PPL
Bhavana Dalvi Mishra
Papers (5)
Converted, Not Equivalent: Benchmarking Codebase Conversion via Observational EquivalenceScientistOne: Towards Human-Level Autonomous Research via Chain-of-EvidenceLiSA: Lifelong Safety Adaptation via Conservative Policy InductionRubricEM: Meta-RL with Rubric-guided Policy Decomposition beyond Verifiable RewardsSkillOS: Learning Skill Curation for Self-Evolving Agents