论文检索

Yilin Zhang, Xinran Zhao, Zora Zhiruo Wang, Chenyang Yang, Jiayi Wei, Tongshuang Wu

Retrieval-Augmented Generation (RAG) has become essential for large-scale code generation, grounding predictions in external code corpora to improve factuality. However, a critical yet underexplored aspect of RAG pipelines is chunking—the process of dividing documents into retrievable units. Existing line-based chunking heuristics often break semantic structures, splitting functions or merging unrelated code, which can degrade generation quality. We propose chunking via Abstract Syntax Trees (cAST), a structure-aware method that recursively breaks large AST nodes into smaller chunks and merges sibling nodes while respecting size limits. This approach generates self-contained, semantically coherent units across programming languages and tasks, improving performance on diverse code generation tasks, e.g., boosting Recall@5 by 4.3 points on RepoEval retrieval and Pass@1 by 2.67 points on SWE-bench generation. Our work highlights the importance of structure-aware chunking for scaling retrieval-enhanced code intelligence.

Sanjay Govindan, Maurice Pagnucco, Yang Song

Large Language Models (LLMs) are trained on diverse and often conflicting knowledge spanning multiple domains and time periods. Some of this knowledge is only valid within specific temporal contexts, such as answering the question, “Who is the President of the United States in 2022?” Ensuring LLMs generate time-appropriate responses is crucial for maintaining relevance and accuracy. In this work we explore activation engineering as a method for temporally aligning LLMs to improve factual recall without any training. Activation engineering has predominantly been used to steer subjective and qualitative outcomes such as toxicity or behavior. Our research is one of few that uncovers the bounds of activation engineering on objective outcomes. We explore an activation engineering technique to anchor LLaMA 2, LLaMA 3.1, Qwen 2 and Gemma 2 to specific points in time and examine the effects of varying injection layers and prompting strategies. Our experiments demonstrate up to a 44% and 16% improvement in relative and explicit prompting respectively, achieving comparable performance to the fine-tuning method proposed by Zhao et al. (2024). Notably, for LLaMA 2 and LLaMA 3.1 our approach achieves similar results to the fine-tuning baseline while being significantly more computationally efficient and requiring no pre-aligned datasets.

Yanhong Li, Min Yang, Xiping Hu, Chengming Li

Recent studies have highlighted the remarkable knowledge retention capabilities of Large Language Models (LLMs) like GPT-4, while simultaneously revealing critical limitations in maintaining knowledge currency and accuracy. Existing knowledge editing methodologies, designed to update specific factual information without compromising general model performance, often encounter two fundamental challenges: parameter conflict during knowledge overwriting and excessive computational overhead. In this paper, we introduce ForGet (Forget for Get), a novel approach grounded in the principle of “forgetting before learning”. By pinpointing the location within the LLM that corresponds to the target knowledge, we first erase the outdated knowledge and then insert the new knowledge at this precise spot. ForGet is the first work to leverage a two-phase gradient-based process for knowledge editing, offering a lightweight solution that also delivers superior results. Experimental findings show that our method achieves more effective knowledge editing at a lower cost compared to previous techniques across various base models.

Abhijit Chakraborty, Chahana Dahal, Vivek Gupta

Federated Retrieval-Augmented Generation (Federated RAG) combines Federated Learning (FL),which enables distributed model training without exposing raw data, with Retrieval-Augmented Generation (RAG), which improves the factual accuracy of language models by grounding outputs in external knowledge. As large language models are increasingly deployed in privacy-sensitive domains such as healthcare, finance, and personalized assistance, Federated RAG offers a promising framework for secure, knowledge-intensive natural language processing (NLP). To the best of our knowledge, this paper presents the first systematic mapping study of Federated RAG, covering literature published between 2020 and 2025. Following Kitchenham’s guidelines for evidence-based software engineering, we develop a structured classification of research focuses, contribution types, and application domains. We analyze architectural patterns, temporal trends, and key challenges, including privacy-preserving retrieval, cross-client heterogeneity, and evaluation limitations. Our findings synthesize a rapidly evolving body of research, identify recurring design patterns, and surface open questions, providing a foundation for future work at the intersection of RAG and federated systems.

Haonan Sheng, Dou Hu, Lingwei Wei, Wei Zhou, Songlin Hu

Large language models are prone to generate hallucinations, which can undermine their reliability in high-stakes applications. Some works on LLM hallucination mitigation use the model’s internal signals to contrast different output during inference stage. However, these works often focus on simple forms of hallucinations, and struggle to effectively mitigate hallucinations. To address the issue, this paper exploits hard negative samples to construct a factually weaker model for improving contrastive decoding. We propose a new inference-time method, Regularized Contrastive Decoding (RCD), to capture correct hallucination signals for mitigating hallucinations in LLMs. RCD learns more diverse hallucination patterns via adversarial-aware fine-tuning and mitigates hallucinations via contrastive decoding. Experiments on four hallucination benchmarks demonstrate that our method achieves better LLM hallucination mitigation performance. Further analysis shows RCD generalizes well across different model sizes, task formats, perturbation methods and training data sizes.

Zeyang Liu, Jingfeng Xue, Xiuqi Yang, Wenbiao Du, Jiarun Fu, Junbao Chen, Wenjie Guo, Yong Wang

Large language models (LLMs) exhibit remarkable text-generation capabilities, yet struggle with factual consistency, motivating growing interest in factuality verification. Existing factuality verification methods typically follow a Decompose-Then-Verify paradigm, which improves granularity but suffers from poor scalability and efficiency. We propose a novel Decompose-Embed-Interact paradigm that shifts factuality verification from costly text-level reasoning to efficient alignment in embedding space, effectively mitigating the scalability bottlenecks and computational inefficiencies inherent to prior approaches. While the proposed paradigm promises scalable verification, its implementation faces three practical challenges: efficient decomposition, factually faithful embedding, and accurate verification in embedding space. To address these challenges, we introduce E-Verify, a lightweight framework that resolves them through three specially designed modules, each aligned with a specific stage of the paradigm and designed to preserve scalability and efficiency. Experiments demonstrate that E-Verify significantly improves both decomposition and verification efficiency while maintaining competitive accuracy. These results confirm that the proposed paradigm enables scalable and fine-grained factuality verification with minimal performance trade-offs.

Ruiling Guo, Xinwei Yang, Chen Huang, Tong Zhang, Yong Hu

The effectiveness of large language models (LLMs) to fact-check misinformation remains uncertain, despite their growing use. To this end, we present CANDY, a benchmark designed to systematically evaluate the capabilities and limitations of LLMs in fact-checking Chinese misinformation. Specifically, we curate a carefully annotated dataset of ~20k instances. Our analysis shows that current LLMs exhibit limitations in generating accurate fact-checking conclusions, even when enhanced with chain-of-thought reasoning and few-shot prompting. To understand these limitations, we develop a taxonomy to categorize flawed LLM-generated explanations for their conclusions and identify factual fabrication as the most common failure mode. Although LLMs alone are unreliable for fact-checking, our findings indicate their considerable potential to augment human performance when deployed as assistive tools in scenarios. Our dataset and code can be accessed at https://github.com/SCUNLP/CANDY.

Xi Ai, Mahardika Krisna Ihsani, Min-Yen Kan

Cross-lingual consistency should be considered to assess cross-lingual transferability, maintain the factuality of the model knowledge across languages, and preserve the parity of language model performance. We are thus interested in analyzing, evaluating, and interpreting cross-lingual consistency for factual knowledge.To facilitate our study, we examine multiple pretrained models and tuned models with code-mixed coreferential statements that convey identical knowledge across languages. Interpretability approaches are leveraged to analyze the behavior of a model in cross-lingual contexts, showing different levels of consistency in multilingual models, subject to language families, linguistic factors, scripts, and a bottleneck in cross-lingual consistency on a particular layer. Code-switching training and cross-lingual word alignment objectives show the most promising results, emphasizing the worthiness of cross-lingual alignment supervision and code-switching strategies for both multilingual performance and cross-lingual consistency enhancement. In addition, experimental results suggest promising result for calibrating consistency on test time via activation patching.

Jeongsoo Lee, Daeyong Kwon, Kyohoon Jin

Retrieval-Augmented Generation (RAG) systems are widely adopted in knowledge-intensive NLP tasks, but current evaluations often overlook the structural complexity and multi-step reasoning required in real-world scenarios. These benchmarks overlook key factors such as the interaction between retrieval difficulty and reasoning depth. To address this gap, we propose GRADE, a novel evaluation framework that models task difficulty along two orthogonal dimensions: (1) reasoning depth, defined by the number of inference steps (hops), and (2) semantic distance between the query and its supporting evidence. We construct a synthetic multi-hop QA dataset from factual news articles by extracting knowledge graphs and augmenting them through semantic clustering to recover missing links, allowing us to generate diverse and difficulty-controlled queries. Central to our framework is a 2D difficulty matrix that combines generator-side and retriever-side difficulty. Experiments across multiple domains and models show that error rates strongly correlate with our difficulty measures, validating their diagnostic utility. GRADE enables fine-grained analysis of RAG performance and provides a scalable foundation for evaluating and improving multi-hop reasoning in real-world applications.

Jianjiang Yang, Yanshu Li, Ziyan Huang

While Multimodal Large Language Models (MLLMs) have achieved remarkable progress in open-ended visual question answering, they remain vulnerable to hallucinations. These are outputs that contradict or misrepresent input semantics, posing a critical challenge to the reliability and factual consistency. Existing methods often rely on external verification or post-hoc correction, lacking an internal mechanism to validate outputs directly during training. To bridge this gap, we propose ReLoop, a unified closed-loop training framework that encourages multimodal consistency for cross-modal understanding in MLLMs. ReLoop adopts a ring-shaped structure that integrates three complementary consistency feedback mechanisms, obliging MLLMs to “seeing twice and thinking backwards”. Specifically, ReLoop employs the frozen Consistency Feedback Plugin (CFP), comprising semantic reconstruction, visual description, and an attention supervision module for attention alignment. These components collectively enforce semantic reversibility, visual consistency, and interpretable attention, enabling the model to correct its outputs during training. Extensive evaluations and analyses demonstrate the effectiveness of ReLoop in reducing hallucination rates across multiple benchmarks, establishing a robust method for hallucination mitigation in MLLMs. We will release our source code and data in the camera-ready version. The code is available at: https://github.com/ZiyanHuang11/Reloop-hallucinations.

Ruobing Yao, Yifei Zhang, Shuang Song, Neng Gao, Chenyang Tu

Retrieval-Augmented Generation (RAG) compensates for the static knowledge limitations of Large Language Models (LLMs) by integrating external knowledge, producing responses with enhanced factual correctness and query-specific contextualization. However, it also introduces new attack surfaces such as corpus poisoning at the same time. Most of the existing defense methods rely on the internal knowledge of the model, which conflicts with the design concept of RAG. To bridge the gap, EcoSafeRAG uses sentence-level processing and bait-guided context diversity detection to identify malicious content by analyzing the context diversity of candidate documents without relying on LLM internal knowledge. Experiments show EcoSafeRAG delivers state-of-the-art security with plug-and-play deployment, simultaneously improving clean-scenario RAG performance while maintaining practical operational costs (relatively 1.2 \times latency, 48%-80% token reduction versus Vanilla RAG).

Sichu Liang, Linhai Zhang, Hongyu Zhu, Wenwen Wang, Yulan He, Deyu Zhou

Medical question answering fundamentally relies on accurate clinical knowledge. The dominant paradigm, Retrieval-Augmented Generation (RAG), acquires expertise conceptual knowledge from large-scale medical corpus to guide general-purpose large language models (LLMs) in generating trustworthy answers. However, existing retrieval approaches often overlook the patient-specific factual knowledge embedded in Electronic Health Records (EHRs), which limits the contextual relevance of retrieved conceptual knowledge and hinders its effectiveness in vital clinical decision-making. This paper introduces RGAR, a recurrence generation-augmented retrieval framework that synergistically retrieves both factual and conceptual knowledge from dual sources (i.e., EHRs and the corpus), allowing mutual refinement through iterative interaction. Across three factual-aware medical QA benchmarks, RGAR establishes new state-of-the-art performance among medical RAG systems. Notably, RGAR enables the Llama-3.1-8B-Instruct model to surpass the considerably larger GPT-3.5 augmented with traditional RAG. Our findings demonstrate the benefit of explicitly mining patient-specific factual knowledge during retrieval, consistently improving generation quality and clinical relevance.

Siyuan Zhang, Yichi Zhang, Yinpeng Dong, Hang Su

Large Language Models (LLMs) often struggle to align their responses with objective facts, resulting in the issue of \textbf{factual hallucinations}, which can be difficult to detect and mislead users without relevant knowledge. Although post-training techniques have been employed to mitigate the issue, existing methods usually suffer from poor generalization and trade-offs in other different capabilities. In this paper, we propose to address these by directly augmenting LLM’s fundamental ability to precisely leverage its knowledge and introduce \textbf{PKUE} (\textbf{P}recise \textbf{K}nowledge \textbf{U}tilization \textbf{E}nhancement), which fine-tunes the model on self-generated responses to precise and simple factual questions through preference optimization. Furthermore, we construct \textbf{FactualBench}, a comprehensive and precise factual QA dataset containing 181k Chinese data spanning 21 domains, to facilitate both evaluation and training. Extensive experiments demonstrate that PKUE significantly improves LLM overall performance, with consistent enhancement across factual tasks of various forms, general tasks beyond factuality, and tasks in different language.

Xiaoying Song, Anirban Saha Anik, Dibakar Barua, Pengcheng Luo, Junhua Ding, Lingzi Hong

Health misinformation spreading online poses a significant threat to public health. Researchers have explored methods for automatically generating counterspeech to health misinformation as a mitigation strategy. Existing approaches often produce uniform responses, ignoring that the health literacy level of the audience could affect the accessibility and effectiveness of counterspeech. We propose a Controlled-Literacy framework using retrieval-augmented generation (RAG) with reinforcement learning (RL) to generate tailored counterspeech adapted to different health literacy levels. In particular, we retrieve knowledge aligned with specific health literacy levels, enabling accessible and factual information to support generation. We design a reward function incorporating subjective user preferences and objective readability-based rewards to optimize counterspeech to the target health literacy level. Experiment results show that Controlled-Literacy outperforms baselines by generating more accessible and user-preferred counterspeech. This research contributes to more equitable and impactful public health communication by improving the accessibility and comprehension of counterspeech to health misinformation.

Jiseung Hong, Grace Byun, Seungone Kim, Kai Shu

Large Language Models (LLMs) are expected to provide helpful and harmless responses, yet they often exhibit sycophancy—conforming to user beliefs regardless of factual accuracy or ethical soundness. Prior research on sycophancy has primarily focused on single-turn factual correctness, overlooking the dynamics of real-world interactions. In this work, we introduce SYCON Bench (SYcophantic CONformity benchmark), a novel evaluation suite that assesses sycophantic behavior in multi-turn, free-form conversational settings. Our benchmark measures how quickly a model conforms to the user (Turn of Flip) and how frequently it shifts its stance under sustained user pressure (Number of Flip). Applying SYCON Bench to 17 LLMs across three real-world scenarios, we find that sycophancy remains a prevalent failure mode. Our analysis shows that alignment tuning amplifies sycophantic behavior, whereas model scaling and reasoning optimization strengthen the model’s ability to resist undesirable user views. Reasoning models generally outperform instruction-tuned models but often fail when they over-index on logical exposition instead of directly addressing the user’s underlying beliefs. Finally, we evaluate four additional prompting strategies and demonstrate that adopting a third-person perspective reduces sycophancy by up to 63.8% in debate scenario.

Yihong Liu, Mingyang Wang, Amir Hossein Kargaran, Felicia Körner, Ercong Nie, Barbara Plank, François Yvon, Hinrich Schuetze

Large Language Models (LLMs) are capable of recalling multilingual factual knowledge present in their pretraining data. However, most studies evaluate only the final model, leaving the development of factual recall and crosslingual consistency throughout pretraining largely unexplored. In this work, we trace how factual recall and crosslingual consistency evolve during pretraining, focusing on OLMo-7B as a case study. We find that both accuracy and consistency improve over time for most languages. We show that this improvement is primarily driven by the fact frequency in the pretraining corpus: more frequent facts are more likely to be recalled correctly, regardless of language. Yet, some low-frequency facts in non-English languages can still be correctly recalled. Our analysis reveals that these instances largely benefit from crosslingual transfer of their English counterparts – an effect that emerges predominantly in the early stages of pretraining. We pinpoint two distinct pathways through which multilingual factual knowledge acquisition occurs: (1) frequency-driven learning, which is dominant and language-agnostic, and (2) crosslingual transfer, which is limited in scale and typically constrained to relation types involving named entities. We release our code and data to facilitate further research at https://github.com/cisnlp/multilingual-fact-tracing.

Linxin Song, Taiwei Shi, Jieyu Zhao

Reinforcement finetuning (RFT) has become a standard approach for enhancing the reasoning capabilities of large language models (LLMs). However, its impact on model trustworthiness remains underexplored. In this work, we identify and systematically study a critical side effect of RFT, which we term the hallucination tax: a degradation in refusal behavior causing models to produce hallucinated answers to unanswerable questions confidently. To investigate this, we introduce SUM (Synthetic Unanswerable Math), a high-quality dataset of unanswerable math problems designed to probe models’ ability to recognize an unanswerable question by reasoning from the insufficient or ambiguous information. Our results show that standard RFT training could reduce model refusal rates by more than 80%, which significantly increases model’s tendency to hallucinate. We further demonstrate that incorporating just 10% SUM during RFT substantially restores appropriate refusal behavior, with minimal accuracy trade-offs on solvable tasks. Crucially, this approach enables LLMs to leverage inference-time compute to reason about their own uncertainty and knowledge boundaries, improving generalization not only to out-of-domain math problems but also to factual question answering tasks.

Jinu Lee, Julia Hockenmaier

Step-by-step reasoning is widely used to enhance the reasoning ability of large language models (LLMs) in complex problems. Evaluating the quality of reasoning traces is crucial for understanding and improving LLM reasoning. However, existing evaluation practices are highly inconsistent, resulting in fragmented progress across evaluator design and benchmark development. To address this gap, this survey provides a comprehensive overview of step-by-step reasoning evaluation, proposing a taxonomy of evaluation criteria with four top-level categories (factuality, validity, coherence, and utility). Based on the taxonomy, we review different datasets, evaluator implementations, and recent findings, leading to promising directions for future research.

Zheng Weihua, Roy Ka-Wei Lee, Zhengyuan Liu, Wu Kui, AiTi Aw, Bowei Zou

Multilingual Large Language Models (MLLMs) demonstrate strong generalization across languages, yet they remain prone to hallucinations, especially in low-resource languages, due to training data imbalances. These hallucinations, which include inaccurate or fabricated outputs, are particularly problematic in domain-specific generation tasks (Chataigner et al., 2024). To address this challenge, we propose CCL-XCoT (Curriculum-based Contrastive Learning-based Cross-lingual Chain-of-Thought), a two-stage fine-tuning framework for mitigating hallucination in MLLMs. Our approach first enhances cross-lingual semantic alignment through curriculum-based contrastive learning combined with next-token prediction during continued pre-training. Building on this foundation, we then introduce a cross-lingual Chain-of-Thought (XCoT) prompting strategy during instruction fine-tuning, which guides the model to reason in a high-resource language before generating answers in the target low-resource language. Experimental results show that CCL-XCoT reduces hallucination rates by up to 62% and substantially improves factual knowledge transfer across language pairs, without relying on external retrieval or multi-model ensembles.

Jiahao Cheng, Tiancheng Su, Jia Yuan, Guoxiu He, Jiawei Liu, Xinqi Tao, Jingwen Xie, Huaxia Li

Large Language Models (LLMs) often exhibit hallucinations, generating factually incorrect or semantically irrelevant content in response to prompts. Chain-of-Thought (CoT) prompting can mitigate hallucinations by encouraging step-by-step reasoning, but its impact on hallucination detection remains underexplored. To bridge this gap, we conduct a systematic empirical evaluation. We begin with a pilot experiment, revealing that CoT reasoning significantly affects the LLM’s internal states and token probability distributions. Building on this, we evaluate the impact of various CoT prompting methods on mainstream hallucination detection methods across both instruction-tuned and reasoning-oriented LLMs. Specifically, we examine three key dimensions: changes in hallucination score distributions, variations in detection accuracy, and shifts in detection confidence. Our findings show that while CoT prompting helps reduce hallucination frequency, it also tends to obscure critical signals used for detection, impairing the effectiveness of various detection methods. Our study highlights an overlooked trade-off in the use of reasoning. Code is publicly available at: https://github.com/ECNU-Text-Computing/cot-hallu-detect .