# Beyond Confidence: The Rhythms of Reasoning in Generative Models

arXiv: https://arxiv.org/abs/2602.10816 · HTML: https://mlsanity.com/paper/2602.10816/

## Key facts

- No citing papers in the map yet (data through 14 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.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 30 of 31)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?](https://mlsanity.com/paper/2202.12837/) (2022)
- [Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity](https://mlsanity.com/paper/2104.08786/) (2021)
- [Ethical and social risks of harm from Language Models](https://mlsanity.com/paper/2112.04359/) (2021)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback](https://mlsanity.com/paper/2305.14975/) (2023)
- [A Survey on In-context Learning](https://mlsanity.com/paper/2301.00234/) (2022)
- [Larger language models do in-context learning differently](https://mlsanity.com/paper/2303.03846/) (2023)
- [Surface Form Competition: Why the Highest Probability Answer Isn't Always Right](https://mlsanity.com/paper/2104.08315/) (2021)
- [The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence](https://mlsanity.com/paper/2002.06177/) (2020)
- [Sensitivity and Generalization in Neural Networks: an Empirical Study](https://mlsanity.com/paper/1802.08760/) (2018)
- [Holistic Evaluation of Language Models](https://mlsanity.com/paper/2211.09110/) (2022)
- [Robust Learning with Jacobian Regularization](https://mlsanity.com/paper/1908.02729/) (2019)
- [Uncertainty Quantification and Confidence Calibration in Large Language Models: A Survey](https://mlsanity.com/paper/2503.15850/) (2025)
- [Calibrating the Confidence of Large Language Models by Eliciting Fidelity](https://mlsanity.com/paper/2404.02655/) (2024)
- [A Survey on Dataset Distillation: Approaches, Applications and Future Directions](https://mlsanity.com/paper/2305.01975/) (2023)
- [Estimating LLM Uncertainty with Evidence](https://mlsanity.com/paper/2502.00290/) (2025)
- [Large language models for artificial general intelligence (AGI): A survey of foundational principles and approaches](https://mlsanity.com/paper/2501.03151/) (2025)
- [Enhancing LLM Robustness to Perturbed Instructions: An Empirical Study](https://mlsanity.com/paper/2504.02733/) (2025)
- [An overview of model uncertainty and variability in LLM-based sentiment analysis. Challenges, mitigation strategies and the role of explainability](https://mlsanity.com/paper/2504.04462/) (2025)
- [RUPBench: Benchmarking Reasoning Under Perturbations for Robustness Evaluation in Large Language Models](https://mlsanity.com/paper/2406.11020/) (2024)
- [Forget What You Know about LLMs Evaluations -- LLMs are Like a Chameleon](https://mlsanity.com/paper/2502.07445/) (2025)
- [Explainable artificial intelligence (XAI): from inherent explainability to large language models](https://mlsanity.com/paper/2501.09967/) (2025)
- [Robustness of Large Language Models to Perturbations in Text](https://mlsanity.com/paper/2407.08989/) (2024)
- [Risk and Response in Large Language Models: Evaluating Key Threat Categories](https://mlsanity.com/paper/2403.14988/) (2024)
- [When an LLM is apprehensive about its answers -- and when its uncertainty is justified](https://mlsanity.com/paper/2503.01688/) (2025)

## 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.