WriteSAE: Sparse Autoencoders for Recurrent State
Paper • 2605.12770 • Published
daily paper
-
-
From Pixels to Concepts: Do Segmentation Models Understand What They Segment?
Paper • 2605.09591 • Published • 2 -
F-GRPO: Factorized Group-Relative Policy Optimization for Unified Candidate Generation and Ranking
Paper • 2605.12995 • Published • 2 -
Learning Agentic Policy from Action Guidance
Paper • 2605.12004 • Published • 12 -
Retrieval is Cheap, Show Me the Code: Executable Multi-Hop Reasoning for Retrieval-Augmented Generation
Paper • 2605.12975 • Published • 9 -
HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution
Paper • 2605.09942 • Published • 15 -
MAP: A Map-then-Act Paradigm for Long-Horizon Interactive Agent Reasoning
Paper • 2605.13037 • Published • 8 -
MemReread: Enhancing Agentic Long-Context Reasoning via Memory-Guided Rereading
Paper • 2605.10268 • Published • 4