LLM Self-Recognition: Steering and Retrieving Activation Signatures

• URL: http://arxiv.org/abs/2606.06315v1
• Date: Thu, 04 Jun 2026 15:54:34 GMT
• Title: LLM Self-Recognition: Steering and Retrieving Activation Signatures
• Abstract summary: Recent advances in interpretability suggest that large language models implicitly encode signals that enable self-recognition of their outputs.<n>We demonstrate that this capability is reliable, even in low-entropy scenarios, and that it can be amplified through targeted intervention.<n>This approach offers a practical alternative to traditional detectors by leveraging the model's natural representation structure for attribution rather than embedding a signal externally.
• Score: 2.992414059774663
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in interpretability suggest that large language models (LLMs) implicitly encode signals in their generated text that enable self-recognition of their outputs. We demonstrate that this capability is reliable, even in low-entropy scenarios, and that it can be amplified through targeted intervention. By steering the internal residual stream during generation with a random sparse vector, we create a detectable fingerprint that enables attribution of a given text to a specific LLM. This signal is recoverable from the activations of an LLM used as a detector, achieving over 98% accuracy across multiple detection settings while preserving the quality of generated text. As AI-generated content proliferates, this approach offers a practical alternative to traditional detectors by leveraging the model's natural representation structure for attribution rather than embedding a signal externally. Our contributions include: (i) establishing reliable self-recognition capabilities in LLMs, (ii) a simple steering mechanism enabling multi-LLM identification with no quality degradation, (iii) demonstrating that activation spaces contain exploitable structure for encoding signals without semantic interference.

Related papers

• LLM Detection as an Intervention: Downstream Impact under Strategic User Behavior [53.63441578662801]
  Detectors operate as an intervention that steers not only the detected attribute itself, but also downstream metrics such as LLM usage and output quality.<n>We develop a stylized model which captures how users strategically choose how much to use the LLM and how to post-process content to reduce the detected attribute.<n>We find that even when reducing the detected attribute improves output quality, introducing an LLM detector can lead users to produce lower quality outputs.
  arXiv  Detail & Related papers  (2026-07-21T17:11:02Z)
• CASS-RTL: Correctness-Aware Subspace Steering for RTL Generation with LLMs [0.0]
  This work proposes CASS-RTL, a first-of-its-kind framework for discovering and leveraging LLMs' correctness-aware components.<n>We observe 10%-20% improvement in pass@1/5/10 accuracy on VerilogEval and 5% improvement on CVDP.
  arXiv  Detail & Related papers  (2026-06-04T04:02:51Z)
• Projectional Decoding: Towards Semantic-Aware LLM Generation [7.143305283235514]
  Large language models (LLMs) are increasingly used to generate software artifacts across many software engineering (SE) tasks.<n>Existing constrained decoding techniques can enforce syntactic correctness and, in some cases, specific semantic rules.<n>We propose projectional decoding, a novel conceptual framework that integrates domain semantics directly into the generation process.
  arXiv  Detail & Related papers  (2026-05-28T15:05:53Z)
• On the Ability of LLMs to Handle Character-Level Perturbations: How Well and How? [22.185338324021117]
  This work investigates the resilience of contemporary LLMs against frequent and structured character-level perturbations.<n>We introduce UCC-Inj, a practical method that inserts invisible Unicode control characters into text to discourage LLM misuse.<n>Despite strong obfuscation that fragments tokenization and reduces the signal-to-noise ratio significantly, many LLMs still maintain notable performance.
  arXiv  Detail & Related papers  (2025-10-16T06:59:58Z)
• Self-Disguise Attack: Induce the LLM to disguise itself for AIGT detection evasion [16.94434185181644]
  Self-Disguise Attack (SDA) is a novel approach that enables Large Language Models to actively disguise its output.<n>We show that the SDA effectively reduces the average detection accuracy of various AIGT detectors across texts generated by three different LLMs.
  arXiv  Detail & Related papers  (2025-08-20T04:17:03Z)
• RepreGuard: Detecting LLM-Generated Text by Revealing Hidden Representation Patterns [50.401907401444404]
  Large language models (LLMs) are crucial for preventing misuse and building trustworthy AI systems.<n>We propose RepreGuard, an efficient statistics-based detection method.<n> Experimental results show that RepreGuard outperforms all baselines with average 94.92% AUROC on both in-distribution (ID) and OOD scenarios.
  arXiv  Detail & Related papers  (2025-08-18T17:59:15Z)
• Your Language Model Can Secretly Write Like Humans: Contrastive Paraphrase Attacks on LLM-Generated Text Detectors [77.82885394684202]
  We propose textbfContrastive textbfParaphrase textbfAttack (CoPA), a training-free method that effectively deceives text detectors.<n>CoPA constructs an auxiliary machine-like word distribution as a contrast to the human-like distribution generated by large language models.<n>Our theoretical analysis suggests the superiority of the proposed attack.
  arXiv  Detail & Related papers  (2025-05-21T10:08:39Z)
• Steered Generation via Gradient Descent on Sparse Features [1.534667887016089]
  We modify the internal structure of large language models (LLMs) by training sparse autoencoders to learn a sparse representation of the query embedding.<n>We demonstrate that manipulating this sparse representation effectively transforms the output toward different stylistic and cognitive targets.
  arXiv  Detail & Related papers  (2025-02-25T21:06:14Z)
• GaussMark: A Practical Approach for Structural Watermarking of Language Models [61.84270985214254]
  GaussMark is a simple, efficient, and relatively robust scheme for watermarking large language models.<n>We show that GaussMark is reliable, efficient, and relatively robust to corruptions such as insertions, deletions, substitutions, and roundtrip translations.
  arXiv  Detail & Related papers  (2025-01-17T22:30:08Z)
• Exploring Automatic Cryptographic API Misuse Detection in the Era of LLMs [60.32717556756674]
  This paper introduces a systematic evaluation framework to assess Large Language Models in detecting cryptographic misuses. Our in-depth analysis of 11,940 LLM-generated reports highlights that the inherent instabilities in LLMs can lead to over half of the reports being false positives. The optimized approach achieves a remarkable detection rate of nearly 90%, surpassing traditional methods and uncovering previously unknown misuses in established benchmarks.
  arXiv  Detail & Related papers  (2024-07-23T15:31:26Z)
• DALD: Improving Logits-based Detector without Logits from Black-box LLMs [56.234109491884126]
  Large Language Models (LLMs) have revolutionized text generation, producing outputs that closely mimic human writing. We present Distribution-Aligned LLMs Detection (DALD), an innovative framework that redefines the state-of-the-art performance in black-box text detection. DALD is designed to align the surrogate model's distribution with that of unknown target LLMs, ensuring enhanced detection capability and resilience against rapid model iterations.
  arXiv  Detail & Related papers  (2024-06-07T19:38:05Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.