mlsanityData to 7 Oct 2026

Exact Feedback Is Not Control: Evaluating Text-based Closed-Loop Revision in LLMs

arXiv 2609.28150 ↗First posted 23 Sep 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Closed-loop revision is increasingly used in large language model (LLM) applications, but failures may reflect incomplete feedback or ineffective responses to correct feedback. We introduce a fixed-budget revision protocol with deterministic verifiers that report all remaining violations across exact-length, lexical, and compositional constraints. Fixing feedback correctness and completeness isolates model-side revision behavior. Across 19 open- and closed-source models, controller-level mean final joint success ranges from 17.4% to 99.8%, with substantial cross-model gaps persisting under identical initial drafts. Controlled experiments reveal reproducible model-specific responses to exact feedback. Post-training and scale reshape these responses without consistently bringing them closer to exact correction. Across all constraint families, failed trajectories often repeat earlier outputs, and prior recurrence is associated with lower subsequent recoverability. Matched-state interventions show that removing earlier dialogue while holding the current draft and feedback fixed changes recurrence escape without reliably improving final success; effects depend on the model, task, and trigger-state composition. Exact feedback makes revision errors observable, but does not make the closed loop reliable. Code and reproduction instructions: https://github.com/kevinjiang0121-cyber/exact-feedback-code.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 29 of its 47 references that are in the map. First posted on arXiv 23 Sep 2026.

Builds on

29 of 47 references.

References · 29

  1. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  2. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  3. 2022Locating and Editing Factual Associations in GPTcited by 2,609
  4. 2024SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineeringcited by 1,628
  5. 2018Hierarchical Neural Story Generationcited by 1,182
  6. 2024ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Toolscited by 1,096
  7. 2022Inner Monologue: Embodied Reasoning through Planning with Language Modelscited by 1,057
  8. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 965
  9. 2025AlphaEvolve: A coding agent for scientific and algorithmic discoverycited by 763
  10. 2023CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquingcited by 751
  11. 2023OpenAssistant Conversations -- Democratizing Large Language Model Alignmentcited by 712
  12. 2024Training Language Models to Self-Correct via Reinforcement Learningcited by 405
17 more
  1. 2025LLMs Get Lost In Multi-Turn Conversationcited by 400
  2. 2023Towards Best Practices of Activation Patching in Language Models: Metrics and Methodscited by 328
  3. 2024When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMscited by 230
  4. 2023Is Self-Repair a Silver Bullet for Code Generation?cited by 191
  5. 2020NeuroLogic Decoding: (Un)supervised Neural Text Generation with Predicate Logic Constraintscited by 146
  6. 2024On the Self-Verification Limitations of Large Language Models on Reasoning and Planning Taskscited by 137
  7. 2023Is This the Subspace You Are Looking for? An Interpretability Illusion for Subspace Activation Patchingcited by 75
  8. 2023COLLIE: Systematic Construction of Constrained Text Generation Taskscited by 69
  9. 2025Falcon-H1: A Family of Hybrid-Head Language Models Redefining Efficiency and Performancecited by 52
  10. 2024LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraintscited by 34
  11. 2025RefineBench: Evaluating Refinement Capability of Language Models via Checklistscited by 13
  12. 2025Feedback Friction: LLMs Struggle to Fully Incorporate External Feedbackcited by 11
  13. 2026The Curse of Multiple Mediators: Hidden Interaction Effects in Activation Patchingcited by 9
  14. 2025Dynamic Weight Grafting: Localizing Finetuned Factual Knowledge in Transformerscited by 3
  15. 2026Self-Correction as Feedback Control: Error Dynamics, Stability Thresholds, and Prompt Interventions in LLMscited by 3
  16. 2026Do Language Models Converge to Themselves? Recursive Self-Refinement as Textual Relaxationcited by 3
  17. 2024Ruler: A Model-Agnostic Method to Control Generated Length for Large Language Modelscited by 1

Built on by

Not cited yet.