5 papers
Calibrating Teacher--Student Discrepancy for On-Policy Distillation
Qiangqiang He, Jin Li, MingCai Chen
On-policy distillation (OPD) improves reasoning models by learning the token-level discrepancy between a stronger teacher and an on-policy student. However, this discrepancy does n…
Test-time Reinforcement Learning for Anomalous Video Understanding
Huining Li, Yuxiang Duan, Jiyang Tan +5
Anomalous video understanding aims to identify abnormal events in videos and interpret their semantic meanings beyond simple anomaly detection. Recent video large language models (…
VITAL-RAG: Invariance Race for Context Allocation in Coding Agents
Zijian Lu, Yonghua Lu, Mingcai Chen +4
Coding agents often retrieve code from an entire repository, but only limited evidence can fit into the final model input. Conventional retrieval-augmented generation (RAG) for cod…
Prior Directions: Separating Geometric Identification from Behavioral Efficacy in Visual Revision
Weile Gong, Zijian Lu, Mingcai Chen +3
Vision-language models can remain anchored to stale verbal priors even when current visual evidence supports a different answer. We use this controlled revision failure to ask whet…
Geometric Risk Control for Vision-Language Model OCR
Weile Gong, Zijian Lu, Mingcai Chen +3
Vision-language models (VLMs) enable flexible generative optical character recognition (OCR), while their open-ended decoders can expose wrong but fluent text with weak visual supp…