6 papers
DRILL: Training World Models to Improve Policies, Not Predict Pixels
Zhaolu Kang, Tailong Luo, Lei Wei +9
World models are increasingly used to reduce environment interaction when post-training vision-language-action (VLA) policies with reinforcement learning. Yet they are typically tr…
PAR: Progress-Aligned Recurrence for Length Generalization in State Tracking
Pan Wang, Yihao Hu, Guangyuan Dong +8
Looped language models can run more iterations at test time, but additional iterations need not help them solve longer problems. For sequential state tracking, final-answer supervi…
Conditional Rank Allocation for Taxonomy-Aware Medical Language Model Adaptation
Guangyuan Dong, Ziwei Hong, Xuehao Zhou +6
Medical question answering spans specialties and clinical operations that may benefit from different adaptation directions. We propose ARBOR, a parameter-efficient method that sele…
Behavior Pack Optimization for Video MLLM Post-Training
Zhaolu Kang, Shiyu Liu, Tailong Luo +12
Video multimodal large language models (MLLMs) keep climbing video question answering benchmarks, yet shuffling the frames, masking the segment that supports the answer, or occludi…
When Images Look Right and Retrieve Wrong: Coverage-Guided Cross-Scale Re-Indexing for Knowledge-Faithful Generative Perception
Guangyuan Dong, Chuang Liu, Haoyu Wang +10
Multimodal information systems increasingly route generated visual content back through the same vision-language index that informed its production, so the output must remain retri…
COMET: Contrastive Motion-Enhanced Temporal Reasoning for Video Multimodal Large Language Models
Chenghua Zhu, Zhaolu Kang, Qifan Shi +8
Video multimodal large language models have advanced significantly, yet fine-grained motion-temporal understanding remains fragile. The core bottleneck is not only sparse frame sam…