NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of ICLR 2027
201 results for “keyword:"self improve"” in ICLR 2027. 52 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Recursive Hypercompression: Phase Transitions from Self-Improvement to Self-Deterioration▼5.8k2.33
[PDF]Freeze the Student, Train the Researcher: Learning the Value of Experiments for Streaming Video Understanding▼5.9k2.50
[PDF]Learning to Repair the Learner: Evaluating Recursive Improvement in Neuro-Symbolic World Models▼4.5k2.67
[PDF]Table-Zero: Self-Improvement for Complex Table Reasoning without Additional Human Supervision▼5.8k2.33
[PDF]Learning to Verify by Doing: A Self-Renewing Policy–Verifier Loop for Sustained Self-Improvement▼1.2k1.67
[PDF]StateWise: Teaching Language Models to Track and Obey Evolving Constraints in Multi-Turn Dialogue▼1.2k1.67
[PDF]Beyond Skill Evolution: Self-Evolving Context Management Policies for Long-Horizon Agent Harnesses▼4.3k2.00
[PDF]Experience Is Not Knowledge: Where the Decision Knowledge in an Agent's Memory Comes From▼1361.33
[PDF]Continual-Skill: Utility Optimization-Driven Co-Evolution of Policies and Skill Libraries▼4.3k2.00
[PDF]Frontier-RSI: Advancing the Agent Capability Frontier through Continual Training and Environment Evolution▼5.9k2.50
[PDF]Gödel Ecology: Productivity-Guided Continuous Co-Evolution of Agents and Evaluators for Collective Recursive Self-Improvement▼4.3k2.00
Withdrawn or desk rejected (52)
[PDF]How Much Evidence Should a Self-Correction Carry? Adaptive Dirichlet Evidence for Self-DistillationWithdrawn1.67
[PDF]Video-RSI: Recursive Self-Improvement of Video Understanding Agents via Harness EvolutionDesk rejected
[PDF]Self-Improving Robot Skills for Long-Horizon Tasks via Recursive State-Aware Credit AssignmentWithdrawn1.33
[PDF]VideoForge: Toward Recursive Self-Improvement Environments for Agentic Video EvaluationWithdrawn2.00