6 papers
PMOPD: Task Ordering, Cycling, and Parameter-Update Subspace Protection in Multi-Teacher On-Policy Distillation
Youzhi Liu, Ruobing Zheng, Boyuan Tong +5
Multi-teacher on-policy distillation (MOPD) has emerged as a popular post-training paradigm for integrating specialized capabilities in frontier language models. Existing OPD resea…
Beyond Retrieval: Progressive Latent Memory Evolution for Streaming Video Understanding
Hongyu Qu, Guangming Yao, Ling Xing +7
Streaming video understanding requires multimodal large language models (MLLMs) to process continuous visual inputs and respond to user queries under strict causality and bounded m…
VideoGAIA: A Benchmark for General AI Assistants on Agentic Video Understanding
Fan Zhang, Guangming Yao, Jinyang Wu +6
Video understanding is a fundamental task for evaluating the capabilities of multimodal large language models (MLLMs). However, existing leading models have already achieved approx…
MemDreamer: Decoupling Perception and Reasoning for Long Video Understanding via Hierarchical Graph Memory and Agentic Retrieval Mechanism
Cong Chen, Guo Gan, Kaixiang Ji +7
Current Vision-Language Models struggle with hours-long videos because processing full-length visual sequences induces prohibitive token explosion and attention dilution. To overco…
SPRI: SVD-Partitioned Residual Initialization for Data-Constrained MoE Upcycling
Weiqiao Shan, Ruixiang Mao, Yuang Li +10
Mixture-of-Experts (MoE) models enable efficient scaling, but training them from scratch remains prohibitively expensive. MoE upcycling mitigates this cost by converting pretrained…
Dual Tuning for Reasoning Efficacy-Driven Data Curation in Multimodal LLM Training
Ruobing Zheng, Tianqi Li, Jianing Li +3
Reasoning post-training improves Large Language Models (LLMs) on complex tasks such as mathematics and coding, but its benefits across diverse multimodal tasks remains uncertain. T…