mlsanityData to 7 Oct 2026

False Fixed Points: Kantian Feedback, Stable Miscalibration, and Representational Compression in LLMs

arXiv 2510.14925 ↗First posted 16 Oct 2025cs.AI cs.CL cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

High-confidence errors in large language models are often treated as fragile failures. We study an alternative: some errors may be false fixed points, locally stable, internally coherent, and confidently wrong. This separates robustness from truth-tracking. We develop the separation through a Kantian commitment-gate framing and a minimal linear feedback model in which stability and correctness can diverge. Across three open-weight models, overconfident wrong items are not systematically more locally fragile than confidently correct items under our hidden-state sensitivity probes. Abstention-aware self-critique reduces overconfident wrong commitments by sacrificing coverage, and C3-R, a rule-based explicit feedback gate, sharpens that tradeoff rather than eliminating it. These results motivate, but do not establish, high signal-to-noise (high-SNR) inertia and representational compression as possible mechanisms for stable miscalibration.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 21 of its 41 references that are in the map. First posted on arXiv 16 Oct 2025.

Builds on

21 of 41 references.

References · 21

  1. 2024Qwen2.5 Technical Reportcited by 4,713
  2. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  3. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  4. 2024Qwen2 Technical Reportcited by 2,076
  5. 2022Survey of Hallucination in Natural Language Generationcited by 2,056
  6. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 787
  7. 2023CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquingcited by 751
  8. 2023ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debatecited by 724
  9. 2019Representation Degeneration Problem in Training Natural Language Generation Modelscited by 250
  10. 2024LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinationscited by 230
  11. 2022RankMe: Assessing the downstream performance of pretrained self-supervised representations by their rankcited by 145
  12. 2024Know Your Limits: A Survey of Abstention in Large Language Modelscited by 116
9 more
  1. 2024LLM Internal States Reveal Hallucination Risk Faced With a Querycited by 80
  2. 2025Reasoning Models Better Express Their Confidencecited by 63
  3. 2024Anisotropy Is Inherent to Self-Attention in Transformerscited by 62
  4. 2025Tracing the Representation Geometry of Language Models from Pretraining to Post-trainingcited by 33
  5. 2025The Illusion of Progress: Re-evaluating Hallucination Detection in LLMscited by 22
  6. 2025Semantic Energy: Detecting LLM Hallucination Beyond Entropycited by 18
  7. 2025Probing the Critical Point (CritPt) of AI Reasoning: a Frontier Physics Research Benchmarkcited by 16
  8. 2025How to Detect and Defeat Molecular Mirage: A Metric-Driven Benchmark for Hallucination in LLM-based Molecular Comprehensioncited by 6
  9. 2025Black-Box Hallucination Detection via Consistency Under the Uncertain Expressioncited by 4

Built on by

Not cited yet.