Towards Mechanistically Understanding Why Memorized Knowledge Fails to Generalize in Large Language Model Finetuning Paper • 2607.08393 • Published 27 days ago • 18
Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents Paper • 2606.06036 • Published Jun 4 • 77
OmniDirector: General Multi-Shot Camera Cloning without Cross-Paired Data Paper • 2606.13432 • Published Jun 11 • 113
Imaginative Perception Tokens Enhance Spatial Reasoning in Multimodal Language Models Paper • 2606.03988 • Published Jun 3 • 126
Mitigating Perceptual Judgment Bias in Multimodal LLM-as-a-Judge via Perceptual Perturbation and Reward Modeling Paper • 2606.02578 • Published Jun 1 • 6
ResearchMath-14K: Scaling Research-Level Mathematics via Agents Paper • 2605.28003 • Published May 27 • 50
Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players Paper • 2605.28816 • Published May 27 • 433