arXiv subjects

Jeff Z. Pan

Publications and source records attributed to Jeff Z. Pan.

At least 55 records · Page 3Linked to original sources

arXiv2025

Prompting Large Language Models with Partial Knowledge for Answering Questions with Unseen Entities

Retrieval-Augmented Generation (RAG) shows impressive performance by supplementing and substituting parametric knowledge in Large Language Models (LLMs). Retrieved knowledge can be divided into three types: explicit answer evidence, implicit answer clue, and insufficient answer context which can be further categorized into totally irrelevant and partially relevant information. Effectively utilizing partially relevant knowledge remains a key challenge for RAG systems, especially in incomplete knowledge base retrieval. Contrary to the conventional view, we propose a new perspective: LLMs can be awakened via partially relevant knowledge already embedded in LLMs. To comprehensively investigate this phenomenon, the triplets located in the gold reasoning path and their variants are used to construct partially relevant knowledge by removing the path that contains the answer. We provide theoretical analysis of the awakening effect in LLMs and support our hypothesis with experiments on two Knowledge Graphs (KGs) Question Answering (QA) datasets. Furthermore, we present a new task, Unseen Entity KGQA, simulating real-world challenges where entity linking fails due to KG incompleteness. Our awakening-based approach demonstrates greater efficacy in practical applications, outperforms traditional methods that rely on embedding-based similarity which are prone to returning noisy information.

cs.CL↗

arXiv2025

Dark Side of Modalities: Reinforced Multimodal Distillation for Multimodal Knowledge Graph Reasoning

The multimodal knowledge graph reasoning (MKGR) task aims to predict the missing facts in the incomplete MKGs by leveraging auxiliary images and descriptions of entities. Existing approaches are trained with single-target objectives, which neglect the probabilistic correlations of entity labels, especially in non-target entities. Moreover, previous studies incorporate all modalities statically or adaptively, overlooking the negative impacts of irrelevant or misleading information in the incompetent modalities. To address these issues, we introduce a novel Reinforced Multimodal Distillation framework, exploiting the Dark Side of Modalities (DSoM) from two perspectives: (1) Dark knowledge from non-target entities: We propose to train a unimodal KGR model through logit distillation to mimic the multimodal soft labels provided by pre-trained multimodal teacher models. The multimodal soft labels could provide rich supervision signals with subtle correlations among both target and non-target entities from multiple perspectives. We further decouple logits into neighbor entities and non-neighbor entities to divide into two types of correlations. (2) Dark side in unhelpful modalities: To exclude the adverse effects of unhelpful modalities, we introduce a reinforced teacher combination mechanism that dynamically selects the optimal set of multimodal teachers for each triple. The agent is trained to maximize the rewards, which are only assigned to the beneficial multimodal combination strategies for the student model. Comprehensive experiments demonstrate the effectiveness of DSoM framework on 5 MKGR datasets. Codes are available at github.com/OreOZhao/DSoM.

cs.MM↗

arXiv2025

Millions of $\text{GeAR}$-s: Extending GraphRAG to Millions of Documents

Recent studies have explored graph-based approaches to retrieval-augmented generation, leveraging structured or semi-structured information -- such as entities and their relations extracted from documents -- to enhance retrieval. However, these methods are typically designed to address specific tasks, such as multi-hop question answering and query-focused summarisation, and therefore, there is limited evidence of their general applicability across broader datasets. In this paper, we aim to adapt a state-of-the-art graph-based RAG solution: $\text{GeAR}$ and explore its performance and limitations on the SIGIR 2025 LiveRAG Challenge.

cs.CL↗

arXiv2025

Can LLMs Evaluate Complex Attribution in QA? Automatic Benchmarking using Knowledge Graphs

Attributed Question Answering (AQA) has attracted wide attention, but there are still several limitations in evaluating the attributions, including lacking fine-grained attribution categories, relying on manual annotations, and failing to compare attributions with only subtle differences. To bridge these gaps, we introduce Complex Attributed Question Answering (CAQA), a large-scale benchmark containing comprehensive attribution categories, automatically generated using Knowledge Graphs (KGs), and complex attribution scenarios. We have conducted extensive experiments to verify the effectiveness of CAQA, including the benchmarking of 25 automatic evaluators, their comparison with human evaluators, the testing of LLM evaluators fine-tuned by CAQA and so on. These experiments also lead to a series of important findings that can benefit the future research of AQA. All the codes and data are publicly accessible at https://github.com/HuuuNan/CAQA-Benchmark.

cs.CL↗

arXiv2025

HyperMono: A Monotonicity-aware Approach to Hyper-Relational Knowledge Representation

In a hyper-relational knowledge graph (HKG), each fact is composed of a main triple associated with attribute-value qualifiers, which express additional factual knowledge. The hyper-relational knowledge graph completion (HKGC) task aims at inferring plausible missing links in a HKG. Most existing approaches to HKGC focus on enhancing the communication between qualifier pairs and main triples, while overlooking two important properties that emerge from the monotonicity of the hyper-relational graphs representation regime. Stage Reasoning allows for a two-step reasoning process, facilitating the integration of coarse-grained inference results derived solely from main triples and fine-grained inference results obtained from hyper-relational facts with qualifiers. In the initial stage, coarse-grained results provide an upper bound for correct predictions, which are subsequently refined in the fine-grained step. More generally, Qualifier Monotonicity implies that by attaching more qualifier pairs to a main triple, we may only narrow down the answer set, but never enlarge it. This paper proposes the HyperMono model for hyper-relational knowledge graph completion, which realizes stage reasoning and qualifier monotonicity. To implement qualifier monotonicity HyperMono resorts to cone embeddings. Experiments on three real-world datasets with three different scenario conditions demonstrate the strong performance of HyperMono when compared to the SoTA.

cs.AI↗

arXiv2025

GeAR: Graph-enhanced Agent for Retrieval-augmented Generation

Retrieval-augmented Generation (RAG) relies on effective retrieval capabilities, yet traditional sparse and dense retrievers inherently struggle with multi-hop retrieval scenarios. In this paper, we introduce GeAR, a system that advances RAG performance through two key innovations: (i) an efficient graph expansion mechanism that augments any conventional base retriever, such as BM25, and (ii) an agent framework that incorporates the resulting graph-based retrieval into a multi-step retrieval framework. Our evaluation demonstrates GeAR's superior retrieval capabilities across three multi-hop question answering datasets. Notably, our system achieves state-of-the-art results with improvements exceeding 10% on the challenging MuSiQue dataset, while consuming fewer tokens and requiring fewer iterations than existing multi-step retrieval systems. The project page is available at https://gear-rag.github.io.

cs.CL↗

arXiv2025

Multi-level Mixture of Experts for Multimodal Entity Linking

Multimodal Entity Linking (MEL) aims to link ambiguous mentions within multimodal contexts to associated entities in a multimodal knowledge base. Existing approaches to MEL introduce multimodal interaction and fusion mechanisms to bridge the modality gap and enable multi-grained semantic matching. However, they do not address two important problems: (i) mention ambiguity, i.e., the lack of semantic content caused by the brevity and omission of key information in the mention's textual context; (ii) dynamic selection of modal content, i.e., to dynamically distinguish the importance of different parts of modal information. To mitigate these issues, we propose a Multi-level Mixture of Experts (MMoE) model for MEL. MMoE has four components: (i) the description-aware mention enhancement module leverages large language models to identify the WikiData descriptions that best match a mention, considering the mention's textual context; (ii) the multimodal feature extraction module adopts multimodal feature encoders to obtain textual and visual embeddings for both mentions and entities; (iii)-(iv) the intra-level mixture of experts and inter-level mixture of experts modules apply a switch mixture of experts mechanism to dynamically and adaptively select features from relevant regions of information. Extensive experiments demonstrate the outstanding performance of MMoE compared to the state-of-the-art. MMoE's code is available at: https://github.com/zhiweihu1103/MEL-MMoE.

cs.CV↗

arXiv2025

Long-Form Information Alignment Evaluation Beyond Atomic Facts

Information alignment evaluators are vital for various NLG evaluation tasks and trustworthy LLM deployment, reducing hallucinations and enhancing user trust. Current fine-grained methods, like FactScore, verify facts individually but neglect inter-fact dependencies, enabling subtle vulnerabilities. In this work, we introduce MontageLie, a challenging benchmark that constructs deceptive narratives by "montaging" truthful statements without introducing explicit hallucinations. We demonstrate that both coarse-grained LLM-based evaluators and current fine-grained frameworks are susceptible to this attack, with AUC-ROC scores falling below 65%. To enable more robust fine-grained evaluation, we propose DoveScore, a novel framework that jointly verifies factual accuracy and event-order consistency. By modeling inter-fact relationships, DoveScore outperforms existing fine-grained methods by over 8%, providing a more robust solution for long-form text alignment evaluation. Our code and datasets are available at https://github.com/dannalily/DoveScore.

cs.CL↗

arXiv2025

Self-Reasoning Language Models: Unfold Hidden Reasoning Chains with Few Reasoning Catalyst

Inference-time scaling has attracted much attention which significantly enhance the performance of Large Language Models (LLMs) in complex reasoning tasks by increasing the length of Chain-of-Thought. These longer intermediate reasoning rationales embody various meta-reasoning skills in human cognition, such as reflection and decomposition, being difficult to create and acquire. In this work, we introduce \textit{Self-Reasoning Language Model} (SRLM), where the model itself can synthesize longer CoT data and iteratively improve performance through self-training. By incorporating a few demonstration examples (i.e., 1,000 samples) on how to unfold hidden reasoning chains from existing responses, which act as a reasoning catalyst, we demonstrate that SRLM not only enhances the model's initial performance but also ensures more stable and consistent improvements in subsequent iterations. Our proposed SRLM achieves an average absolute improvement of more than $+2.5$ points across five reasoning tasks: MMLU, GSM8K, ARC-C, HellaSwag, and BBH on two backbone models. Moreover, it brings more improvements with more times of sampling during inference, such as absolute $+7.89$ average improvement with $64$ sampling times, revealing the in-depth, diverse and creative reasoning paths in SRLM against the strong baseline.

cs.CL↗

arXiv2025

Rethinking Stateful Tool Use in Multi-Turn Dialogues: Benchmarks and Challenges

Existing benchmarks that assess Language Models (LMs) as Language Agents (LAs) for tool use primarily focus on stateless, single-turn interactions or partial evaluations, such as tool selection in a single turn, overlooking the inherent stateful nature of interactions in multi-turn applications. To fulfill this gap, we propose \texttt{DialogTool}, a multi-turn dialogue dataset with stateful tool interactions considering the whole life cycle of tool use, across six key tasks in three stages: 1) \textit{tool creation}; 2) \textit{tool utilization}: tool awareness, tool selection, tool execution; and 3) \textit{role-consistent response}: response generation and role play. Furthermore, we build \texttt{VirtualMobile} -- an embodied virtual mobile evaluation environment to simulate API calls and assess the robustness of the created APIs\footnote{We will use tools and APIs alternatively, there are no significant differences between them in this paper.}. Taking advantage of these artifacts, we conduct comprehensive evaluation on 13 distinct open- and closed-source LLMs and provide detailed analysis at each stage, revealing that the existing state-of-the-art LLMs still cannot perform well to use tools over long horizons.

cs.CL↗

arXiv2025

Masking in Multi-hop QA: An Analysis of How Language Models Perform with Context Permutation

Multi-hop Question Answering (MHQA) adds layers of complexity to question answering, making it more challenging. When Language Models (LMs) are prompted with multiple search results, they are tasked not only with retrieving relevant information but also employing multi-hop reasoning across the information sources. Although LMs perform well on traditional question-answering tasks, the causal mask can hinder their capacity to reason across complex contexts. In this paper, we explore how LMs respond to multi-hop questions by permuting search results (retrieved documents) under various configurations. Our study reveals interesting findings as follows: 1) Encoder-decoder models, such as the ones in the Flan-T5 family, generally outperform causal decoder-only LMs in MHQA tasks, despite being significantly smaller in size; 2) altering the order of gold documents reveals distinct trends in both Flan T5 models and fine-tuned decoder-only models, with optimal performance observed when the document order aligns with the reasoning chain order; 3) enhancing causal decoder-only models with bi-directional attention by modifying the causal mask can effectively boost their end performance. In addition to the above, we conduct a thorough investigation of the distribution of LM attention weights in the context of MHQA. Our experiments reveal that attention weights tend to peak at higher values when the resulting answer is correct. We leverage this finding to heuristically improve LMs' performance on this task. Our code is publicly available at https://github.com/hwy9855/MultiHopQA-Reasoning.

cs.CL↗

arXiv2025

Multi-modal Knowledge Graph Generation with Semantics-enriched Prompts

Multi-modal Knowledge Graphs (MMKGs) have been widely applied across various domains for knowledge representation. However, the existing MMKGs are significantly fewer than required, and their construction faces numerous challenges, particularly in ensuring the selection of high-quality, contextually relevant images for knowledge graph enrichment. To address these challenges, we present a framework for constructing MMKGs from conventional KGs. Furthermore, to generate higher-quality images that are more relevant to the context in the given knowledge graph, we designed a neighbor selection method called Visualizable Structural Neighbor Selection (VSNS). This method consists of two modules: Visualizable Neighbor Selection (VNS) and Structural Neighbor Selection (SNS). The VNS module filters relations that are difficult to visualize, while the SNS module selects neighbors that most effectively capture the structural characteristics of the entity. To evaluate the quality of the generated images, we performed qualitative and quantitative evaluations on two datasets, MKG-Y and DB15K. The experimental results indicate that using the VSNS method to select neighbors results in higher-quality images that are more relevant to the knowledge graph.

cs.AI↗

arXiv2025

BUCA: A Binary Classification Approach to Unsupervised Commonsense Question Answering

Unsupervised commonsense reasoning (UCR) is becoming increasingly popular as the construction of commonsense reasoning datasets is expensive, and they are inevitably limited in their scope. A popular approach to UCR is to fine-tune language models with external knowledge (e.g., knowledge graphs), but this usually requires a large number of training examples. In this paper, we propose to transform the downstream multiple choice question answering task into a simpler binary classification task by ranking all candidate answers according to their reasonableness. To this end, for training the model, we convert the knowledge graph triples into reasonable and unreasonable texts. Extensive experimental results show the effectiveness of our approach on various multiple choice question answering benchmarks. Furthermore, compared with existing UCR approaches using KGs, ours is less data hungry. Our code is available at https://github.com/probe2/BUCA.

cs.CL↗

arXiv2025

N2C2: Nearest Neighbor Enhanced Confidence Calibration for Cross-Lingual In-Context Learning

Recent advancements of in-context learning (ICL) show language models can significantly improve their performance when demonstrations are provided. However, little attention has been paid to model calibration and prediction confidence of ICL in cross-lingual scenarios. To bridge this gap, we conduct a thorough analysis of ICL for cross-lingual sentiment classification. Our findings suggest that ICL performs poorly in cross-lingual scenarios, exhibiting low accuracy and presenting high calibration errors. In response, we propose a novel approach, N2C2, which employs a -nearest neighbors augmented classifier for prediction confidence calibration. N2C2 narrows the prediction gap by leveraging a datastore of cached few-shot instances. Specifically, N2C2 integrates the predictions from the datastore and incorporates confidence-aware distribution, semantically consistent retrieval representation, and adaptive neighbor combination modules to effectively utilize the limited number of supporting instances. Evaluation on two multilingual sentiment classification datasets demonstrates that N2C2 outperforms traditional ICL. It surpasses fine tuning, prompt tuning and recent state-of-the-art methods in terms of accuracy and calibration errors.

cs.CL↗

arXiv2025

Similarity-Based Domain Adaptation with LLMs

Unsupervised domain adaptation leverages abundant labeled data from various source domains to generalize onto unlabeled target data. Prior research has primarily focused on learning domain-invariant features across the source and target domains. However, these methods often require training a model using source domain data, which is time-consuming and can limit model usage for applications with different source data. This paper introduces a simple framework that utilizes the impressive generalization capabilities of Large Language Models (LLMs) for target data annotation without the need of source model training, followed by a novel similarity-based knowledge distillation loss. Our extensive experiments on cross-domain text classification reveal that our framework achieves impressive performance, specifically, 2.44\% accuracy improvement when compared to the SOTA method.

cs.CL↗

arXiv2025

MiCEval: Unveiling Multimodal Chain of Thought's Quality via Image Description and Reasoning Steps

Multimodal Chain of Thought (MCoT) is a popular prompting strategy for improving the performance of multimodal large language models (MLLMs) across a range of complex reasoning tasks. Despite its popularity, there is a notable absence of automated methods for evaluating the quality of reasoning steps in MCoT. To address this gap, we propose Multimodal Chain-of-Thought Evaluation (MiCEval), a framework designed to assess the correctness of reasoning chains by evaluating the quality of both the description and each reasoning step. The evaluation of the description component focuses on the accuracy of the image descriptions, while the reasoning step evaluates the quality of each step as it is conditionally generated based on the preceding steps. MiCEval is built upon a fine-grained dataset with annotations that rate each step according to correctness, relevance, and informativeness. Extensive experiments on four state-of-the-art MLLMs show that step-wise evaluations using MiCEval align more closely with human judgments compared to existing methods based on cosine similarity or fine-tuning approaches. MiCEval datasets and code can be found in https://github.com/alenai97/MiCEval.

cs.CL↗

arXiv2025

An Extensive Evaluation of PDDL Capabilities in off-the-shelf LLMs

In recent advancements, large language models (LLMs) have exhibited proficiency in code generation and chain-of-thought reasoning, laying the groundwork for tackling automatic formal planning tasks. This study evaluates the potential of LLMs to understand and generate Planning Domain Definition Language (PDDL), an essential representation in artificial intelligence planning. We conduct an extensive analysis across 20 distinct models spanning 7 major LLM families, both commercial and open-source. Our comprehensive evaluation sheds light on the zero-shot LLM capabilities of parsing, generating, and reasoning with PDDL. Our findings indicate that while some models demonstrate notable effectiveness in handling PDDL, others pose limitations in more complex scenarios requiring nuanced planning knowledge. These results highlight the promise and current limitations of LLMs in formal planning tasks, offering insights into their application and guiding future efforts in AI-driven planning paradigms.

cs.AI↗

arXiv2025

GenTool: Enhancing Tool Generalization in Language Models through Zero-to-One and Weak-to-Strong Simulation

Large Language Models (LLMs) can enhance their capabilities as AI assistants by integrating external tools, allowing them to access a wider range of information. While recent LLMs are typically fine-tuned with tool usage examples during supervised fine-tuning (SFT), questions remain about their ability to develop robust tool-usage skills and can effectively generalize to unseen queries and tools. In this work, we present GenTool, a novel training framework that prepares LLMs for diverse generalization challenges in tool utilization. Our approach addresses two fundamental dimensions critical for real-world applications: Zero-to-One Generalization, enabling the model to address queries initially lacking a suitable tool by adopting and utilizing one when it becomes available, and Weak-to-Strong Generalization, allowing models to leverage enhanced versions of existing tools to solve queries. To achieve this, we develop synthetic training data simulating these two dimensions of tool usage and introduce a two-stage fine-tuning approach: optimizing tool ranking, then refining tool selection. Through extensive experiments across four generalization scenarios, we demonstrate that our method significantly enhances the tool-usage capabilities of LLMs ranging from 1B to 8B parameters, achieving performance that surpasses GPT-4o. Furthermore, our analysis also provides valuable insights into the challenges LLMs encounter in tool generalization.

cs.CL↗