Research map / Large language models / Language, safety and interpretability
How Can You Tell if Your Large Language Model Could Be a Closet Antisemite? An Explainability-Based Audit Framework for Implicit Bias
Arka Dutta, Reza Fayyazi, Shanchieh Yang, Ashiqur R. KhudaBukhsh
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use probes from DSBIC instead of DTRH with a different identity-group set (ablation) | SHAP-based bias audit · KL-Bias ranking of 5 open LLMs | KL-Bias score ordering: PCC 0.8996; model ordering unchanged | No clear effect |
| upgrade model version within family (Gemma to Gemma2) | Gemma · DTRH probe bias audit | Islamophobic and antisemitic aggregate bias score | Hurt |
| paraphrase the audit prompt with GPT-4o (four variants) (ablation) | SHAP-based bias audit · 10 random probes | pairwise KL-divergence of aggregate bias scores | No clear effect |
Closest papers by meaning
- Investigating and Mitigating Undesirable Biases in Large Language Models (AAAI 2025 · similarity 0.89)
- StereoTales: A Multilingual Framework for Open-Ended Stereotype Discovery in LLMs (NeurIPS 2026 · similarity 0.88)
- Bias and Volatility: A Statistical Framework for Evaluating Large Language Model's Stereotypes and the Associated Generation Inconsistency (NeurIPS 2024 · similarity 0.88)
- B-score: Detecting biases in large language models using response history (ICML 2025 · similarity 0.87)
- Underestimated Privacy Risks for Minority Populations in Large Language Model Unlearning (ICML 2025 · similarity 0.87)
- Examining Alignment of Large Language Models through Representative Heuristics: the case of political stereotypes (ICLR 2025 · similarity 0.87)
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.
