mlsanityData to 7 Oct 2026

Beyond Confidence: The Rhythms of Reasoning in Generative Models

arXiv 2602.10816 ↗First posted 11 Feb 2026cs.CL cs.AIcited by 0

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

Large Language Models (LLMs) exhibit impressive capabilities yet suffer from sensitivity to slight input context variations, hampering reliability. Conventional metrics like accuracy and perplexity fail to assess local prediction robustness, as normalized output probabilities can obscure the underlying resilience of an LLM's internal state to perturbations. We introduce the Token Constraint Bound (δTCB), a novel metric that quantifies the maximum internal state perturbation an LLM can withstand before its dominant next-token prediction significantly changes. Intrinsically linked to output embedding space geometry, δTCB provides insights into the stability of the model's internal predictive commitment. Our experiments show δTCB correlates with effective prompt engineering and uncovers critical prediction instabilities missed by perplexity during in-context learning and text generation. δTCB offers a principled, complementary approach to analyze and potentially improve the contextual stability of LLM predictions.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 31 of its 38 references that are in the map. First posted on arXiv 11 Feb 2026.

Builds on

31 of 38 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Uncertainty Quantification and Confidence Calibration in Large Language Models: A Survey20 Mar

Other references · 30

  1. 2024The Llama 3 Herd of Modelscited by 14,192
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  3. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  4. 2020Scaling Laws for Neural Language Modelscited by 6,411
  5. 2022Training Compute-Optimal Large Language Modelscited by 2,927
  6. 2022Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?cited by 1,413
  7. 2021Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivitycited by 1,122
  8. 2021Ethical and social risks of harm from Language Modelscited by 957
  9. 2023Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedbackcited by 788
  10. 2023Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formattingcited by 778
  11. 2022A Survey on In-context Learningcited by 717
  12. 2023Larger language models do in-context learning differentlycited by 386
18 more
  1. 2021Surface Form Competition: Why the Highest Probability Answer Isn't Always Rightcited by 256
  2. 2020The Next Decade in AI: Four Steps Towards Robust Artificial Intelligencecited by 194
  3. 2018Sensitivity and Generalization in Neural Networks: an Empirical Studycited by 187
  4. 2019Robust Learning with Jacobian Regularizationcited by 127
  5. 2022Holistic Evaluation of Language Modelscited by 115
  6. 2024Calibrating the Confidence of Large Language Models by Eliciting Fidelitycited by 47
  7. 2023A Survey on Dataset Distillation: Approaches, Applications and Future Directionscited by 32
  8. 2025Estimating LLM Uncertainty with Evidencecited by 26
  9. 2025Large language models for artificial general intelligence (AGI): A survey of foundational principles and approachescited by 18
  10. 2025Enhancing LLM Robustness to Perturbed Instructions: An Empirical Studycited by 17
  11. 2025An overview of model uncertainty and variability in LLM-based sentiment analysis. Challenges, mitigation strategies and the role of explainabilitycited by 15
  12. 2024RUPBench: Benchmarking Reasoning Under Perturbations for Robustness Evaluation in Large Language Modelscited by 13
  13. 2025Explainable artificial intelligence (XAI): from inherent explainability to large language modelscited by 12
  14. 2024Robustness of Large Language Models to Perturbations in Textcited by 11
  15. 2025Forget What You Know about LLMs Evaluations -- LLMs are Like a Chameleoncited by 11
  16. 2024Risk and Response in Large Language Models: Evaluating Key Threat Categoriescited by 10
  17. 2025When an LLM is apprehensive about its answers -- and when its uncertainty is justifiedcited by 4
  18. 2025FedEAT: A Robustness Optimization Framework for Federated LLMscited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Beyond Exponential Decay: Rethinking Error Accumulation in Large Language Models
  2. 2024Calibrating Reasoning in Language Models with Internal Consistency
  3. 2026Beyond Confidence: Stability-Aware Test-Time Adaptation for LLM Reasoning
  4. 2025Breach in the Shield: Unveiling the Vulnerabilities of Large Language Models
  5. 2025Randomly Sampled Language Reasoning Problems Elucidate Limitations of In-Context Learning
  6. 2026Dissecting Failure Dynamics in Large Language Model Reasoning
  7. 2024Beyond the Black Box: A Statistical Model for LLM Reasoning and Inference
  8. 2023Self-Consistency of Large Language Models under Ambiguity