# Internal Flow Signatures for Self-Checking and Refinement in LLMs

arXiv: https://arxiv.org/abs/2602.01897 · HTML: https://mlsanity.com/paper/2602.01897/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 42 references that are in the map.
- First posted on arXiv 2 Feb 2026.

## Builds on (most cited first, 28 of 28)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [WebGPT: Browser-assisted question-answering with human feedback](https://mlsanity.com/paper/2112.09332/) (2021)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [BERT Rediscovers the Classical NLP Pipeline](https://mlsanity.com/paper/1905.05950/) (2019)
- [On Layer Normalization in the Transformer Architecture](https://mlsanity.com/paper/2002.04745/) (2020)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [On Faithfulness and Factuality in Abstractive Summarization](https://mlsanity.com/paper/2005.00661/) (2020)
- [Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation](https://mlsanity.com/paper/2302.09664/) (2023)
- [SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models](https://mlsanity.com/paper/2303.08896/) (2023)
- [How Contextual are Contextualized Word Representations? Comparing the Geometry of BERT, ELMo, and GPT-2 Embeddings](https://mlsanity.com/paper/1909.00512/) (2019)
- [The Internal State of an LLM Knows When It's Lying](https://mlsanity.com/paper/2304.13734/) (2023)
- [Eliciting Latent Predictions from Transformers with the Tuned Lens](https://mlsanity.com/paper/2303.08112/) (2023)
- [Active Retrieval Augmented Generation](https://mlsanity.com/paper/2305.06983/) (2023)
- [INSIDE: LLMs' Internal States Retain the Power of Hallucination Detection](https://mlsanity.com/paper/2402.03744/) (2024)
- [RARR: Researching and Revising What Language Models Say, Using Language Models](https://mlsanity.com/paper/2210.08726/) (2022)
- [Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs](https://mlsanity.com/paper/2406.15927/) (2024)
- [A Survey of Confidence Estimation and Calibration in Large Language Models](https://mlsanity.com/paper/2311.08298/) (2023)
- [Future Lens: Anticipating Subsequent Tokens from a Single Hidden State](https://mlsanity.com/paper/2311.04897/) (2023)
- [LLM Internal States Reveal Hallucination Risk Faced With a Query](https://mlsanity.com/paper/2407.03282/) (2024)
- [Adaptive Activation Steering: A Tuning-Free LLM Truthfulness Improvement Method for Diverse Hallucinations Categories](https://mlsanity.com/paper/2406.00034/) (2024)
- [Explaining How Transformers Use Context to Build Predictions](https://mlsanity.com/paper/2305.12535/) (2023)

## Abstract (arXiv)

Large language models can generate fluent answers that are unfaithful to the provided context, while many safeguards rely on external verification or a separate judge after generation. We introduce internal flow signatures that audit decision formation from depthwise dynamics at a fixed inter-block monitoring boundary. The method stabilizes token-wise motion via bias-centered monitoring, then summarizes trajectories in compact moving readout-aligned subspaces constructed from the top token and its close competitors within each depth window. Neighboring window frames are aligned by an orthogonal transport, yielding depth-comparable transported step lengths, turning angles, and subspace drift summaries that are invariant to within-window basis choices. A lightweight GRU validator trained on these signatures performs self-checking without modifying the base model. Beyond detection, the validator localizes a culprit depth event and enables a targeted refinement: the model rolls back to the culprit token and clamps an abnormal transported step at the identified block while preserving the orthogonal residual. The resulting pipeline provides actionable localization and low-overhead self-checking from internal decision dynamics. Code is available at textttgithub.com/EavnJeong/Internal-Flow-Signatures-for-Self-Checking-and-Refinement-in-LLMs.