论文检索

Bohao Chu, Meijie Li, Sameh Frihat, Chengyu Gu, Georg Lodde, Elisabeth Livingstone, Norbert Fuhr

While document summarization with LLMs has enhanced access to textual information, concerns about the factual accuracy of these summaries persist (e.g., hallucination), especially in the medical domain. Tracing source evidence from which summaries are derived enables users to assess their accuracy, thereby alleviating this concern. In this paper, we introduce TracSum, a novel benchmark for traceable, aspect-based summarization, in which generated summaries are paired with sentence-level citations, enabling users to trace back to the original context. First, we annotate 500 medical abstracts for seven key medical aspects, yielding 3.5K summary-citations pairs. We then propose a fine-grained evaluation framework for this new task, designed to assess the completeness and consistency of generated content using four metrics. Finally, we introduce a summarization pipeline, Track-Then-Sum, which serves as a baseline method for comparison. In experiments, we evaluate both this baseline and a set of LLMs on TracSum, and conduct a human evaluation to assess the evaluation results. The findings demonstrate that TracSum can serve as an effective benchmark for traceable, aspect-based summarization tasks. We also observe that explicitly performing sentence-level tracking prior to summarization enhances generation accuracy, while incorporating the full context further improves summary completeness. Source code and dataset are available at https://github.com/chubohao/TracSum.

Melanie Subbiah, Akankshya Mishra, Grace Kim, Liyan Tang, Greg Durrett, Kathleen McKeown

Determining faithfulness of a claim to a source document is an important problem across many domains. This task is generally treated as a binary judgment of whether the claim is supported or unsupported in relation to the source. In many cases, though, whether a claim is supported can be ambiguous. For instance, it may depend on making inferences from given evidence, and different people can reasonably interpret the claim as either supported or unsupported based on their agreement with those inferences. Forcing binary labels upon such claims lowers the reliability of evaluation. In this work, we reframe the task to manage the subjectivity involved with factuality judgments of ambiguous claims. We introduce LLM-generated edits of summaries as a method of providing a nuanced evaluation of claims: how much does a summary need to be edited to be unambiguous? Whether a claim gets rewritten and how much it changes can be used as an automatic evaluation metric, the Ambiguity Rewrite Metric (ARM), with a much richer feedback signal than a binary judgment of faithfulness. We focus on the area of narrative summarization as it is particularly rife with ambiguity and subjective interpretation. We show that ARM produces a 21% absolute improvement in annotator agreement on claim faithfulness, indicating that subjectivity is reduced.

Sicong Leng, Yun Xing, Zesen Cheng, Yang Zhou, Hang Zhang, Xin Li, Deli Zhao, Shijian Lu, Chunyan Miao, Lidong Bing

Recent advancements in large multimodal models (LMMs) have significantly enhanced performance across diverse tasks, with ongoing efforts to further integrate additional modalities such as video and audio. However, most existing LMMs remain vulnerable to hallucinations, the discrepancy between the factual multimodal input and the generated textual output, which has limited their applicability in various real-world scenarios. This paper presents the first systematic investigation of hallucinations in LMMs involving the three most common modalities: language, visual, and audio. Our study reveals two key contributors to hallucinations: overreliance on unimodal priors and spurious inter-modality correlations. To address these challenges, we introduce the benchmark The Curse of Multi-Modalities (CMM), which comprehensively evaluates hallucinations in LMMs, providing a detailed analysis of their underlying issues. Our findings highlight key vulnerabilities, including imbalances in modality integration and biases from training data, underscoring the need for balanced cross-modal learning and enhanced hallucination mitigation strategies. Based on our observations and findings, we suggest potential research directions that could enhance the reliability of LMMs.

Deep Learning · Everything Else

Bowen Dong, Minheng Ni, Zitong Huang, Guanglei Yang, Wangmeng Zuo, Lei Zhang

Multimodal hallucination in multimodal large language models (MLLMs) restricts the correctness of MLLMs. However, multimodal hallucinations are multi-sourced and arise from diverse causes. Existing benchmarks fail to adequately distinguish between perception-induced hallucinations and reasoning-induced hallucinations. This failure constitutes a significant issue and hinders the diagnosis of multimodal reasoning failures within MLLMs. To address this, we propose the MIRAGE benchmark, which isolates reasoning hallucinations by constructing questions where input images are correctly perceived by MLLMs yet reasoning errors persist. MIRAGE introduces multi-granular evaluation metrics: accuracy, factuality, and LLMs hallucination score for hallucination quantification. Our analysis reveals strong correlations between question types and specific hallucination patterns, particularly systematic failures of MLLMs in spatial reasoning involving complex relationships (\emph{e.g.}, complex geometric patterns across images). This highlights a critical limitation in the reasoning capabilities of current MLLMs and provides targeted insights for hallucination mitigation on specific types. To address these challenges, we propose Logos, a method that combines curriculum reinforcement fine-tuning to encourage models to generate logic-consistent reasoning chains by stepwise reducing learning difficulty, and collaborative hint inference to reduce reasoning complexity. Logos establishes a baseline on MIRAGE, and reduces the logical hallucinations in original base models. Link: \url{https://bit.ly/25mirage}.

Applications · Language, Speech and Dialog

Jiaqi Wei, Hao Zhou, Xiang Zhang, Di Zhang, Zijie Qiu, Noah Wei, Jinzhe Li, Wanli Ouyang, Siqi Sun

Retrieval-augmented generation (RAG) has become a widely adopted paradigm for enabling knowledge-grounded large language models (LLMs). However, standard RAG pipelines often fail to ensure that model reasoning remains consistent with the evidence retrieved, leading to factual inconsistencies or unsupported conclusions. In this work, we reinterpret RAG as \textit{Retrieval-Augmented Reasoning} and identify a central but underexplored problem: \textit{Reasoning Misalignment}—the divergence between an LLM's internal reasoning trajectory and the evidential constraints provided by retrieval. To address this issue, we propose \textsc{AlignRAG}, a novel iterative framework grounded in \textit{Critique-Driven Alignment (CDA)}. We further introduce \textsc{AlignRAG-auto}, an autonomous variant that dynamically terminates refinement, removing the need to pre-specify the number of critique iterations. At the heart of \textsc{AlignRAG} lies a \textit{contrastive critique synthesis} mechanism that generates retrieval-sensitive critiques while mitigating self-bias. This mechanism trains a dedicated retrieval-augmented \textit{Critic Language Model (CLM)} using labeled critiques that distinguish between evidence-aligned and misaligned reasoning. Empirical evaluations show that our approach significantly improves reasoning fidelity. Our 8B-parameter CLM improves performance over the Self-Refine baseline by \textbf{12.1\%} on out-of-domain tasks and outperforms a standard 72B-parameter CLM by \textbf{2.2\%}. Furthermore, \textsc{AlignRAG-auto} achieves this state-of-the-art performance while dynamically determining the optimal number of refinement steps, enhancing efficiency and usability. \textsc{AlignRAG} remains compatible with existing RAG architectures as a \textit{plug-and-play} module and demonstrates strong robustness under both informative and noisy retrieval scenarios. Overall, \textsc{AlignRAG} offers a principled solution for aligning model reasoning with retrieved evidence, substantially improving the factual reliability and robustness of RAG systems. Our source code is provided at \href{https://github.com/upup-wei/RAG-ReasonAlignment}{link}.

Deep Learning · Algorithms

Yiqun Chen, Lingyong Yan, Weiwei Sun, Xinyu Ma, Yi Zhang, Shuaiqiang Wang, Dawei Yin, Yiming Yang, Jiaxin Mao

Retrieval-augmented generation (RAG) is widely utilized to incorporate external knowledge into large language models, thereby enhancing factuality and reducing hallucinations in question-answering (QA) tasks. A standard RAG pipeline consists of several components, such as query rewriting, document retrieval, document filtering, and answer generation. However, these components are typically optimized separately through supervised fine-tuning, which can lead to misalignments between the objectives of individual components and the overarching aim of generating accurate answers. Although recent efforts have explored using reinforcement learning (RL) to optimize specific RAG components, these approaches often focus on simple pipelines with only two components or do not adequately address the complex interdependencies and collaborative interactions among the modules. To overcome these limitations, we propose treating the complex RAG pipeline with multiple components as a multi-agent cooperative task, in which each component can be regarded as an RL agent. Specifically, we present MMOA-RAG\footnote{The code of MMOA-RAG is on \url{https://github.com/chenyiqun/MMOA-RAG}.}, \textbf{M}ulti-\textbf{M}odule joint \textbf{O}ptimization \textbf{A}lgorithm for \textbf{RAG}, which employs multi-agent reinforcement learning to harmonize all agents' goals toward a unified reward, such as the F1 score of the final answer. Experiments conducted on various QA benchmarks demonstrate that MMOA-RAG effectively boost the overall performance of the pipeline and outperforms existing baselines. Furthermore, comprehensive ablation studies validate the contributions of individual components and demonstrate MMOA-RAG can be adapted to different RAG pipelines and benchmarks.

Computer Vision · Vision Models & Multimodal

Wufei Ma, Yu-Cheng Chou, Qihao Liu, Xingrui Wang, Celso de Melo, Jianwen Xie, Alan Yuille

Despite recent advances on multi-modal models, 3D spatial reasoning remains a challenging task for state-of-the-art open-source and proprietary models. Recent studies explore data-driven approaches and achieve enhanced spatial reasoning performance by fine-tuning models on 3D-related visual question-answering data. However, these methods typically perform spatial reasoning in an implicit manner and often fail on questions that are trivial to humans, even with long chain-of-thought reasoning. In this work, we introduce SpatialReasoner, a novel large vision-language model (LVLM) that addresses 3D spatial reasoning with explicit 3D representations shared between multiple stages--3D perception, computation, and reasoning. Explicit 3D representations provide a coherent interface that supports advanced 3D spatial reasoning and improves the generalization ability to novel question types. Furthermore, by analyzing the explicit 3D representations in multi-step reasoning traces of SpatialReasoner, we study the factual errors and identify key shortcomings of current LVLMs. Results show that our SpatialReasoner achieves improved performance on a variety of spatial reasoning benchmarks, outperforming Gemini 2.0 by 9.2% on 3DSRBench, and generalizes better when evaluating on novel 3D spatial reasoning questions. Our study bridges the 3D parsing capabilities of prior visual foundation models with the powerful reasoning abilities of large language models, opening new directions for 3D spatial reasoning.

Reinforcement Learning · Deep RL

Junyi Li, Hwee Tou Ng

Large language models (LLMs) have significantly advanced in reasoning tasks through reinforcement learning (RL) optimization, achieving impressive capabilities across various challenging benchmarks. However, our empirical analysis reveals a critical drawback: reasoning-oriented RL fine-tuning significantly increases the prevalence of hallucinations. We theoretically analyze the RL training dynamics, identifying high-variance gradient, entropy-induced randomness, and susceptibility to spurious local optima as key factors leading to hallucinations. To address this drawback, we propose Factuality-aware Step-wise Policy Optimization (FSPO), an innovative RL fine-tuning algorithm incorporating explicit factuality verification at each reasoning step. FSPO leverages automated verification against given evidence to dynamically adjust token-level advantage values, incentivizing factual correctness throughout the reasoning process. Experiments across mathematical reasoning and hallucination benchmarks using Qwen2.5 and Llama models demonstrate that FSPO effectively reduces hallucinations while enhancing reasoning accuracy, substantially improving both reliability and performance.

Deep Learning · Generative Models and Autoencoders

Guanghan Wang, Yair Schiff, Subham Sahoo, Volodymyr Kuleshov

Part of the success of diffusion models stems from their ability to perform iterative refinement, i.e., repeatedly correcting outputs during generation. However, modern masked discrete diffusion lacks this capability: when a token is generated, it cannot be updated again, even when it introduces an error. Here, we address this limitation by introducing the remasking diffusion model (ReMDM) sampler, a method that can be applied to pretrained masked diffusion models in a principled way and that is derived from a discrete diffusion model with a custom remasking backward process. Most interestingly, ReMDM endows discrete diffusion with a form of inference-time compute scaling. By increasing the number of sampling steps, ReMDM generates natural language outputs that approach the quality of autoregressive models, whereas when the computation budget is limited, ReMDM better maintains quality. ReMDM also improves sample quality of masked diffusion models for discretized images, and in scientific domains such as molecule design, ReMDM facilitates diffusion guidance and pushes the Pareto frontier of controllability relative to classical masking and uniform noise diffusion. When applied to large pretrained diffusion language models, ReMDM boosts the model’s performance on downstream tasks requiring factual knowledge grasp and reasoning ability.

Computer Vision · Vision Models & Multimodal

Yuxuan Luo, Ryan Yuan, Junwen Chen, Haonan Cai, Ziyi Yue, Yuwei Yang, Fatima Zohra Daha, Ji Li, Zhouhui Lian

In this paper, we introduce knowledge image generation as a new task, alongside the Massive Multi-Discipline Multi-Tier Knowledge-Image Generation Benchmark (MMMG) to probe the reasoning capability of image generation models.Knowledge images have been central to human civilization and to the mechanisms of human learning—a fact underscored by dual-coding theory and the picture-superiority effect.Generating such images is challenging, demanding multimodal reasoning that fuses world knowledge with pixel-level grounding into clear explanatory visuals.To enable comprehensive evaluation, MMMG offers $4,456$ expert-validated (knowledge) image-prompt pairs spanning $10$ disciplines, $6$ educational levels, and diverse knowledge formats such as charts, diagrams, and mind maps. To eliminate confounding complexity during evaluation, we adopt a unified Knowledge Graph (KG) representation. Each KG explicitly delineates a target image’s core entities and their dependencies.We further introduce MMMG-Score to evaluate generated knowledge images. This metric combines factual fidelity, measured by graph-edit distance between KGs, with visual clarity assessment.Comprehensive evaluations of $21$ state-of-the-art text-to-image generation models expose serious reasoning deficits—low entity fidelity, weak relations, and clutter—with GPT-4o achieving an MMMG-Score of only $50.20$, underscoring the benchmark’s difficulty.To spur further progress, we release FLUX-Reason (MMMG-Score of $34.45$), an effective and open baseline that combines a reasoning LLM with diffusion models and is trained on $16,000$ curated knowledge image–prompt pairs.

Applications · Language, Speech and Dialog

Liang Wang, Haonan Chen, Nan Yang, Xiaolong Huang, Zhicheng Dou, Furu Wei

This paper introduces an approach for training o1-like RAG models that retrieve and reason over relevant information step by step before generating the final answer. Conventional RAG methods usually perform a single retrieval step before the generation process, which limits their effectiveness in addressing complex queries due to imperfect retrieval results. In contrast, our proposed method, CoRAG (Chain-of-Retrieval Augmented Generation), allows the model to dynamically reformulate the query based on the evolving state. To train CoRAG effectively, we utilize rejection sampling to automatically generate intermediate retrieval chains, thereby augmenting existing RAG datasets that only provide the correct final answer. At test time, we propose various decoding strategies to scale the model's test-time compute by controlling the length and number of sampled retrieval chains. Experimental results across multiple benchmarks validate the efficacy of CoRAG, particularly in multi-hop question answering tasks, where we observe more than $10$ points improvement in EM score compared to strong baselines. On the KILT benchmark, CoRAG establishes a new state-of-the-art performance across a diverse range of knowledge-intensive tasks. Furthermore, we offer comprehensive analyses to understand the scaling behavior of CoRAG, laying the groundwork for future research aimed at developing factual and grounded foundation models.

Computer Vision · Vision Models & Multimodal

Constantin Venhoff, Ashkan Khakzar, Sonia Joseph, Philip Torr, Neel Nanda

Training vision language models (VLMs) aims to align visual representations from a vision encoder with the textual representations of a pretrained large language model (LLM). However, many VLMs exhibit reduced factual recall performance compared to their LLM backbones, raising the question of how effective multimodal fine-tuning is at extending existing mechanisms within the LLM to visual inputs. We argue that factual recall based on visual inputs requires VLMs to solve a two-hop problem: (1) forming entity representations from visual inputs, and (2) recalling associated factual knowledge based on these entity representations. By benchmarking 14 VLMs with various architectures (LLaVA, Native, Cross-Attention), sizes (7B-124B parameters), and training setups on factual recall tasks against their original LLM backbone models, we find that 11 of 14 models exhibit factual recall degradation. We select three models exhibiting high- and two models with low performance degradation, and use attribution patching, activation patching, and probing to show that degraded VLMs struggle to use the existing factual recall circuit of their LLM backbone, because they resolve the first hop too late in the computation. In contrast, high-performing VLMs resolve entity representations early enough to reuse the existing factual recall mechanism. Finally, we demonstrate two methods to recover performance: patching entity representations from the LLM backbone into the VLM, and prompting with chain-of-thought reasoning. Our results highlight that the speed of early entity resolution critically determines how effective VLMs are in using preexisting LLM mechanisms. More broadly, our work illustrates how mechanistic analysis can explain and unveil systematic failures in multimodal alignment.

Social Aspects · Fairness, Equity, Justice and Safety

Herun Wan, Jiaying Wu, Minnan Luo, Zhi Zeng, Zhixiong Su

Misinformation detectors often rely on superficial cues (i.e., shortcuts) that correlate with misinformation in training data but fail to generalize to the diverse and evolving nature of real-world misinformation. This issue is exacerbated by large language models (LLMs), which can easily generate convincing misinformation using simple prompts. We introduce TruthOverTricks, a unified evaluation paradigm for measuring shortcut learning in misinformation detection. TruthOverTricks categorizes shortcut behaviors into intrinsic shortcut induction and extrinsic shortcut injection, and evaluates seven representative detectors across 14 popular benchmarks, along with two new factual misinformation datasets, NQ-Misinfo and Streaming-Misinfo. Empirical results reveal that existing detectors suffer severe performance degradation when exposed to both naturally occurring and adversarially crafted shortcuts. To address this, we propose the Shortcut Mitigation Framework (SMF), an LLM-augmented data augmentation framework that mitigates shortcut reliance through paraphrasing, factual summarization, and sentiment normalization. SMF consistently enhances robustness across 16 benchmarks, forcing models to rely on deeper semantic understanding rather than shortcut cues.

Social Aspects · Trustworthy Machine Learning

William Overman, Mohsen Bayati

Modern language‑model deployments must often balance competing objectives—for example, helpfulness versus harmlessness, cost versus accuracy, and reward versus safety. We introduce Conformal Arbitrage, a post‑hoc framework that learns a data‑driven threshold to mediate between a Primary model optimized for a primary objective and a more conservative Guardian—which could be another model or a human domain expert—aligned with a guardrail objective. The threshold is calibrated with conformal risk control, yielding finite‑sample, distribution‑free guarantees that the long‑run frequency of undesirable events (such as factual errors or safety violations) does not exceed a user‑specified quota. Because Conformal Arbitrage operates wholly at the API level—without requiring access to model logits or updating model weights—it complements weight‑based alignment techniques and integrates seamlessly with existing cost‑aware cascades. Empirically, Conformal Arbitrage traces an efficient frontier, allowing users to define an acceptable performance level for one objective while maximizing utility in another. We observe that our method outperforms (in terms of accuracy) cost-matched random routing between models. These properties make Conformal Arbitrage a practical, theoretically grounded tool for trustworthy and economical deployment of large language models across a broad range of potentially competing objectives.

Deep Learning · Robustness

Sanjana Ramprasad, Byron Wallace

Modern LLMs can now produce highly readable abstractive summaries, to the point that traditional automated metrics for evaluating summary quality, such as ROUGE, have saturated. However, LLMs still sometimes introduce inaccuracies into summaries, i.e., information inconsistent with or unsupported by the corresponding source. Measuring the occurrence of these often subtle factual inconsistencies automatically has proved challenging. This in turn has motivated development of metrics intended to measure the factual consistency of generated summaries against sources. But are these approaches measuring what they purport to? Or are they mostly exploiting artifacts? In this work, we stress test a range of automatic factuality metrics—including specialized model-based approaches and LLM-based prompting methods—to probe what they actually capture. Using a shallow classifier to separate “easy” examples for factual evaluation—where surface features suffice—from “hard” cases requiring deeper reasoning, we find that all metrics show substantial performance drops on the latter. Furthermore, some metrics are more sensitive to benign, fact-preserving edits than to factual corrections. Building on this observation, we demonstrate that most automatic factuality metrics can be gamed—that is, their scores can be artificially inflated by appending innocuous, content-free sentences to summaries. Among the metrics tested, the LLM prompt-based ChatGPT-DA approach is the most robust and reliable; however, it exhibits a notable caveat: it likely relies more on parametric knowledge than on the provided source when making judgments. Taken together, our findings call into question the reliability of current factuality metrics and prompt a broader reflection on what these metrics are truly measuring. We conclude with concrete recommendations for improving both benchmark design and metric robustness, particularly in light of their vulnerability to superficial manipulations.

Data-centric AI · Data-centric AI methods and tools

Liangliang Zhang, Zhuorui Jiang, Hongliang Chi, Haoyang Chen, Mohammed ElKoumy, Fali Wang, Qiong Wu, Zhengyi Zhou, Shirui Pan, Suhang Wang 等

Knowledge Graph Question Answering (KGQA) systems rely on high-quality benchmarks to evaluate complex multi-hop reasoning. However, despite their widespread use, popular datasets such as WebQSP and CWQ suffer from critical quality issues, including inaccurate or incomplete ground-truth annotations, poorly constructed questions that are ambiguous, trivial, or unanswerable, and outdated or inconsistent knowledge. Through a manual audit of 16 popular KGQA datasets—including WebQSP and CWQ—we find that the average factual correctness rate is only 57%. To address these issues, we introduce KGQAGen, an LLM-in-the-loop framework that systematically resolves these pitfalls. KGQAGen combines structured knowledge grounding, LLM-guided generation, and symbolic verification to produce challenging and verifiable QA instances. Using KGQAGen, we construct KGQAGen-10k, a 10K-scale benchmark grounded in Wikidata, and evaluate a diverse set of KG-RAG models. Experimental results demonstrate that even state-of-the-art systems struggle on this benchmark, highlighting its ability to expose limitations of existing models. Our findings advocate for more rigorous benchmark construction and position KGQAGen as a scalable framework for advancing KGQA evaluation.

Data-centric AI · Data-centric AI methods and tools

Cathy Jiao, Yijun Pan, Emily Xiao, Daisy Sheng, Niket Jain, Hanzhang Zhao, Ishita Dasgupta, Jiaqi Ma, Chenyan Xiong

Data attribution methods quantify the influence of training data on model outputs and are becoming increasingly relevant for a wide range of LLM research and applications, including dataset curation, model interpretability, data valuation. However, there remain critical gaps in systematic LLM-centric evaluation of data attribution methods. To this end, we introduce DATE-LM (Data Attribution Evaluation in Language Models), a unified benchmark for evaluating data attribution methods through real-world LLM applications. DATE-LM measures attribution quality through three key tasks — training data selection, toxicity/bias filtering, and factual attribution. Our benchmark is designed for ease of use, enabling researchers to configure and run large-scale evaluations across diverse tasks and LLM architectures. Furthermore, we use DATE-LM to conduct a large-scale evaluation of existing data attribution methods. Our findings show that no single method dominates across all tasks, data attribution methods have trade-offs with simpler baselines, and method performance is sensitive to task-specific evaluation design. Finally, we release a public leaderboard for quick comparison of methods and to facilitate community engagement. We hope DATE-LM serves as a foundation for future data attribution research in LLMs.

Deep Learning · Attention Mechanisms

Pengbo Wang, Chaozhuo Li, Chenxu Wang, Liwen Zheng, Litian Zhang, Xi Zhang

LLMs have demonstrated unprecedented capabilities in natural language processing, yet their practical deployment remains hindered by persistent factuality and faithfulness hallucinations. While existing methods address these hallucination types independently, they inadvertently induce performance trade-offs, as interventions targeting one type often exacerbate the other. Through empirical and theoretical analysis of activation space dynamics in LLMs, we reveal that these hallucination categories share overlapping subspaces within neural representations, presenting an opportunity for concurrent mitigation. To harness this insight, we propose SPACE, a unified framework that jointly enhances factuality and faithfulness by editing shared activation subspaces. SPACE establishes a geometric foundation for shared subspace existence through dual-task feature modeling, then identifies and edits these subspaces via a hybrid probe strategy combining spectral clustering and attention head saliency scoring. Experimental results across multiple benchmark datasets demonstrate the superiority of our approach.

Applications · Language, Speech and Dialog

Pengcheng Huang, Zhenghao Liu, Yukun Yan, Haiyan Zhao, Xiaoyuan Yi, Hao Chen, Zhiyuan Liu, Maosong Sun, Tong Xiao, Ge Yu 等

Large language models (LLMs) integrated with retrieval-augmented generation (RAG) have improved factuality by grounding outputs in external evidence. However, they remain susceptible to unfaithful generation, where outputs contradict retrieved context despite its relevance and accuracy. Existing approaches aiming to improve faithfulness primarily focus on enhancing the utilization of external context, but often overlook the persistent influence of internal parametric knowledge during generation. In this work, we investigate the internal mechanisms behind unfaithful generation and identify a subset of mid-to-deep feed-forward networks (FFNs) that are disproportionately activated in such cases. Building on this insight, we propose Parametric Knowledge Muting through FFN Suppression (ParamMute), a framework that improves contextual faithfulness by suppressing the activation of unfaithfulness-associated FFNs and calibrating the model toward retrieved knowledge. To evaluate our approach, we introduce CoFaithfulQA, a benchmark specifically designed to evaluate faithfulness in scenarios where internal knowledge conflicts with accurate external evidence. Experimental results show that ParamMute significantly enhances faithfulness across both CoFaithfulQA and the established ConFiQA benchmark, achieving substantial reductions in reliance on parametric memory. These findings underscore the importance of mitigating internal knowledge dominance and provide a new direction for improving LLM trustworthiness in RAG. All codes are available at https://github.com/OpenBMB/ParamMute.

Applications · Language, Speech and Dialog

Xingang Guo, Yaxin Li, XiangYi Kong, YILAN JIANG, Xiayu Zhao, Zhihua Gong, Yufan Zhang, Daixuan Li, Tianle Sang, Beixiao Zhu 等

Modern engineering, spanning electrical, mechanical, aerospace, civil, and computer disciplines, stands as a cornerstone of human civilization and the foundation of our society. However, engineering design poses a fundamentally different challenge for large language models (LLMs) compared with traditional textbook-style problem solving or factual question answering. Although existing benchmarks have driven progress in areas such as language understanding, code synthesis, and scientific problem solving, real-world engineering design demands the synthesis of domain knowledge, navigation of complex trade-offs, and management of the tedious processes that consume much of practicing engineers' time. Despite these shared challenges across engineering disciplines, no benchmark currently captures the unique demands of engineering design work. In this work, we introduce EngDesign, an Engineering Design benchmark that evaluates LLMs' abilities to perform practical design tasks across nine engineering domains. Unlike existing benchmarks that focus on factual recall or question answering, EngDesign uniquely emphasizes LLMs' ability to synthesize domain knowledge, reason under constraints, and generate functional, objective-oriented engineering designs. Each task in EngDesign represents a real-world engineering design problem, accompanied by a detailed task description specifying design goals, constraints, and performance requirements. EngDesign pioneers a simulation-based evaluation paradigm that moves beyond textbook knowledge to assess genuine engineering design capabilities and shifts evaluation from static answer checking to dynamic, simulation-driven functional verification, marking a crucial step toward realizing the vision of engineering Artificial General Intelligence (AGI).