Demystifying Agent Skills: Why They Work-Until They Don't
Paper • 2608.14036 • Published • 167
August/September_2026
-
-
Modular Cognitive Architecture Emerges in Large Language Models
Paper • 2608.13567 • Published • 17 -
Cognitive Convergence: Deep Similarities Between Large Language Models and Human Cognition
Paper • 2607.26179 • Published • 1 -
Where Animacy Lives in Large Language Models: Tracing the Circuits of the Animacy Concept
Paper • 2607.20995 • Published • 1 -
Learn What's Left, Not What's Mastered: Saturation Aware Advantage Reweighting for Multi-Reward Policy Optimization
Paper • 2608.16072 • Published • 148 -
WikiSkill: Compiling Agent Experience into Persistent Knowledge for Skill Evolution
Paper • 2608.27454 • Published • 37 -
Fast Weight Attention for Continual Learning
Paper • 2608.27763 • Published • 31 -
Drift-Constrained Optimization: Only Direction Matters in Fine-Tuning Instruct Models
Paper • 2609.13680 • Published • 13 -
ImpossibleRubrics: Stress-Testing Generated Rubrics as Reward Signals
Paper • 2609.16816 • Published • 11