The Detection--Extraction Gap: Models Know the Answer Before They Can Say It Paper • 2604.06613 • Published Apr 8 • 2
SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning Paper • 2602.08234 • Published Feb 9 • 76
