arXiv subjects

Ruirui Chen

Publications and source records attributed to Ruirui Chen.

At least 19 recordsLinked to original sources

arXiv2026

M$^3$R-Bench: A Unified Benchmark for Evidence-Grounded Multimodal Metaphor Understanding

Multimodal metaphor understanding requires identifying cross-domain Target-Source mappings and inferring sentiment from visual and textual evidence. However, existing benchmarks largely evaluate these subtasks separately and lack evidence-grounded explanations, limiting their ability to assess coherent cross-modal reasoning. To enable unified and evidence-grounded evaluation, we introduce M$^3$R-Bench, a benchmark of 1,000 human-annotated image-text instances with joint labels for metaphor occurrence, Target-Source mapping, and sentiment, alongside stage-wise explanations for evidence identification, mapping establishment, and sentiment inference. Evaluations reveal that existing models frequently rely on textual shortcuts and produce Target-Source mappings inconsistent with visual evidence. To address this evidence-mapping mismatch, we propose M$^3$R-Reasoner, which combines curriculum-based reasoning supervision with task-aware reinforcement learning. Built on an 8B-parameter backbone, M$^3$R-Reasoner achieves 37.90% Joint Label Accuracy (JLA) metric, outperforming the strongest proprietary baseline by 19.50 percentage points. It further surpasses GPT-5.5 in image-text evidence and sentiment justification scores by 28.45 and 28.11 points, respectively. The dataset and code are available at https://github.com/hongshi4/m3r-bench-artifact.

cs.CL↗

arXiv2026

Are Diffusion Language Models Good Database Analysts?

Diffusion Language Models (DLMs) enable iterative global refinement, but their effectiveness in Natural Language to SQL (NL2SQL) remains unclear because heterogeneous architectures and evaluation setups hinder fair comparison with autoregressive (AR) models. Their iterative denoising process permits sequence-wide revision, offering a different way to handle the long-range structural dependencies in SQL. We present a systematic study of DLM-based NL2SQL under a unified framework that standardizes generation and execution across DLM architectures. We further introduce SQL-D1, an agentic inference framework combining database-aware retrieval, multi-trajectory denoising, execution-guided verification, and selection based on execution denotations. Across diverse benchmark settings, strong DLMs are competitive with some open-source AR models and offer flexible trade-offs between efficiency and accuracy, although they still trail the strongest AR systems on difficult queries. Test-time scaling yields substantial gains, whereas standard supervised fine-tuning yields unstable outcomes across the evaluated DLMs. Stronger model variants exhibit fewer schema-grounding errors, but failures involving complex clauses and query nesting persist. These results clarify the capabilities and limitations of DLMs for database analysis and provide a standardized foundation for future work. Our code is available at https://anonymous.4open.science/r/SQL-D1-82DF.

cs.DB↗

arXiv2026

ConsultMind:Towards Automated Diagnostic Consultation via Uncertainty-Aware Reasoning

Diagnostic consultation is an online sequential decision-making process in which clinicians gather evidence through patient interaction until a diagnosis is sufficiently supported. Automating this process requires adaptive inquiry and interpretable decisions. Bayesian networks offer a natural foundation by updating diagnostic posteriors as evidence accumulates, but their use in open-ended consultation raises two challenges: linking diagnostic hypotheses to potential inquiries and translating evolving posteriors into consultation decisions. We introduce AutoDisym, an automated pipeline that integrates diagnostic knowledge with heterogeneous diagnosis-labeled clinical narratives to construct a Disorder--Symptom Bayesian Network (DSBN). Building on the DSBN, we propose ConsultMind, an uncertainty-aware framework that updates disorder posteriors after each response and uses posterior uncertainty to guide inquiry and diagnosis. We evaluate both methods across psychiatry, respiratory medicine, fever clinics, and three public datasets. The results show that AutoDisym can automatically construct high-quality DSBNs and that ConsultMind consistently improves diagnostic performance and explanation soundness. For example, AutoDisym achieves macro-averaged F1 scores of 81.37 for canonical symptoms and 72.19 for manifestations using GPT-5.6-Sol. ConsultMind improves Top-1 and Top-3 diagnostic accuracy by up to 22.15 and 37.89 percentage points, respectively. Physician evaluation further shows that ConsultMind improves the quality of ranking explanations, differential diagnoses, and diagnosis rationales across LLMs of different scales. This work offers a promising approach to automatic diagnostic consultation.

cs.AI↗

arXiv2026

Selective Knowledge Edit Reversal via Gated Singular Vector Shrinkage

Knowledge editing provides an efficient way to update factual knowledge in large language models. However, malicious edits may introduce safety risks, making it necessary to reverse undesirable editing effects. Existing reversal methods for parameter-modifying edits mainly focus on global removal, which may also erase beneficial edits that should be preserved. In this paper, we study selective reversal of edited knowledge, where the goal is to reverse targeted edited facts while preserving the remaining edited facts. Based on the hypothesis that each edit is sparsely encoded within the dominant subspace of the edited matrix, we propose a spectral-based reversal framework that locates edit-sensitive components within the dominant singular subspace of edited weights. Experiments across multiple settings demonstrate the effectiveness of our method in reversing selected edits while preserving unrelated edited facts. These results suggest that different edits are sparsely encoded within dominant singular components and can be separable when the number of edits is moderate, making selective spectral reversal a promising direction for locating edit-specific components and repairing edited language models.

cs.CL↗

arXiv2026

CoMMET: A Psychologically Grounded Benchmark for Evaluating Theory of Mind in Multimodal LLMs

Theory of Mind (ToM)-the ability to reason about the mental states of oneself and others-is a cornerstone of human social intelligence. As Multimodal Large Language Models (MLLMs) become ubiquitous in real-world applications, validating their capacity for this level of social reasoning is essential for effective and natural interactions. However, existing benchmarks for assessing ToM in MLLMs are limited; most rely solely on text inputs and focus narrowly on belief-related tasks. In this paper, we propose a new multimodal benchmark dataset, CoMMET, a comprehensive mental states and moral evaluation task inspired by the Theory of Mind Booklet Task. CoMMET expands the scope of evaluation by covering a broader range of mental states and introducing multi-turn testing. To the best of our knowledge, this is the first psychology-grounded benchmark to evaluate MLLMs across multiple mental states in a multimodal, open-ended, and multi-turn setting. Through a comprehensive assessment of MLLMs across different families and sizes, we analyze the strengths and limitations of current models and identify directions for future improvement.

cs.CL↗

arXiv2026

Are Large Language Models Effective Knowledge Graph Constructors?

Knowledge graphs (KGs) are widely used in knowledge-intensive applications, yet it remains unclear how effectively current large language models (LLMs) can construct document-grounded KGs in a zero-shot, schema-free setting without relying on complex task-specific frameworks. We introduce Detail-to-Abstract Hierarchical Knowledge Graph (D2A-HKG) construction framework, which decomposes KG construction into three stages: initial extraction, splitting, and abstraction, and evaluates the resulting graphs from both semantic and structural perspectives. Using seven frontier LLMs, we benchmark zero-shot KG construction on CMW-Lit, a dataset derived from published paediatric research articles on children's mental well-being. CMW-Lit provides a challenging test bed due to its heterogeneous evidence, interconnected factors, and complex, statistically qualified relationships. Our results show that state-of-the-art LLMs can generally produce relevant and document-faithful triples with limited hallucination, while exhibiting substantially different extraction behaviors across the construction stages. These findings provide empirical insight into the strengths and limitations of frontier LLMs for direct knowledge graph construction. We further release CMW-Lit and the resulting knowledge graphs as resources for future research, with the generated graphs providing a strong foundation for expert refinement and downstream knowledge-intensive applications.

cs.CL↗

arXiv2026

MELLA: Bridging Linguistic Capability and Cultural Groundedness for Low-Resource Language MLLMs

Multimodal Large Language Models (MLLMs) perform strongly in high-resource languages, yet often produce fluent but culturally "thin" descriptions in low-resource settings. We argue that this failure is not merely a linguistic limitation: culture-specific visual knowledge depends on native visual-textual alignments that translation-centric pipelines rarely provide. We present MELLA, a multimodal dataset across eight low-resource languages, designed to support linguistic fluency and cultural groundedness. MELLA uses a dual-source strategy that combines native web image-alt-text pairs for culture-grounded supervision with generated-and-translated image descriptions for linguistically rich supervision, explicitly separating two learning signals often conflated in multilingual multimodal data. Through controlled diagnostic fine-tuning on multiple MLLM backbones, we show that MELLA mitigates cultural hallucination by helping models recognize and articulate culturally specific entities overlooked by translation-based adaptation. Our findings highlight data alignment, rather than model modification alone, as a path toward culturally grounded multimodal understanding in low-resource languages. The dataset is available at https://opendatalab.com/applyMultilingualCorpus.

cs.CV↗

arXiv2026

CausalChaos! Dataset for Comprehensive Causal Action Question Answering Over Longer Causal Chains Grounded in Dynamic Visual Scenes

Causal video question answering (QA) has garnered increasing interest, yet existing datasets often lack depth in causal reasoning. To address this gap, we capitalize on the unique properties of cartoons and construct CausalChaos!, a novel, challenging causal Why-QA dataset built upon the iconic "Tom and Jerry" cartoon series. Cartoons use the principles of animation that allow animators to create expressive, unambiguous causal relationships between events to form a coherent storyline. Utilizing these properties, along with thought-provoking questions and multi-level answers (answer and detailed causal explanation), our questions involve causal chains that interconnect multiple dynamic interactions between characters and visual scenes. These factors demand models to solve more challenging, yet well-defined causal relationships. We also introduce hard incorrect answer mining, including a causally confusing version that is even more challenging. While models perform well, there is much room for improvement, especially, on open-ended answers. We identify more advanced/explicit causal relationship modeling & joint modeling of vision and language as the immediate areas for future efforts to focus upon. Along with the other complementary datasets, our new challenging dataset will pave the way for these developments in the field. Project Page: https://github.com/LUNAProject22/CausalChaos

cs.CV↗

arXiv2026

ReTool-Video: Recursive Tool-Using Video Agents with Meta-Augmented Tool Grounding

Video understanding requires active evidence seeking, motivating tool-augmented video agents for temporal reasoning, cross-modal understanding, and complex question answering. Existing video agents have improved video reasoning with retrieval, memory, frame inspection, and verifier tools, but they still face two limitations: (1) a coarse tool space that lacks fine-grained operations for compositional reasoning; and (2) a flat action space that forces high-level video intents into primitive executable tool calls. In this paper, we address these challenges with two complementary designs. First, we construct a MetaAug-Video Tool Library (MVTL), an extensible tool library with 134 registered tools, including 26 base tools for general multimodal signal processing and 108 meta tools for filtering, aggregation, reranking, formatting, and other intermediate-result operations. MVTL supports dual-level access to both structured video information and raw modal evidence, enabling diverse video reasoning scenarios. Second, we propose ReTool-Video, a recursive tool-using method that grounds high-level video intents into executable tool chains. In ReTool-Video, matched actions are executed directly, while unmatched intents are delegated to a resolver for parameter repair, tool substitution, or decomposition. This allows abstract actions such as temporal merging, cross-modal verification, or repeated-event aggregation to be progressively translated into concrete multimodal operations at runtime. Experiments on MVBench, MLVU, and Video-MME w/o sub. show that ReTool-Video consistently outperforms strong baselines. Further analysis demonstrates that recursive grounding and fine-grained meta tools improve the stability and effectiveness of complex video understanding.

cs.CV↗

arXiv2025

From Grounding to Manipulation: Case Studies of Foundation Model Integration in Embodied Robotic Systems

Foundation models (FMs) are increasingly used to bridge language and action in embodied agents, yet the operational characteristics of different FM integration strategies remain under-explored -- particularly for complex instruction following and versatile action generation in changing environments. This paper examines three paradigms for building robotic systems: end-to-end vision-language-action (VLA) models that implicitly integrate perception and planning, and modular pipelines incorporating either vision-language models (VLMs) or multimodal large language models (LLMs). We evaluate these paradigms through two focused case studies: a complex instruction grounding task assessing fine-grained instruction understanding and cross-modal disambiguation, and an object manipulation task targeting skill transfer via VLA finetuning. Our experiments in zero-shot and few-shot settings reveal trade-offs in generalization and data efficiency. By exploring performance limits, we distill design implications for developing language-driven physical agents and outline emerging challenges and opportunities for FM-powered robotics in real-world conditions.

cs.RO↗

arXiv2025

Theory of Mind in Large Language Models: Assessment and Enhancement

Theory of Mind (ToM)-the ability to reason about the mental states of oneself and others-is a cornerstone of human social intelligence. As Large Language Models (LLMs) become increasingly integrated into daily life, understanding their ability to interpret and respond to human mental states is crucial for enabling effective interactions. In this paper, we review LLMs' ToM capabilities by analyzing both evaluation benchmarks and enhancement strategies. For evaluation, we focus on recently proposed and widely used story-based benchmarks. For enhancement, we provide an in-depth analysis of recent methods aimed at improving LLMs' ToM abilities. Furthermore, we outline promising directions for future research to further advance these capabilities and better adapt LLMs to more realistic and diverse scenarios. Our survey serves as a valuable resource for researchers interested in evaluating and advancing LLMs' ToM capabilities.

cs.CL↗

arXiv2025

Sub-Connected Hybrid Beamfocusing Design for RSMA-Enabled Near-Field Communications with Imperfect CSI and SIC

Near-field spherical waves inherently encode both direction and distance information, enabling spotlight-like beam focusing for targeted interference mitigation. However, whether such beam focusing can fully eliminate interference under perfect and imperfect channel state information (CSI), rendering advanced interference management schemes unnecessary, remains an open question. To address this, we investigate rate-splitting multiple access (RSMA)-enabled near-field communications (NFC) under imperfect SCI. Our transmit scheme employs a sub-connected hybrid analog-digital (HAD) architecture to reduce hardware overhead while incorporating imperfect successive interference cancellation (SIC) for practical implementation. A minimum rate maximization problem is formulated by jointly optimizing the analog beamfocuser, the digital beamfocuser, and the common rate allocation. To solve the non-convex problem, we develop a penalty-based block coordinate descent (BCD) algorithm, deriving closed-form expressions for the optimal analog and digital beamfocusers solutions. Furthermore, to reduce computational complexity, we propose a low-complexity algorithm, where analog and digital beamfocusers are designed in two separate stages. Simulation results underscore that: 1) beamfocusing alone is insufficient to fully suppress interference even under perfect CSI; 2) RSMA exhibits superior interference management over SDMA under imperfect CSI and SIC conditions; 3) sub-connected HAD architecture delivers near-optimal digital beamfocusing performance with fewer radio frequency chains.

cs.IT↗

arXiv2025

Beyond Output Matching: Bidirectional Alignment for Enhanced In-Context Learning

Large language models (LLMs) have shown impressive few-shot generalization on many tasks via in-context learning (ICL). Despite their success in showing such emergent abilities, the scale and complexity of larger models also lead to unprecedentedly high computational demands and deployment challenges. In reaction, researchers explore transferring the powerful capabilities of larger models to more efficient and compact models by typically aligning the output of smaller (student) models with that of larger (teacher) models. Existing methods either train student models on the generated outputs of teacher models or imitate their token-level probability distributions. However, these distillation methods pay little to no attention to the input, which also plays a crucial role in ICL. Based on the finding that the performance of ICL is highly sensitive to the selection of demonstration examples, we propose Bidirectional Alignment (BiAlign) to fully leverage the models' preferences for ICL examples to improve the ICL abilities of student models. Specifically, we introduce the alignment of input preferences between student and teacher models by incorporating a novel ranking loss, in addition to aligning the token-level output distribution. With extensive experiments and analysis, we demonstrate that BiAlign can consistently outperform existing baselines on a variety of tasks involving language understanding, reasoning, and coding.

cs.CL↗

arXiv2025

Relevant or Random: Can LLMs Truly Perform Analogical Reasoning?

Analogical reasoning is a unique ability of humans to address unfamiliar challenges by transferring strategies from relevant past experiences. One key finding in psychology is that compared with irrelevant past experiences, recalling relevant ones can help humans better handle new tasks. Coincidentally, the NLP community has also recently found that self-generating relevant examples in the context can help large language models (LLMs) better solve a given problem than hand-crafted prompts. However, it is yet not clear whether relevance is the key factor eliciting such capability, i.e., can LLMs benefit more from self-generated relevant examples than irrelevant ones? In this work, we systematically explore whether LLMs can truly perform analogical reasoning on a diverse set of reasoning tasks. With extensive experiments and analysis, we show that self-generated random examples can surprisingly achieve comparable or even better performance on certain tasks, e.g., 4% performance boost on GSM8K with random biological examples. We find that the accuracy of self-generated examples is the key factor and subsequently design two novel methods with improved performance and significantly reduced inference costs. Overall, we aim to advance a deeper understanding of LLM analogical reasoning and hope this work stimulates further research in the design of self-generated contexts.

cs.CL↗

arXiv2024

LLM-Based Multi-Hop Question Answering with Knowledge Graph Integration in Evolving Environments

The important challenge of keeping knowledge in Large Language Models (LLMs) up-to-date has led to the development of various methods for incorporating new facts. However, existing methods for such knowledge editing still face difficulties with multi-hop questions that require accurate fact identification and sequential logical reasoning, particularly among numerous fact updates. To tackle these challenges, this paper introduces Graph Memory-based Editing for Large Language Models (GMeLLo), a straightforward and effective method that merges the explicit knowledge representation of Knowledge Graphs (KGs) with the linguistic flexibility of LLMs. Beyond merely leveraging LLMs for question answering, GMeLLo employs these models to convert free-form language into structured queries and fact triples, facilitating seamless interaction with KGs for rapid updates and precise multi-hop reasoning. Our results show that GMeLLo significantly surpasses current state-of-the-art (SOTA) knowledge editing methods in the multi-hop question answering benchmark, MQuAKE, especially in scenarios with extensive knowledge edits.

cs.CL↗

arXiv2024

Air-to-Ground Cooperative OAM Communications

For users in hotspot region, orbital angular momentum (OAM) can realize multifold increase of spectrum efficiency (SE), and the flying base station (FBS) can rapidly support the real-time communication demand. However, the hollow divergence and alignment requirement impose crucial challenges for users to achieve air-to-ground OAM communications, where there exists the line-of-sight path. Therefore, we propose the air-to-ground cooperative OAM communication (ACOC) scheme, which can realize OAM communications for users with size-limited devices. The waist radius is adjusted to guarantee the maximum intensity at the cooperative users (CUs). We derive the closed-form expression of the optimal FBS position, which satisfies the antenna alignment for two cooperative user groups (CUGs). Furthermore, the selection constraint is given to choose two CUGs composed of four CUs. Simulation results are provided to validate the optimal FBS position and the SE superiority of the proposed ACOC scheme.

cs.IT↗

arXiv2024

Precoding Based Downlink OAM-MIMO Communications with Rate Splitting

Orbital angular momentum (OAM) and rate splitting (RS) are the potential key techniques for the future wireless communications. As a new orthogonal resource, OAM can achieve the multifold increase of spectrum efficiency to relieve the scarcity of the spectrum resource, but how to enhance the privacy performance imposes crucial challenge for OAM communications. RS technique divides the information into private and common parts, which can guarantee the privacies for all users. In this paper, we integrate the RS technique into downlink OAM-MIMO communications, and study the precoding optimization to maximize the sum capacity. First, the concentric uniform circular arrays (UCAs) are utilized to construct the downlink transmission framework of OAM-MIMO communications with RS. Particularly, users in the same user pair utilize RS technique to obtain the information and different user pairs use different OAM modes. Then, we derive the OAM-MIMO channel model, and formulate the sum capacity maximization problem. Finally, based on the fractional programming, the optimal precoding matrix is obtained to maximize the sum capacity by using quadratic transformation. Extensive simulation results show that by using the proposed precoding optimization algorithm, OAM-MIMO communications with RS can achieve higher sum capacity than the traditional communication schemes.

cs.IT↗

arXiv2024

Joint Power Allocation and Placement Scheme for UAV-assisted IoT with QoS Guarantee

In the disaster and remote regions, unmanned aerial vehicles (UAVs) can assist the data acquisition for Internet of Things (IoT). How to cover massive IoT devices (IDs), which require diverse quality-of-service (QoS), is a crucial challenge. For UAV-assisted IoT, this paper studies the deployment scheme with QoS guarantee to place multiple UAVs for covering all ground IDs and maximizing the average data rate of UAVs. First, for the ground ID, we propose the QoS demand based power allocation (QDPA) algorithm to solve the diversity of QoS with respect to data rate demand. Then, the data rate maximization placement (DRMP) algorithm is proposed to optimize the placement of single UAV. Finally, based on QDPA and DRMP algorithms, we propose the joint power allocation and placement (JPAP) scheme with QoS guarantee, which can cover massive IDs, to deploy multiple UAVs for maximizing the average UAV data rate. Simulation experiments are conducted to verify the superiority of our proposed JPAP scheme, which can minimize the UAV number and maximize the average data rate of UAVs.

cs.IT↗