpus. Despite the simplicity of this task, our evaluation in full-supervised setting, cross-lingual transfer setting and prompting of large language model setting show that there is still a large gap between the performance of high-resource and low-resource languages when multilingual evaluation is scaled to numerous world languages. We found that languages unseen during the pre-training of multilingual language models, under-represented language families (like Nilotic and Altantic-Congo), and languages from the regions of Africa, Americas, Oceania and South East Asia, often have the lowest performance on our topic classification dataset. We hope our dataset will encourage a more inclusive evaluation of multilingual language models on a more diverse set of languages. https://github.com/dadelani/sib-200, Comment: Accepted to EACL 2024 (main conference)

Weakly supervised learning is a popular approach for training machine learning models in low-resource settings. Instead of requesting high-quality yet costly human annotations, it allows training models with noisy annotations obtained from various weak sources. Recently, many sophisticated approaches have been proposed for robust training under label noise, reporting impressive results. In this paper, we revisit the setup of these approaches and find that the benefits brought by these approaches are significantly overestimated. Specifically, we find that the success of existing weakly supervised learning approaches heavily relies on the availability of clean validation samples which, as we show, can be leveraged much more efficiently by simply training on them. After using these clean labels in training, the advantages of using these sophisticated approaches are mostly wiped out. This remains true even when reducing the size of the available clean data to just five samples per class, making these approaches impractical. To understand the true value of weakly supervised learning, we thoroughly analyze diverse NLP datasets and tasks to ascertain when and why weakly supervised approaches work. Based on our findings, we provide recommendations for future research., Comment: ACL 2023, oral presentation

Cross-lingual transfer is important for developing high-quality chatbots in multiple languages due to the strongly imbalanced distribution of language resources. A typical approach is to leverage off-the-shelf machine translation (MT) systems to utilize either the training corpus or developed models from high-resource languages. In this work, we investigate whether it is helpful to utilize MT at all in this task. To do so, we simulate a low-resource scenario assuming access to limited Chinese dialog data in the movie domain and large amounts of English dialog data from multiple domains. Experiments show that leveraging English dialog corpora can indeed improve the naturalness, relevance and cross-domain transferability in Chinese. However, directly using English dialog corpora in its original form, surprisingly, is better than using its translated version. As the topics and wording habits in daily conversations are strongly culture-dependent, MT can reinforce the bias from high-resource languages, yielding unnatural generations in the target language. Considering the cost of translating large amounts of text and the strong effects of the translation quality, we suggest future research should rather focus on utilizing the original English data for cross-lingual transfer in dialog generation. We perform extensive human evaluations and ablation studies. The analysis results, together with the collected dataset, are presented to draw attention towards this area and benefit future research.

Product Question Answering (PQA) systems are key in e-commerce applications to provide responses to customers' questions as they shop for products. While existing work on PQA focuses mainly on English, in practice there is need to support multiple customer languages while leveraging product information available in English. To study this practical industrial task, we present xPQA, a large-scale annotated cross-lingual PQA dataset in 12 languages across 9 branches, and report results in (1) candidate ranking, to select the best English candidate containing the information to answer a non-English question; and (2) answer generation, to generate a natural-sounding non-English answer based on the selected English candidate. We evaluate various approaches involving machine translation at runtime or offline, leveraging multilingual pre-trained LMs, and including or excluding xPQA training data. We find that (1) In-domain data is essential as cross-lingual rankers trained on other domains perform poorly on the PQA task; (2) Candidate ranking often prefers runtime-translation approaches while answer generation prefers multilingual approaches; (3) Translating offline to augment multilingual models helps candidate ranking mainly on languages with non-Latin scripts; and helps answer generation mainly on languages with Latin scripts. Still, there remains a significant performance gap between the English and the cross-lingual test sets., Comment: ACL 2023 industry track. Dataset available in https://github.com/amazon-science/contextual-product-qa

Identifying the fault in propellers is important to keep quadrotors operating safely and efficiently. The simulation-to-reality (sim-to-real) UAV fault diagnosis methods provide a cost-effective and safe approach to detecting propeller faults. However, due to the gap between simulation and reality, classifiers trained with simulated data usually underperform in real flights. In this work, a novel difference-based deep convolutional neural network (DDCNN) model is presented to address the above issue. It uses the difference features extracted by deep convolutional neural networks to reduce the sim-to-real gap. Moreover, a new domain adaptation (DA) method is presented to further bring the distribution of the real-flight data closer to that of the simulation data. The experimental results demonstrate that the DDCNN+DA model can increase the accuracy from 52.9% to 99.1% in real-world UAV fault detection., Comment: 20xx IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, including reprinting/republishing this material for advertising or promotional purposes, collecting new collected works for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

Large Language Models pre-trained with self-supervised learning have demonstrated impressive zero-shot generalization capabilities on a wide spectrum of tasks. In this work, we present WeLM: a well-read pre-trained language model for Chinese that is able to seamlessly perform different types of tasks with zero or few-shot demonstrations. WeLM is trained with 10B parameters by "reading" a curated high-quality corpus covering a wide range of topics. We show that WeLM is equipped with broad knowledge on various domains and languages. On 18 monolingual (Chinese) tasks, WeLM can significantly outperform existing pre-trained models with similar sizes and match the performance of models up to 25 times larger. WeLM also exhibits strong capabilities in multi-lingual and code-switching understanding, outperforming existing multilingual language models pre-trained on 30 languages. Furthermore, We collected human-written prompts for a large set of supervised datasets in Chinese and fine-tuned WeLM with multi-prompted training. The resulting model can attain strong generalization on unseen types of tasks and outperform the unsupervised WeLM in zero-shot learning. Finally, we demonstrate that WeLM has basic skills at explaining and calibrating the decisions from itself, which can be promising directions for future research. Our models can be applied from https://welm.weixin.qq.com/docs/api/.

Owing to the lack of corpora for low-resource languages, current works on dialogue generation have mainly focused on English. In this paper, we present mDIA, the first large-scale multilingual benchmark for dialogue generation across low- to high-resource languages. It covers real-life conversations in 46 languages across 19 language families. We present baseline results obtained by fine-tuning the multilingual, non-dialogue-focused pre-trained model mT5 as well as English-centric, dialogue-focused pre-trained chatbot DialoGPT. The results show that mT5-based models perform better on sacreBLEU and BertScore but worse on diversity. Even though promising results are found in few-shot and zero-shot scenarios, there is a large gap between the generation quality in English and other languages. We hope that the release of mDIA could encourage more works on multilingual dialogue generation to promote language diversity., Comment: The dataset and processing scripts are available in https://github.com/DoctorDream/mDIA

Dense retrieval (DR) approaches based on powerful pre-trained language models (PLMs) achieved significant advances and have become a key component for modern open-domain question-answering systems. However, they require large amounts of manual annotations to perform competitively, which is infeasible to scale. To address this, a growing body of research works have recently focused on improving DR performance under low-resource scenarios. These works differ in what resources they require for training and employ a diverse set of techniques. Understanding such differences is crucial for choosing the right technique under a specific low-resource scenario. To facilitate this understanding, we provide a thorough structured overview of mainstream techniques for low-resource DR. Based on their required resources, we divide the techniques into three main categories: (1) only documents are needed; (2) documents and questions are needed; and (3) documents and question-answer pairs are needed. For every technique, we introduce its general-form algorithm, highlight the open issues and pros and cons. Promising directions are outlined for future research.

Training deep neural networks (DNNs) under weak supervision has attracted increasing research attention as it can significantly reduce the annotation cost. However, labels from weak supervision can be noisy, and the high capacity of DNNs enables them to easily overfit the label noise, resulting in poor generalization. Recent methods leverage self-training to build noise-resistant models, in which a teacher trained under weak supervision is used to provide highly confident labels for teaching the students. Nevertheless, the teacher derived from such frameworks may have fitted a substantial amount of noise and therefore produce incorrect pseudo-labels with high confidence, leading to severe error propagation. In this work, we propose Meta Self-Refinement (MSR), a noise-resistant learning framework, to effectively combat label noise from weak supervision. Instead of relying on a fixed teacher trained with noisy labels, we encourage the teacher to refine its pseudo-labels. At each training step, MSR performs a meta gradient descent on the current mini-batch to maximize the student performance on a clean validation set. Extensive experimentation on eight NLP benchmarks demonstrates that MSR is robust against label noise in all settings and outperforms state-of-the-art methods by up to 11.4% in accuracy and 9.26% in F1 score., Comment: EACL 2023 (long paper)

Recent advances in the pre-training of language models leverage large-scale datasets to create multilingual models. However, low-resource languages are mostly left out in these datasets. This is primarily because many widely spoken languages are not well represented on the web and therefore excluded from the large-scale crawls used to create datasets. Furthermore, downstream users of these models are restricted to the selection of languages originally chosen for pre-training. This work investigates how to optimally leverage existing pre-trained models to create low-resource translation systems for 16 African languages. We focus on two questions: 1) How can pre-trained models be used for languages not included in the initial pre-training? and 2) How can the resulting translation models effectively transfer to new domains? To answer these questions, we create a new African news corpus covering 16 languages, of which eight languages are not part of any existing evaluation dataset. We demonstrate that the most effective strategy for transferring both to additional languages and to additional domains is to fine-tune large pre-trained models on small quantities of high-quality translation data., Comment: Accepted to NAACL 2022 (added evaluation data for amh, kin, nya, sna, xho)

Legal judgment prediction (LJP) applies Natural Language Processing (NLP) techniques to predict judgment results based on fact descriptions automatically. Recently, large-scale public datasets and advances in NLP research have led to increasing interest in LJP. Despite a clear gap between machine and human performance, impressive results have been achieved in various benchmark datasets. In this paper, to address the current lack of comprehensive survey of existing LJP tasks, datasets, models and evaluations, (1) we analyze 31 LJP datasets in 6 languages, present their construction process and define a classification method of LJP with 3 different attributes; (2) we summarize 14 evaluation metrics under four categories for different outputs of LJP tasks; (3) we review 12 legal-domain pretrained models in 3 languages and highlight 3 major research directions for LJP; (4) we show the state-of-art results for 8 representative datasets from different court cases and discuss the open challenges. This paper can provide up-to-date and comprehensive reviews to help readers understand the status of LJP. We hope to facilitate both NLP researchers and legal professionals for further joint efforts in this problem., Comment: 25 pages, 6 figures and 12 tables

In conversational QA, models have to leverage information in previous turns to answer upcoming questions. Current approaches, such as Question Rewriting, struggle to extract relevant information as the conversation unwinds. We introduce the Common Ground (CG), an approach to accumulate conversational information as it emerges and select the relevant information at every turn. We show that CG offers a more efficient and human-like way to exploit conversational information compared to existing approaches, leading to improvements on Open Domain Conversational QA., Comment: Accepted at NLP for ConvAI

Text generation aims to produce human-like natural language output for down-stream tasks. It covers a wide range of applications like machine translation, document summarization, dialogue generation and so on. Recently deep neural network-based end-to-end architectures have been widely adopted. The end-to-end approach conflates all sub-modules, which used to be designed by complex handcrafted rules, into a holistic encode-decode architecture. Given enough training data, it is able to achieve state-of-the-art performance yet avoiding the need of language/domain-dependent knowledge. Nonetheless, deep learning models are known to be extremely data-hungry, and text generated from them usually suffer from low diversity, interpretability and controllability. As a result, it is difficult to trust the output from them in real-life applications. Deep latent-variable models, by specifying the probabilistic distribution over an intermediate latent process, provide a potential way of addressing these problems while maintaining the expressive power of deep neural networks. This dissertation presents how deep latent-variable models can improve over the standard encoder-decoder model for text generation., Comment: phd thesis

Reasoning is central to human intelligence. However, fallacious arguments are common, and some exacerbate problems such as spreading misinformation about climate change. In this paper, we propose the task of logical fallacy detection, and provide a new dataset (Logic) of logical fallacies generally found in text, together with an additional challenge set for detecting logical fallacies in climate change claims (LogicClimate). Detecting logical fallacies is a hard problem as the model must understand the underlying logical structure of the argument. We find that existing pretrained large language models perform poorly on this task. In contrast, we show that a simple structure-aware classifier outperforms the best language model by 5.46% on Logic and 4.51% on LogicClimate. We encourage future work to explore this task as (a) it can serve as a new reasoning challenge for language models, and (b) it can have potential applications in tackling the spread of misinformation. Our dataset and code are available at https://github.com/causalNLP/logical-fallacy, Comment: EMNLP 2021 Findings

Recent research has achieved impressive progress in the session-based recommendation. However, information such as item knowledge and click time interval, which could be potentially utilized to improve the performance, remains largely unexploited. In this paper, we propose a framework called Knowledge-enhanced Session-based Recommendation with Temporal Transformer (KSTT) to incorporate such information when learning the item and session embeddings. Specifically, a knowledge graph, which models contexts among items within a session and their corresponding attributes, is proposed to obtain item embeddings through graph representation learning. We introduce time interval embedding to represent the time pattern between the item that needs to be predicted and historical click, and use it to replace the position embedding in the original transformer (called temporal transformer). The item embeddings in a session are passed through the temporal transformer network to get the session embedding, based on which the final recommendation is made. Extensive experiments demonstrate that our model outperforms state-of-the-art baselines on four benchmark datasets.