Yiwei Wang

ChainMPQ: Interleaved Text-Image Reasoning Chains for Mitigating Relation Hallucinations

Video-STAR: Reinforcing Open-Vocabulary Action Recognition with Tools

Fastcar: Cache Attentive Replay for Fast Auto-Regressive Video Generation on the Edge

ContextNav: Towards Agentic Multimodal In-Context Learning

Unveiling the Potential of Diffusion Large Language Model in Controllable Generation

Parameters vs. Context: Fine-Grained Control of Knowledge Reliance in Language Models

Efficiently Disentangling CLIP for Multi-Object Perception

RefineX: Learning to Refine Pre-training Data at Scale from Expert-Guided Programs

WavefrontDiffusion: Dynamic Decoding Schedule for Improved Reasoning

STDR: Spatio-Temporal Decoupling for Real-Time Dynamic Scene Rendering

Mitigating Coordinate Prediction Bias from Positional Encoding Failures

Are LLMs Really Not Knowledgeable? Mining the Submerged Knowledge in LLMs' Memory

Visual CoT Makes VLMs Smarter but More Fragile

FrameMind: Frame-Interleaved Chain-of-Thought for Video Reasoning via Reinforcement Learning

$A^2R^2$: Advancing Img2LaTeX Conversion via Visual Reasoning with Attention-Guided Refinement

Focusing by Contrastive Attention: Enhancing VLMs' Visual Reasoning

a1: Steep Test-time Scaling Law via Environment Augmented Generation

Blockwise SFT for Diffusion Language Models: Reconciling Bidirectional Attention and Autoregressive Decoding

Thinking with Sound: Audio Chain-of-Thought Enables Multimodal Reasoning for Large Audio-Language Models

Beyond the Shot: Rethinking Cinematography Understanding with Foundational Skill Evaluation

SQLAgent: Learning to Explore Before Generating as a Data Engineer

Dynamic Infilling Anchors for Format-Constrained Generation in Diffusion LLMs

Depth-Breadth Synergy in RLVR: Unlocking LLM Reasoning Gains with Adaptive Exploration

When Symbols Speak: Understanding Logo Triggered Texts in Vision-Language Models

Innate Reasoning is Not Enough: In-Context Learning Enhances Reasoning Large Language Models with Less Overthinking

Lost in Time: Systematic Temporal Bias in Large Audio Language Models

Critique to Verify: Accurate and Honest Test-Time Scaling with RL-Trained Verifiers

Detecting and Mitigating Insertion Hallucination in Video-to-Audio Generation

TreeRPO: Tree Relative Policy Optimization

Generalist Scanner Meets Specialist Locator: A Synergistic Coarse-to-Fine Framework for Robust GUI Grounding

Do "New Snow Tablets" Contain Snow? Large Language Models Over-Rely on Names to Identify Ingredients of Chinese Drugs

Structured Attention Matters to Multimodal LLMs in Document Understanding

DraftAttention: Fast Video Diffusion via Low-Resolution Attention Guidance

Symbolic or Numerical? Understanding Physics Problem Solving in Reasoning LLMs

OptiBench Meets ReSocratic: Measure and Improve LLMs for Optimization Modeling

Is Factuality Enhancement a Free Lunch For LLMs? Better Factuality Can Lead to Worse Context-Faithfulness

How does Watermarking Affect Visual Language Models in Document Understanding?

Peripheral Memory for LLMs: Integration of Sequential Memory Banks with Adaptive Querying

Texture or Semantics? Vision-Language Models Get Lost in Font Recognition

MQuAKE-Remastered: Multi-Hop Knowledge Editing Can Only Be Advanced with Reliable Evaluations

Text Speaks Louder than Vision: ASCII Art Reveals Textual Biases in Vision-Language Models

Towards Robust Heterogeneous Graph Explanations under Structural Perturbations

HiddenGuard: Fine-Grained Safe Generation with Specialized Representation Router

UniTST: Effectively Modeling Inter-Series and Intra-Series Dependencies for Multivariate Time Series Forecasting

OIDA-QA: A Multimodal Benchmark for Analyzing the Opioid Industry Document Archive

Decoding by Contrasting Knowledge: Enhancing LLMs' Confidence on Edited Facts

Robust Heterogeneous Graph Neural Network Explainer with Graph Information Bottleneck