Open multimodal reasoning models have benefited from large-scale reasoning supervision, yet reliable post-training remains challenging due to uneven data quality, inefficient supervision construction, imbalanced difficulty, and cross-domain interference. We introduce MMVistaReason (MVR), an open-data post-training reci...
Jue-Kai Lin, Hong-Lin Lin, Yu-Qian Yuan et al.· 0 citations
World modeling under partial observation requires reasoning about the complete worlds that remain compatible with limited visual evidence. Occluded objects and unseen regions can leave several world states possible; additional views can exclude alternatives and strengthen the conclusions supported by the observations....
Wei Wang, Wen-Qiao Zhang, Yu-Tong Lin et al.· 0 citations
Experiments on WebShop and ALFWorld with different backbones show that TRIAL outperforms GRPO across all eight combinations of backbone, environment, and evaluation metric, while achieving the best or tied-best performance among six methods on six of them.
Hao-Yu Zheng, Yun Zhu, Qing Wang et al.· 1 citation
A large-scale robotic manipulation dataset and benchmark for diagnosing embodied reasoning in VLA models, and establishes RoboSPA as a challenging diagnostic benchmark for developing more capable, reliable, and generalizable embodied agents.
Zhen-Xuan Fan, Bo Zhang, Yu-Tong Lin et al.· 0 citations
This work illuminates the limitations of current MLLMs in electrophysiology and provides a foundational benchmark for long-term medical MLLMs, including Holtercare-Bench, a multimodal benchmark that evaluates models on temporal localization, clinical diagnosis, and global summarization.
EmbodiedSkills is a unified framework that treats each skill decision as an execution proposal: the runtime checks its prerequisites before execution and verifies the outcome afterward and provides a trainable and inspectable agent layer for turning low-level VLA policies into closed-loop embodied systems.
Wei Wang, Wen-Qiao Zhang, Yu-Tong Lin et al.· 3 citations
Results show that self-contamination is a trainable component of the LiC gap, and propose MAIGO, an on-policy self-distillation method that reduces this contamination using history-cleaned references from the model's own policy.
A novel Dual Self-Consistency Reinforcement Learning optimization paradigm is introduced, which utilizes Round-Trip Verification to penalize degenerate code and boost overall self-consistency in TikZ code.
Juekai Lin, Yun Zhu, Honglin Lin et al.· arXiv.org· 5 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
