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
198 results for “keyword:"self improvement"” in ICLR 2027. 52 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]When Does Memory Become Capability? Agent Self-Improvement through Memory Evolution and Consolidation▲5.4k3.00
[PDF]HarnessWeaver: Continual Self-Improvement through Task-Conditioned Harness Composition and Model Routing▼4.3k2.00
[PDF]Policy-Level Recursive Self-Improvement for Embodied AI with a Criticality World Model▲3.0k2.33
[PDF]Consensus Is Not Correctness: Reliability-Aware Self-Teaching for Unsupervised Self-Distillation▲11.20
[PDF]DIVE: Unlocking Self-Improvement in Frozen Language Models Through Diversity-Driven Skill Evolution▲3952.67
[PDF]Divide and Evolve:Orchestrating Agent Harness Evolution with Diagnostic Multi-Agent System▼1.2k1.67
[PDF]FinTriad: A Multimodal, Multi-Agent System with Adaptive Memory for Tri-Horizon Portfolio Construction▼1.3k1.75
[PDF]HEIR: When the Student Inherits the Teacher's Seat Pareto-Competitive Reference Selection for On-Policy Self-Distillation of Reasoning LLMs▼5.3k1.33
[PDF]Bifrost: Steering Strategic Trajectories to Bridge Contextual Gaps for Self-Improving Agents▲9.2k2.67
[PDF]RSI-Wild: Benchmarking AI Data Researchers with In-the-Wild Agentic Trajectories for Recursive Self-Improvement▼4.3k2.00
[PDF]Gen-RSI: Recursive Self-Improvement of Physical Video Generators through World Proxies▲6.6k3.33
[PDF]CreativeBench2.0: Benchmarking Creative Method Discovery and Foresight Toward Recursive Self-Improvement▼4.5k2.67
[PDF]EffectPreconception: Guiding ML Research Agents through Effect Prediction and Observation▲3.0k2.33
Withdrawn or desk rejected (52)
[PDF]GymHarness: Recursive Self-Improvement of Agents with Recursively Verified VerifiersWithdrawn1.67