AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

One Probe Won’t Catch Them All: Towards Targeted Deception Detection

Vikram Natarajan, Devina Jain, Shivam Arora, Satvik Golechha, Joseph Isaac Bloom

ICML 2026 · Large language models › Language, safety and interpretability · approach: Data and hallucination detection · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use a single validation-selected taxonomy/framing prompt instead of the default promptinstruction-pair probe · Mean over 10 deception datasetsAUC: 0.689 -> 0.721Improved
match the taxonomy prompt per dataset (oracle) instead of the default promptinstruction-pair probe · Mean over 10 deception datasetsAUC: 0.689 -> 0.797Improved
use pressure-dishonesty promptinstruction-pair probe · MASK known factsAUC: 0.374 -> 0.697Improved
use best taxonomy promptinstruction-pair probe · MASK doubling down known factsAUC: 0.565 -> 0.727Improved
use concealment-focused promptinstruction-pair probe · SandbaggingAUC: 0.518 -> 0.657Improved
use information-fabrication taxonomy promptinstruction-pair probe · MASK disinformationAUC: 0.842 -> 0.901Improved
use best taxonomy promptinstruction-pair probe · Liar's Bench Convincing GameAUC: 0.984 -> 0.985No clear effect
vary system prompt in probe training (ablation)instruction-pair probe · ANOVA of probe AUC variance (Gemma-2-9B-IT)variance explained: 70.6% of varianceImproved
vary layer ID (20, 25, 31) (ablation)instruction-pair probe · ANOVA of probe AUC variance (Gemma-2-9B-IT)variance explained: 2.7% of varianceNo clear effect
vary training fact dataset (ablation)instruction-pair probe · ANOVA of probe AUC variance (Gemma-2-9B-IT)variance explained: 2.5% of varianceNo clear effect
vary token selection strategy (ablation)instruction-pair probe · ANOVA of probe AUC variance (Gemma-2-9B-IT)variance explained: 0.6% of varianceNo clear effect
vary system prompt in probe training (ablation)instruction-pair probe · ANOVA of probe AUC variance (Llama-3.3-70B-Instruct)variance explained: 71.0% of varianceImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy