arXiv subjects

Yuchen Wu

Publications and source records attributed to Yuchen Wu.

At least 55 records · Page 3Linked to original sources

arXiv2025

AffordBot: 3D Fine-grained Embodied Reasoning via Multimodal Large Language Models

Effective human-agent collaboration in physical environments requires understanding not only what to act upon, but also where the actionable elements are and how to interact with them. Existing approaches often operate at the object level or disjointedly handle fine-grained affordance reasoning, lacking coherent, instruction-driven grounding and reasoning. In this work, we introduce a new task: Fine-grained 3D Embodied Reasoning, which requires an agent to predict, for each referenced affordance element in a 3D scene, a structured triplet comprising its spatial location, motion type, and motion axis, based on a task instruction. To solve this task, we propose AffordBot, a novel framework that integrates Multimodal Large Language Models (MLLMs) with a tailored chain-of-thought (CoT) reasoning paradigm. To bridge the gap between 3D input and 2D-compatible MLLMs, we render surround-view images of the scene and project 3D element candidates into these views, forming a rich visual representation aligned with the scene geometry. Our CoT pipeline begins with an active perception stage, prompting the MLLM to select the most informative viewpoint based on the instruction, before proceeding with step-by-step reasoning to localize affordance elements and infer plausible interaction motions. Evaluated on the SceneFun3D dataset, AffordBot achieves state-of-the-art performance, demonstrating strong generalization and physically grounded reasoning with only 3D point cloud input and MLLMs.

cs.CV↗

arXiv2025

Beyond One-Way Influence: Bidirectional Opinion Dynamics in Multi-Turn Human-LLM Interactions

Large language model (LLM)-powered chatbots are increasingly used for opinion exploration. Prior research examined how LLMs alter user views, yet little work extended beyond one-way influence to address how user input can affect LLM responses and how such bi-directional influence manifests throughout the multi-turn conversations. This study investigates this dynamic through 50 controversial-topic discussions with participants (N=266) across three conditions: static statements, standard chatbot, and personalized chatbot. Results show that human opinions barely shifted, while LLM outputs changed more substantially, narrowing the gap between human and LLM stance. Personalization amplified these shifts in both directions compared to the standard setting. Analysis of multi-turn conversations further revealed that exchanges involving participants' personal stories were most likely to trigger stance changes for both humans and LLMs. Our work highlights the risk of over-alignment in human-LLM interaction and the need for careful design of personalized chatbots to more thoughtfully and stably align with users.

cs.HC↗

arXiv2025

Beyond Likes: How Normative Feedback Complements Engagement Signals on Social Media

Many online platforms incorporate engagement signals, such as likes, into their interface design to boost engagement. However, these signals can unintentionally elevate content that may not support normatively desirable behavior, especially when toxic content correlates strongly with popularity indicators. In this study, we propose structured prosocial feedback as a complementary signal, which highlights content quality based on normative criteria. We design and implement an LLM-based feedback system, which evaluates user comments based on principles from positive psychology, such as individual well-being. A pre-registered user study then examines how existing peer-based (popularity) and the new expert-based feedback interact to shape users' reposting behavior in a social media setting. Results show that peer feedback increases conformity to popularity cues, while expert feedback shifts choices toward normatively higher-quality content. This illustrates the added value of normative cues and underscores the potential benefits of incorporating such signals into platform feedback systems to foster healthier online environments.

cs.HC↗

arXiv2025

Robust Knowledge Editing via Explicit Reasoning Chains for Distractor-Resilient Multi-Hop QA

Large language models (LLMs) encode vast amounts of world knowledge but remain static once trained, making the timely integration of emerging facts prohibitively expensive via full retraining. Knowledge-editing techniques have thus emerged to inject or overwrite specific facts into LLMs, yet they either over-rely on superficial cues or incur complex, iterative pipelines that collapse under noisy, multi-hop conditions. We introduce Reason-KE, an end-to-end reasoning-chain-based editing framework that steers a pretrained LLM through four structured stages-fact acknowledgment, relevance determination, selective application, and final reasoning-to filter distractors in a single pass. Trained on MQuAKE-CF with up to four irrelevant facts, Reason-KE elevates Qwen2.5-7B's multi-hop QA accuracy to 90.2% while suffering merely a 6.3% drop under heavy distraction and <1% when answers are leaked. Our quantitative analysis confirms Reason-KE's resilience and efficiency, establishing a new state-of-the-art for reliable LLM knowledge updates.

cs.CL↗

arXiv2025

StructVRM: Aligning Multimodal Reasoning with Structured and Verifiable Reward Models

Existing Vision-Language Models often struggle with complex, multi-question reasoning tasks where partial correctness is crucial for effective learning. Traditional reward mechanisms, which provide a single binary score for an entire response, are too coarse to guide models through intricate problems with multiple sub-parts. To address this, we introduce StructVRM, a method that aligns multimodal reasoning with Structured and Verifiable Reward Models. At its core is a model-based verifier trained to provide fine-grained, sub-question-level feedback, assessing semantic and mathematical equivalence rather than relying on rigid string matching. This allows for nuanced, partial credit scoring in previously intractable problem formats. Extensive experiments demonstrate the effectiveness of StructVRM. Our trained model, Seed-StructVRM, achieves state-of-the-art performance on six out of twelve public multimodal benchmarks and our newly curated, high-difficulty STEM-Bench. The success of StructVRM validates that training with structured, verifiable rewards is a highly effective approach for advancing the capabilities of multimodal models in complex, real-world reasoning domains.

cs.AI↗

arXiv2025

Seed-Prover: Deep and Broad Reasoning for Automated Theorem Proving

LLMs have demonstrated strong mathematical reasoning abilities by leveraging reinforcement learning with long chain-of-thought, yet they continue to struggle with theorem proving due to the lack of clear supervision signals when solely using natural language. Dedicated domain-specific languages like Lean provide clear supervision via formal verification of proofs, enabling effective training through reinforcement learning. In this work, we propose \textbf{Seed-Prover}, a lemma-style whole-proof reasoning model. Seed-Prover can iteratively refine its proof based on Lean feedback, proved lemmas, and self-summarization. To solve IMO-level contest problems, we design three test-time inference strategies that enable both deep and broad reasoning. Seed-Prover proves $78.1\%$ of formalized past IMO problems, saturates MiniF2F, and achieves over 50\% on PutnamBench, outperforming the previous state-of-the-art by a large margin. To address the lack of geometry support in Lean, we introduce a geometry reasoning engine \textbf{Seed-Geometry}, which outperforms previous formal geometry engines. We use these two systems to participate in IMO 2025 and fully prove 5 out of 6 problems. This work represents a significant advancement in automated mathematical reasoning, demonstrating the effectiveness of formal verification with long chain-of-thought reasoning.

cs.AI↗

arXiv2025

OceanVive: An Immersive Visualization System for Communicating Complex Oceanic Phenomena

Communicating the complexity of oceanic phenomena-such as hypoxia and acidification-poses a persistent challenge for marine science. Despite advances in sensing technologies and computational models, conventional formats like static visualizations and text-based reports often fall short in conveying the dynamics of ocean changes. To address this gap, we present OceanVive, an immersive and interactive visualization system that transforms complex ocean datasets into navigable spatial narratives. OceanVive incorporates an exploratory panel on a table-sized tablet for managing immersive content on a large screen and integrates adaptive visual encodings, contextual storytelling, and intuitive navigation pathways to support effective communication. We validate the system through expert interviews, demonstrating its potential to enhance science communication and promote deeper public understanding.

cs.HC↗

arXiv2025

CriticLean: Critic-Guided Reinforcement Learning for Mathematical Formalization

Translating natural language mathematical statements into formal, executable code is a fundamental challenge in automated theorem proving. While prior work has focused on generation and compilation success, little attention has been paid to the critic phase-the evaluation of whether generated formalizations truly capture the semantic intent of the original problem. In this paper, we introduce CriticLean, a novel critic-guided reinforcement learning framework that elevates the role of the critic from a passive validator to an active learning component. Specifically, first, we propose the CriticLeanGPT, trained via supervised fine-tuning and reinforcement learning, to rigorously assess the semantic fidelity of Lean 4 formalizations. Then, we introduce CriticLeanBench, a benchmark designed to measure models' ability to distinguish semantically correct from incorrect formalizations, and demonstrate that our trained CriticLeanGPT models can significantly outperform strong open- and closed-source baselines. Building on the CriticLean framework, we construct FineLeanCorpus, a dataset comprising over 285K problems that exhibits rich domain diversity, broad difficulty coverage, and high correctness based on human evaluation. Overall, our findings highlight that optimizing the critic phase is essential for producing reliable formalizations, and we hope our CriticLean will provide valuable insights for future advances in formal mathematical reasoning.

cs.CL↗

arXiv2025

VIS-Shepherd: Constructing Critic for LLM-based Data Visualization Generation

Data visualization generation using Large Language Models (LLMs) has shown promising results but often produces suboptimal visualizations that require human intervention for improvement. In this work, we introduce VIS-Shepherd, a specialized Multimodal Large Language Model (MLLM)-based critic to evaluate and provide feedback for LLM-generated data visualizations. At the core of our approach is a framework to construct a high-quality visualization critique dataset, where we collect human-created visualization instances, synthesize corresponding LLM-generated instances, and construct high-quality critiques. We conduct both model-based automatic evaluation and human preference studies to evaluate the effectiveness of our approach. Our experiments show that even small (7B parameters) open-source MLLM models achieve substantial performance gains by leveraging our high-quality visualization critique dataset, reaching levels comparable to much larger open-source or even proprietary models. Our work demonstrates significant potential for MLLM-based automated visualization critique and indicates promising directions for enhancing LLM-based data visualization generation. Our project page: https://github.com/bopan3/VIS-Shepherd.

cs.CV↗

arXiv2025

Learning to Lead: Incentivizing Strategic Agents in the Dark

We study an online learning version of the generalized principal-agent model, where a principal interacts repeatedly with a strategic agent possessing private types, private rewards, and taking unobservable actions. The agent is non-myopic, optimizing a discounted sum of future rewards and may strategically misreport types to manipulate the principal's learning. The principal, observing only her own realized rewards and the agent's reported types, aims to learn an optimal coordination mechanism that minimizes strategic regret. We develop the first provably sample-efficient algorithm for this challenging setting. Our approach features a novel pipeline that combines (i) a delaying mechanism to incentivize approximately myopic agent behavior, (ii) an innovative reward angle estimation framework that uses sector tests and a matching procedure to recover type-dependent reward functions, and (iii) a pessimistic-optimistic LinUCB algorithm that enables the principal to explore efficiently while respecting the agent's incentive constraints. We establish a near optimal $\tilde{O}(\sqrt{T}) $ regret bound for learning the principal's optimal policy, where $\tilde{O}(\cdot) $ omits logarithmic factors. Our results open up new avenues for designing robust online learning algorithms for a wide range of game-theoretic settings involving private types and strategic agents.

cs.LG↗

arXiv2025

Imago Obscura: An Image Privacy AI Co-pilot to Enable Identification and Mitigation of Risks

Users often struggle to navigate the privacy / publicity boundary in sharing images online: they may lack awareness of image privacy risks and/or the ability to apply effective mitigation strategies. To address this challenge, we introduce and evaluate Imago Obscura, an AI-powered, image-editing copilot that enables users to identify and mitigate privacy risks with images they intend to share. Driven by design requirements from a formative user study with 7 image-editing experts, Imago Obscura enables users to articulate their image-sharing intent and privacy concerns. The system uses these inputs to surface contextually pertinent privacy risks, and then recommends and facilitates application of a suite of obfuscation techniques found to be effective in prior literature -- e.g., inpainting, blurring, and generative content replacement. We evaluated Imago Obscura with 15 end-users in a lab study and found that it greatly improved users' awareness of image privacy risks and their ability to address those risks, allowing them to make more informed sharing decisions.

cs.HC↗

arXiv2025

Edit Once, Update Everywhere: A Simple Framework for Cross-Lingual Knowledge Synchronization in LLMs

Knowledge editing allows for efficient adaptation of large language models (LLMs) to new information or corrections without requiring full retraining. However, prior methods typically focus on either single-language editing or basic multilingual editing, failing to achieve true cross-linguistic knowledge synchronization. To address this, we present a simple and practical state-of-the-art (SOTA) recipe Cross-Lingual Knowledge Democracy Edit (X-KDE), designed to propagate knowledge from a dominant language to other languages effectively. Our X-KDE comprises two stages: (i) Cross-lingual Edition Instruction Tuning (XE-IT), which fine-tunes the model on a curated parallel dataset to modify in-scope knowledge while preserving unrelated information, and (ii) Target-language Preference Optimization (TL-PO), which applies advanced optimization techniques to ensure consistency across languages, fostering the transfer of updates. Additionally, we contribute a high-quality, cross-lingual dataset, specifically designed to enhance knowledge transfer across languages. Extensive experiments on the Bi-ZsRE and MzsRE benchmarks show that X-KDE significantly enhances cross-lingual performance, achieving an average improvement of +8.19%, while maintaining high accuracy in monolingual settings.

cs.CL↗

arXiv2025

Three-dimensional neural network driving self-interference digital holography enables high-fidelity, non-scanning volumetric fluorescence microscopy

We present a deep learning driven computational approach to overcome the limitations of self-interference digital holography that imposed by inferior axial imaging performances. We demonstrate a 3D deep neural network model can simultaneously suppresses the defocus noise and improves the spatial resolution and signal-to-noise ratio of conventional numerical back-propagation-obtained holographic reconstruction. Compared with existing 2D deep neural networks used for hologram reconstruction, our 3D model exhibits superior performance in enhancing the resolutions along all the three spatial dimensions. As the result, 3D non-scanning volumetric fluorescence microscopy can be achieved, using 2D self-interference hologram as input, without any mechanical and opto-electronic scanning and complicated system calibration. Our method offers a high spatiotemporal resolution 3D imaging approach which can potentially benefit, for example, the visualization of dynamics of cellular structure and measurement of 3D behavior of high-speed flow field.

physics.optics↗

arXiv2025

Lower Bounds for the Convergence of Tensor Power Iteration on Random Overcomplete Models

Tensor decomposition serves as a powerful primitive in statistics and machine learning, and has numerous applications in problems such as learning latent variable models or mixture of Gaussians. In this paper, we focus on using power iteration to decompose an overcomplete random tensor. Past work studying the properties of tensor power iteration either requires a non-trivial data-independent initialization, or is restricted to the undercomplete regime. Moreover, several papers implicitly suggest that logarithmically many iterations (in terms of the input dimension) are sufficient for the power method to recover one of the tensor components. Here we present a novel analysis of the dynamics of tensor power iteration from random initialization in the overcomplete regime, where the tensor rank is much greater than its dimension. Surprisingly, we show that polynomially many steps are necessary for convergence of tensor power iteration to any of the true component, which refutes the previous conjecture. On the other hand, our numerical experiments suggest that tensor power iteration successfully recovers tensor components for a broad range of parameters in polynomial time. To further complement our empirical evidence, we prove that a popular objective function for tensor decomposition is strictly increasing along the power iteration path. Our proof is based on the Gaussian conditioning technique, which has been applied to analyze the approximate message passing (AMP) algorithm. The major ingredient of our argument is a conditioning lemma that allows us to generalize AMP-type analysis to non-proportional limit and polynomially many iterations of the power method.

cs.LG↗

arXiv2025

Sharp Analysis of Power Iteration for Tensor PCA

We investigate the power iteration algorithm for the tensor PCA model introduced in Richard and Montanari (2014). Previous work studying the properties of tensor power iteration is either limited to a constant number of iterations, or requires a non-trivial data-independent initialization. In this paper, we move beyond these limitations and analyze the dynamics of randomly initialized tensor power iteration up to polynomially many steps. Our contributions are threefold: First, we establish sharp bounds on the number of iterations required for power method to converge to the planted signal, for a broad range of the signal-to-noise ratios. Second, our analysis reveals that the actual algorithmic threshold for power iteration is smaller than the one conjectured in literature by a polylog(n) factor, where n is the ambient dimension. Finally, we propose a simple and effective stopping criterion for power iteration, which provably outputs a solution that is highly correlated with the true signal. Extensive numerical experiments verify our theoretical results.

cs.LG↗

arXiv2025

High temperature surface state in Kondo insulator U$_3$Bi$_4$Ni$_3$

The resurgence of interest in Kondo insulators has been driven by two major mysteries: the presence of metallic surface states and the observation of quantum oscillations. To further explore these mysteries, it is crucial to investigate another similar system beyond the two existing ones, SmB$_6$ and YbB$_{12}$. Here, we address this by reporting on a Kondo insulator, U$_3$Bi$_4$Ni$_3$. Our transport measurements reveal that a surface state emerges below 250 K and dominates transport properties below 150 K, which is well above the temperature scale of SmB$_6$ and YbB$_{12}$. At low temperatures, the surface conductivity is about one order of magnitude higher than the bulk. The robustness of the surface state indicates that it is inherently protected. The similarities and differences between U$_3$Bi$_4$Ni$_3$ and the other two Kondo insulators will provide valuable insights into the nature of metallic surface states in Kondo insulators and their interplay with strong electron correlations.

cond-mat.str-el↗

arXiv2024

Trinity: Synchronizing Verbal, Nonverbal, and Visual Channels to Support Academic Oral Presentation Delivery

Academic Oral Presentation (AOP) allows English-As-Foreign-Language (EFL) students to express ideas, engage in academic discourse, and present research findings. However, while previous efforts focus on training efficiency or speech assistance, EFL students often face the challenge of seamlessly integrating verbal, nonverbal, and visual elements into their presentations to avoid coming across as monotonous and unappealing. Based on a need-finding survey, a design study, and an expert interview, we introduce Trinity, a hybrid mobile-centric delivery support system that provides guidance for multichannel delivery on-the-fly. On the desktop side, Trinity facilitates script refinement and offers customizable delivery support based on large language models (LLMs). Based on the desktop configuration, Trinity App enables a remote mobile visual control, multi-level speech pace modulation, and integrated delivery prompts for synchronized delivery. A controlled between-subject user study suggests that Trinity effectively supports AOP delivery and is perceived as significantly more helpful than baselines, without excessive cognitive load.

cs.HC↗

arXiv2024

Hierarchical Search-Based Cooperative Motion Planning

Cooperative path planning, a crucial aspect of multi-agent systems research, serves a variety of sectors, including military, agriculture, and industry. Many existing algorithms, however, come with certain limitations, such as simplified kinematic models and inadequate support for multiple group scenarios. Focusing on the planning problem associated with a nonholonomic Ackermann model for Unmanned Ground Vehicles (UGV), we propose a leaderless, hierarchical Search-Based Cooperative Motion Planning (SCMP) method. The high-level utilizes a binary conflict search tree to minimize runtime, while the low-level fabricates kinematically feasible, collision-free paths that are shape-constrained. Our algorithm can adapt to scenarios featuring multiple groups with different shapes, outlier agents, and elaborate obstacles. We conduct algorithm comparisons, performance testing, simulation, and real-world testing, verifying the effectiveness and applicability of our algorithm. The implementation of our method will be open-sourced at https://github.com/WYCUniverStar/SCMP.

cs.RO↗