Houqiang Li

Professor@University of Science and Technology of China·中国·OpenReview

研究方向

sign language recognition; video captioning · image segmentation; object detection; video tracking;image retrieval · image and video intelligent coding;

Houqiang Li

Exploring Mode Connectivity in Krylov Subspace for Domain Generalization

VisuLogic: A Benchmark for Evaluating Visual Reasoning in Multi-modal Large Language Models

Primary-Fine Decoupling for Action Generation in Robotic Imitation

EchoGen: Generating Visual Echoes in Any Scene via Feed-Forward Subject-Driven Auto-Regressive Model

Flatness Guided Test-Time Adaptation for Vision-Language Models

Disentangling Length Bias in Preference Learning via Response-Conditioned Modeling

T2G-Reasoner: Deep Reasoning for Text-to-Gloss Translation

Envision the Future in Open-World Dynamic Tasks by a Hierarchical World Model with Residual Enhanced Foresight

Integer-Centric Neural Video Compression

ScaleWeaver: Weaving Efficient Controllable T2I Generation with Multi-Scale Reference Attention

Toward Experiment-Guided Hypothesis Ranking via Simulated Experimental Feedback

Optimizing Distributional Geometry Alignment with Optimal Transport for Generative Dataset Distillation

Enhancing the Outcome Reward-based RL Training of MLLMs with Self-Consistency Sampling

Uni-Sign: Toward Unified Sign Language Understanding at Scale

Robust Multimodal Large Language Models Against Modality Conflict

EG4D: Explicit Generation of 4D Object without Score Distillation

Mitigating Hallucination in VideoLLMs via Temporal-Aware Activation Engineering

Deep Video Discovery: Agentic Search with Tool Use for Long-form Video Understanding

MotionRL: Align Text-to-Motion Generation to Human Preferences with Multi-Reward Reinforcement Learning

Super Robot View Transformer

Unveiling the Flow of Input-label Mappings for In-context Learning