DECOR: Auditing LLM Deception via Information Manipulation Theory
For AI safety researchers, DECOR provides an interpretable method to detect and analyze how LLMs manipulate information, addressing the need for fine-grained deception auditing beyond coarse-grained black-box methods.
DECOR is a multi-agent framework for fine-grained auditing of strategic deception in LLM responses, achieving state-of-the-art performance on single-turn and multi-turn deception detection benchmarks across 15 frontier models.
Large language models can deceive by subtly manipulating truthful information -- omitting key facts, shifting focus, or obscuring meaning -- making such behavior difficult to detect. Existing black-box methods rely on coarse-grained judgments, offering limited interpretability and failing to pinpoint which facts were distorted and how. We introduce DECOR, a multi-agent framework grounded in Information Manipulation Theory for fine-grained auditing of strategic deception in LLM responses. DECOR decomposes input contexts into atomic informational units and scores each unit against the response across four dimensions of manipulation, producing interpretable manipulation profiles that are aggregated into a global deception index. We comprehensively evaluate DECOR on both single-turn and multi-turn deception detection benchmarks spanning real-world domains, and show that DECOR achieves state-of-the-art performance on both, outperforming competitive baselines. The framework generalizes across 15 frontier models, and ablation studies confirm the contribution of each key design component. Our findings demonstrate that fine-grained, theory-grounded auditing of information manipulation offers an effective and interpretable path for LLM deception detection.