MG-Thinker is presented, a post-training RL framework that advances a new MRG paradigm featuring such hierarchical reasoning, supported by a curated 25K MRG dataset with task-adaptive Chain-of-Thought annotations that elicit multi-perspective evidence before conclusion.
Heyuan Huang, Chi Chen, Zong-Hao Guo et al.· 0 citations
MM-OPD is introduced, a multimodal on-policy self-distillation framework for symbolic-to-visual correction that transfers guidance from symbolic-conditioned behavior to the image-conditioned policy through residual token-level targets, steering the model toward correct visual evidence.
STD is proposed, a hierarchical token pruning framework that adapts token selection mechanisms to the functional role of each network stage, and introduces a Stability-Adaptive Trigger in deep layers to execute pruning only during semantically stable phases.
Shuo Zhang, Jin-Tao Tong, Yi-Xiong Zou et al.· Proceedings of the 2026 Conf...· 0 citations
A compositional-learning-based method is proposed to learn two primitive sets (a common set and a discriminative set), which alleviates the regional shortcut by constraining the model to learn and utilize the common primitive set for base- and novel-class recognition.
Haichen Zhou, Y. Lyu, Yi-Xiong Zou et al.· IEEE transactions on multime...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.