ARM-Thinker: Reinforcing Multimodal Generative Reward Models with Agentic Tool Use and Visual Reasoning Paper • 2512.05111 • Published 3 days ago • 42
MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing Paper • 2509.22186 • Published Sep 26 • 136
CMPhysBench: A Benchmark for Evaluating Large Language Models in Condensed Matter Physics Paper • 2508.18124 • Published Aug 25 • 48
Intern-S1: A Scientific Multimodal Foundation Model Paper • 2508.15763 • Published Aug 21 • 256
Mol-R1: Towards Explicit Long-CoT Reasoning in Molecule Discovery Paper • 2508.08401 • Published Aug 11 • 42
The Imitation Game: Turing Machine Imitator is Length Generalizable Reasoner Paper • 2507.13332 • Published Jul 17 • 48
VIKI-R: Coordinating Embodied Multi-Agent Cooperation via Reinforcement Learning Paper • 2506.09049 • Published Jun 10 • 37
NovelSeek: When Agent Becomes the Scientist -- Building Closed-Loop System from Hypothesis to Verification Paper • 2505.16938 • Published May 22 • 120
MIG Datasets Collection Processed data pools and sampled datasets released with MIG. • 13 items • Updated Apr 27 • 2