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
30 results for “keyword:"self correction"” in ICLR 2027. 21 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Escaping Answer Anchoring: Fixing Shortcut Tokens in Reasoning Models via Context- Replay Gaps▼4.3k2.00
[PDF]From Self-Correction to Self-Evolution: Agentic Test-Time Learning for Mathematical Reasoning▼4.3k2.00
[PDF]Which Revisions Are Worth Applying? Risk-Aware Repair Decisions for Translation Self-Correction▲4012.67
[PDF]Inject a Turn, Not a Token: Learning Where to Redirect LLM Reasoning for Self-Correction▲4012.67
[PDF]DACPO: Drift-Adaptive Constrained Policy Optimization for Cue-Free Safety Recovery in Large Reasoning Models▲4012.67
[PDF]Rollback the World, Keep the Reflection: Rollback-Induced Reflection for Long-Horizon LLM Agents▲13k2.33
Withdrawn or desk rejected (21)
[PDF]Verifier-Native Control and Trajectory Evaluation for Operations Research Repair AgentsWithdrawn2.00
[PDF]Foreign Exchange and Proof Diversity: Which Failures to Borrow and Which Retries to BuyDesk rejected
[PDF]Read Together, Write Apart: Decoupling Rollout Access from Update Ownership in Multi-Role LLM TrainingDesk rejected2.00
[PDF]When Revision Backfires: Consequence-Aware Verification for Multi-Role LLM ReasoningDesk rejected2.67
[PDF]Hypothesis Anchoring: Evidence Use Without Belief Revision on Self-Generated HypothesesDesk rejected