Decoding Hidden Deception in Reasoning LLMs: Activation Explainers for Deception Auditing
Quick Answer
This paper shows that STATEWITNESS, an activation explainer for deception auditing in reasoning LLMs, achieves a 0.916 mean AUROC across seven datasets, outperforming existing monitors by up to 25%.
Quick Take
It provides detailed insights into suspicious responses, enhancing interpretability and alignment tools for AI safety.
Key Points
- STATEWITNESS improves deception detection with 11.6% gain over the best black-box text monitor.
- The model provides query-level answers and structured reports for human inspection.
- It reduces missed deceptive examples when combined with existing monitors.
- Evaluated on two reasoning across seven deception datasets.
- Potential building block for broader interpretability and alignment tools.
Paper Resources
Source Excerpt
arXiv:2606. 17478v1 Announce Type: new Abstract: As acquire stronger reasoning capabilities, deceptive behavior becomes an increasingly serious safety concern. Existing deception monitors either score visible transcripts or derive scalar probe scores from representation vectors, leaving little inspectable evidence about why a response is suspicious. We introduce STATEWITNESS, an activation explainer for deception auditing.
A separate decoder reads a target model's hidden states, then answers natural-language queries or emits structured reports about them. We evaluate STATEWITNESS on two target reasoning LLMs across seven deception datasets. …
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →TriAgent: Divergence-Aware Committees for Cost-Efficient Financial Sentiment Analysis
TriAgent introduces a cost-efficient multi-agent system for financial sentiment analysis, combining VADER, FinBERT, and Qwen2.5. It achieves an F1 score of ~0.87 with significant savings of $9.3M/year at a 10M-user scale compared to GPT-4o-mini, while also detecting hallucinations with an AUC of 0.90.