arXiv subjects

Peifeng Li

Publications and source records attributed to Peifeng Li.

At least 19 recordsLinked to original sources

arXiv2026

Hint-Guided Diversified Policy Optimization for LLM Reasoning

Recent developments in Large Language Models (LLMs) have showcased impressive reasoning capabilities, with Reinforcement Learning with Verifiable Rewards (RLVR) being a promising enhancement strategy. However, existing reward mechanisms are constrained to the outcome-level correctness and lack explicit signals to guide the model to consider diverse solutions. In contrast, human problem solving typically involves evaluating multiple potential approaches and selecting the most reliable solution, a cognitive process that current RLVR frameworks do not explicitly incentivize. Inspired by this, we propose Hint-Guided Diversified Policy Optimization (HDPO), allowing the model to first list all potential candidate solution outlines as hints and then select the most reliable one for further reasoning. HDPO comprises two stages of Cold Start for Structured Reasoning and Hint-Guided Diversified Reinforcement Learning to incentivize the model to generate diverse and reliable solutions following the ``propose-select-think'' trajectory. Experimental results show that HDPO effectively boosts LLM reasoning and enhances the diversity of candidate solutions as well as the LLM's ability to identify reliable solutions.

cs.CL↗

arXiv2026

OmniSelect: Dynamic Modality-Aware Token Compression for Efficient Omni-modal Large Language Models

Omnimodal large language models (OmniLLMs) have recently gained increasing attention for unified audio-video understanding. However, processing long multimodal token sequences introduces substantial computational overhead, making efficient token compression crucial. Existing methods typically rely on fixed, modality-specific guidance, which fails to account for the varying importance of modalities across different queries. To address this limitation, we propose $\textbf{OmniSelect}$, a training-free, modality-adaptive token pruning framework that dynamically selects appropriate compression strategies for multimodal inputs. Specifically, we leverage a lightweight AudioCLIP model to estimate cross-modal relevance and categorize each input into three pruning regimes: Audio-Centric, Video-Centric, and Uniform pruning. Based on these relevance scores, OmniSelect further performs fine-grained token pruning within each temporal group, adaptively allocating pruning ratios to preserve informative tokens across modalities. By explicitly modeling modality preference and enabling dynamic strategy selection, OmniSelect effectively avoids the pitfalls of one-size-fits-all compression. Extensive experiments demonstrate that our method achieves efficient multimodal token reduction while maintaining strong performance, without requiring any additional training.

cs.CV↗

arXiv2026

DraDDP: A Multimodal Multi-Party Dialogue Discourse Parsing Dataset

Multi-party dialogue discourse parsing aims to identify dependency structures and relation types between utterances in conversations. Previous studies are mostly limited to textual modality or two-party dialogue, failing to meet the multimodal and multi-party settings. In this paper, we construct the first publicly available English multimodal dataset DraDDP for multi-party dialogue discourse parsing, based on American TV dramas. DraDDP contains 495 dialogue segments with 6,374 utterances and 9.1 hours of parallel video content, covering rich multi-party interaction scenarios. Moreover, we establish comprehensive benchmarks by evaluating this task on DraDDP and conducting in-depth analysis on the impact of different modalities. Experimental results demonstrate the value of multimodal information in capturing dialogue structures and relation types. We will publicly release the dataset, annotation guidelines, and code to promote future research in multimodal dialogue understanding.

cs.CL↗

arXiv2026

When to Trust Tools? Adaptive Tool Trust Calibration For Tool-Integrated Math Reasoning

Large reasoning models (LRMs) have achieved strong performance enhancement through scaling test time computation, but due to the inherent limitations of the underlying language models, they still have shortcomings in tasks that require precise computation and extensive knowledge reserves. Tool-Integrated Reasoning (TIR) has emerged as a promising paradigm that incorporates tool call and execution within the reasoning trajectory. Although recent works have released some powerful open-source TIR models, our analysis reveals that these models still suffer from critical deficiencies. We find that when the reasoning of the model conflicts with the tool results, the model tends to believe in its own reasoning. And there are cases where the tool results are correct but are ignored by the model, resulting in incorrect answers, which we define as "Tool Ignored''. This indicates that the model does not know when to trust or ignore the tool. To overcome these limitations, We introduce Adaptive Tool Trust Calibration (ATTC), a novel framework that guides the model to adaptively choose to trust or ignore the tool results based on the confidence score of generated code blocks. The experimental results from various open-source TIR models of different sizes and across multiple datasets demonstrate that ATTC effectively reduces the "Tool Ignored" issue, resulting in a performance increase of 4.1% to 7.5%.

cs.CL↗

arXiv2026

Multi-Faceted Self-Consistent Preference Alignment for Query Rewriting in Conversational Search

Conversational Query Rewriting (CQR) aims to rewrite ambiguous queries to achieve more efficient conversational search. Early studies have predominantly focused on the rewriting in isolation, ignoring the feedback from query rewrite, passage retrieval and response generation in the rewriting process. To address this issue, we propose Multi-Faceted Self-Consistent Preference Aligned CQR (MSPA-CQR). Specifically, we first construct self-consistent preference alignment data from three dimensions (rewriting, retrieval, and response) to generate more diverse rewritten queries. Then we propose prefix guided multi-faceted direct preference optimization to learn preference information from three different dimensions. The experimental results show that our MSPA-CQR is effective in both in- and out-of-distribution scenarios.

cs.CL↗

arXiv2026

Discourse Coherence and Response-Guided Context Rewriting for Multi-Party Dialogue Generation

Previous research on multi-party dialogue generation has predominantly leveraged structural information inherent in dialogues to directly inform the generation process. However, the prevalence of colloquial expressions and incomplete utterances in dialogues often impedes comprehension and weakens the fidelity of dialogue structure representations, which is particularly pronounced in multi-party dialogues. In this work, we propose a novel framework DRCR (Discourse coherence and Response-guided Context Rewriting) to improve multi-party dialogue generation through dialogue context rewriting. Specifically, DRCR employs two complementary feedback signals, discourse coherence and response quality, to construct preference data for both context rewriting and response generation. Moreover, we propose a dynamic self-evolution learning method that allows the rewriter and responder to continuously enhance their capabilities through mutual interaction in an iterative training loop. Comprehensive experiments conducted on four multi-party dialogue datasets substantiate the effectiveness of DRCR.

cs.CL↗

arXiv2025

ICR: Iterative Clarification and Rewriting for Conversational Search

Most previous work on Conversational Query Rewriting employs an end-to-end rewriting paradigm. However, this approach is hindered by the issue of multiple fuzzy expressions within the query, which complicates the simultaneous identification and rewriting of multiple positions. To address this issue, we propose a novel framework ICR (Iterative Clarification and Rewriting), an iterative rewriting scheme that pivots on clarification questions. Within this framework, the model alternates between generating clarification questions and rewritten queries. The experimental results show that our ICR can continuously improve retrieval performance in the clarification-rewriting iterative process, thereby achieving state-of-the-art performance on two popular datasets.

cs.CL↗

arXiv2025

Improving Dialogue Discourse Parsing through Discourse-aware Utterance Clarification

Dialogue discourse parsing aims to identify and analyze discourse relations between the utterances within dialogues. However, linguistic features in dialogues, such as omission and idiom, frequently introduce ambiguities that obscure the intended discourse relations, posing significant challenges for parsers. To address this issue, we propose a Discourse-aware Clarification Module (DCM) to enhance the performance of the dialogue discourse parser. DCM employs two distinct reasoning processes: clarification type reasoning and discourse goal reasoning. The former analyzes linguistic features, while the latter distinguishes the intended relation from the ambiguous one. Furthermore, we introduce Contribution-aware Preference Optimization (CPO) to mitigate the risk of erroneous clarifications, thereby reducing cascading errors. CPO enables the parser to assess the contributions of the clarifications from DCM and provide feedback to optimize the DCM, enhancing its adaptability and alignment with the parser's requirements. Extensive experiments on the STAC and Molweni datasets demonstrate that our approach effectively resolves ambiguities and significantly outperforms the state-of-the-art (SOTA) baselines.

cs.CL↗

arXiv2025

Enhancing Goal-oriented Proactive Dialogue Systems via Consistency Reflection and Correction

Goal-oriented proactive dialogue systems are designed to guide user conversations seamlessly towards specific objectives by planning a goal-oriented path. However, previous research has focused predominantly on optimizing these paths while neglecting the inconsistencies that may arise between generated responses and dialogue contexts, including user profiles, dialogue history, domain knowledge, and subgoals. To address this issue, we introduce a model-agnostic two-stage Consistency Reflection and Correction (CRC) framework. Specifically, in the consistency reflection stage, the model is prompted to reflect on the discrepancies between generated responses and dialogue contexts, identifying inconsistencies and suggesting possible corrections. In the consistency correction stage, the model generates responses that are more consistent with the dialogue context based on these reflection results. We conducted experiments on various model architectures with different parameter sizes, including encoder-decoder models (BART, T5) and decoder-only models (GPT-2, DialoGPT, Phi3, Mistral and LLaMA3), and the experimental results on three datasets demonstrate that our CRC framework significantly improves the consistency between generated responses and dialogue contexts.

cs.CL↗

arXiv2025

FMNV: A Dataset of Media-Published News Videos for Fake News Detection

News media, particularly video-based platforms, have become deeply embed-ded in daily life, concurrently amplifying the risks of misinformation dissem-ination. Consequently, multimodal fake news detection has garnered signifi-cant research attention. However, existing datasets predominantly comprise user-generated videos characterized by crude editing and limited public en-gagement, whereas professionally crafted fake news videos disseminated by media outlets-often politically or virally motivated-pose substantially greater societal harm. To address this gap, we construct FMNV, a novel da-taset exclusively composed of news videos published by media organizations. Through empirical analysis of existing datasets and our curated collection, we categorize fake news videos into four distinct types. Building upon this taxonomy, we employ Large Language Models (LLMs) to automatically generate deceptive content by manipulating authentic media-published news videos. Furthermore, we propose FMNVD, a baseline model featuring a dual-stream architecture that integrates spatio-temporal motion features from a 3D ResNeXt-101 backbone and static visual semantics from CLIP. The two streams are fused via an attention-based mechanism, while co-attention modules refine the visual, textual, and audio features for effective multi-modal aggregation. Comparative experiments demonstrate both the generali-zation capability of FMNV across multiple baselines and the superior detec-tion efficacy of FMNVD. This work establishes critical benchmarks for de-tecting high-impact fake news in media ecosystems while advancing meth-odologies for cross-modal inconsistency analysis. Our dataset is available in https://github.com/DennisIW/FMNV.

cs.CV↗

arXiv2025

Incomplete Utterance Rewriting with Editing Operation Guidance and Utterance Augmentation

Although existing fashionable generation methods on Incomplete Utterance Rewriting (IUR) can generate coherent utterances, they often result in the inclusion of irrelevant and redundant tokens in rewritten utterances due to their inability to focus on critical tokens in dialogue context. Furthermore, the limited size of the training datasets also contributes to the insufficient training of the IUR model. To address the first issue, we propose a multi-task learning framework EO-IUR (Editing Operation-guided Incomplete Utterance Rewriting) that introduces the editing operation labels generated by sequence labeling module to guide generation model to focus on critical tokens. Furthermore, we introduce a token-level heterogeneous graph to represent dialogues. To address the second issue, we propose a two-dimensional utterance augmentation strategy, namely editing operation-based incomplete utterance augmentation and LLM-based historical utterance augmentation. The experimental results on three datasets demonstrate that our EO-IUR outperforms previous state-of-the-art (SOTA) baselines in both open-domain and task-oriented dialogue. The code will be available at https://github.com/Dewset/EO-IUR.

cs.CL↗

arXiv2025

Two-stage Incomplete Utterance Rewriting on Editing Operation

Previous work on Incomplete Utterance Rewriting (IUR) has primarily focused on generating rewritten utterances based solely on dialogue context, ignoring the widespread phenomenon of coreference and ellipsis in dialogues. To address this issue, we propose a novel framework called TEO (\emph{Two-stage approach on Editing Operation}) for IUR, in which the first stage generates editing operations and the second stage rewrites incomplete utterances utilizing the generated editing operations and the dialogue context. Furthermore, an adversarial perturbation strategy is proposed to mitigate cascading errors and exposure bias caused by the inconsistency between training and inference in the second stage. Experimental results on three IUR datasets show that our TEO outperforms the SOTA models significantly.

cs.CL↗

arXiv2025

Exploring Model Editing for LLM-based Aspect-Based Sentiment Classification

Model editing aims at selectively updating a small subset of a neural model's parameters with an interpretable strategy to achieve desired modifications. It can significantly reduce computational costs to adapt to large language models (LLMs). Given its ability to precisely target critical components within LLMs, model editing shows great potential for efficient fine-tuning applications. In this work, we investigate model editing to serve an efficient method for adapting LLMs to solve aspect-based sentiment classification. Through causal interventions, we trace and determine which neuron hidden states are essential for the prediction of the model. By performing interventions and restorations on each component of an LLM, we identify the importance of these components for aspect-based sentiment classification. Our findings reveal that a distinct set of mid-layer representations is essential for detecting the sentiment polarity of given aspect words. Leveraging these insights, we develop a model editing approach that focuses exclusively on these critical parts of the LLM, leading to a more efficient method for adapting LLMs. Our in-domain and out-of-domain experiments demonstrate that this approach achieves competitive results compared to the currently strongest methods with significantly fewer trainable parameters, highlighting a more efficient and interpretable fine-tuning strategy.

cs.CL↗

arXiv2025

Multimodal Fake News Video Explanation: Dataset, Analysis and Evaluation

Multimodal fake news videos are difficult to interpret because they require comprehensive consideration of the correlation and consistency between multiple modes. Existing methods deal with fake news videos as a classification problem, but it's not clear why news videos are identified as fake. Without proper explanation, the end user may not understand the underlying meaning of the falsehood. Therefore, we propose a new problem - Fake news video Explanation (FNVE) - given a multimodal news post containing a video and title, our goal is to generate natural language explanations to reveal the falsity of the news video. To that end, we developed FakeVE, a new dataset of 2,672 fake news video posts that can definitively explain four real-life fake news video aspects. In order to understand the characteristics of fake news video explanation, we conducted an exploratory analysis of FakeVE from different perspectives. In addition, we propose a Multimodal Relation Graph Transformer (MRGT) based on the architecture of multimodal Transformer to benchmark FakeVE. The empirical results show that the results of the various benchmarks (adopted by FakeVE) are convincing and provide a detailed analysis of the differences in explanation generation of the benchmark models.

cs.CV↗

arXiv2024

Rethinking the Key Factors for the Generalization of Remote Sensing Stereo Matching Networks

Stereo matching, a critical step of 3D reconstruction, has fully shifted towards deep learning due to its strong feature representation of remote sensing images. However, ground truth for stereo matching task relies on expensive airborne LiDAR data, thus making it difficult to obtain enough samples for supervised learning. To improve the generalization ability of stereo matching networks on cross-domain data from different sensors and scenarios, in this paper, we dedicate to study key training factors from three perspectives. (1) For the selection of training dataset, it is important to select data with similar regional target distribution as the test set instead of utilizing data from the same sensor. (2) For model structure, cascaded structure that flexibly adapts to different sizes of features is preferred. (3) For training manner, unsupervised methods generalize better than supervised methods, and we design an unsupervised early-stop strategy to help retain the best model with pre-trained weights as the basis. Extensive experiments are conducted to support the previous findings, on the basis of which we present an unsupervised stereo matching network with good generalization performance. We release the source code and the datasets at https://github.com/Elenairene/RKF_RSSM to reproduce the results and encourage future work.

cs.CV↗

arXiv2024

Official-NV: An LLM-Generated News Video Dataset for Multimodal Fake News Detection

News media, especially video news media, have penetrated into every aspect of daily life, which also brings the risk of fake news. Therefore, multimodal fake news detection has recently garnered increased attention. However, the existing datasets are comprised of user-uploaded videos and contain an excess amounts of superfluous data, which introduces noise into the model training process. To address this issue, we construct a dataset named Official-NV, comprising officially published news videos. The crawl officially published videos are augmented through the use of LLMs-based generation and manual verification, thereby expanding the dataset. We also propose a new baseline model called OFNVD, which captures key information from multimodal features through a GLU attention mechanism and performs feature enhancement and modal aggregation via a cross-modal Transformer. Benchmarking the dataset and baselines demonstrates the effectiveness of our model in multimodal news detection.

cs.CV↗

arXiv2024

Quantifying Self-diagnostic Atomic Knowledge in Chinese Medical Foundation Model: A Computational Analysis

Foundation Models (FMs) have the potential to revolutionize the way users self-diagnose through search engines by offering direct and efficient suggestions. Recent studies primarily focused on the quality of FMs evaluated by GPT-4 or their ability to pass medical exams, no studies have quantified the extent of self-diagnostic atomic knowledge stored in FMs' memory, which is the basis of foundation models to provide factual and reliable suggestions. In this paper, we first constructed a benchmark of Self-diagnostic Atomic Knowledge (SdAK), including the most common types of atomic knowledge involved in self-diagnostic queries, with 17 atomic types and a total of 14, 048 pieces of atomic knowledge. Then, we evaluated both generic and open-source Chinese medical FMs on the benchmark. The experimental results showcase that generic FMs perform better than medical FMs in terms of self-diagnostic atomic knowledge. Error analysis revealed that both generic and medical FMs are sycophantic, e.g., always catering to users' claims when it comes to unknown knowledge. We further explored different types of data commonly adopted for fine-tuning medical FMs, i.e., real-world, semi-distilled, and distilled data, and found that distilled data can benefit FMs most. The code and data are available at https://github.com/FreedomIntelligence/SDAK.

cs.CL↗

arXiv2024

Advancing Topic Segmentation and Outline Generation in Chinese Texts: The Paragraph-level Topic Representation, Corpus, and Benchmark

Topic segmentation and outline generation strive to divide a document into coherent topic sections and generate corresponding subheadings, unveiling the discourse topic structure of a document. Compared with sentence-level topic structure, the paragraph-level topic structure can quickly grasp and understand the overall context of the document from a higher level, benefitting many downstream tasks such as summarization, discourse parsing, and information retrieval. However, the lack of large-scale, high-quality Chinese paragraph-level topic structure corpora restrained relative research and applications. To fill this gap, we build the Chinese paragraph-level topic representation, corpus, and benchmark in this paper. Firstly, we propose a hierarchical paragraph-level topic structure representation with three layers to guide the corpus construction. Then, we employ a two-stage man-machine collaborative annotation method to construct the largest Chinese Paragraph-level Topic Structure corpus (CPTS), achieving high quality. We also build several strong baselines, including ChatGPT, to validate the computability of CPTS on two fundamental tasks (topic segmentation and outline generation) and preliminarily verified its usefulness for the downstream task (discourse parsing).

cs.CL↗