Research map / Large language models / Language, safety and interpretability
Bias Association Discovery Framework for Open-Ended LLM Generations
Jinhao Pan, Chahat Raj, Ziwei Zhu
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 Negative sentiment-constrained prompt instead of two-character base prompt (ablation) | BADF generation · Gender (Female), Llama3.2-3B | number of bias associations: 277 -> 524 | Improved |
| use Balanced-Valence prompt instead of two-character base prompt (ablation) | BADF generation · Gender (Female), Llama3.2-3B | number of bias associations: 277 -> 423 | Improved |
| use open-box (patchscope) generation instead of black-box (ablation) | BADF generation · Gender (Female), Llama3.2-3B | number of bias associations: 277 -> 332 | Improved |
| use open-box (patchscope) generation instead of black-box (ablation) | BADF generation · Religions, Llama3.2-3B | number of bias associations: Muslim 832 -> 318 | Hurt |
| use two-character instead of single-character base setting (ablation) | BADF generation · Gender (Female), Llama3.2-3B | number of bias associations: 169 -> 277 | Improved |
| use Qwen3-8B instead of Llama3.2-3B as the generator | BADF generation · Gender (Male), two-character base | number of bias associations: 167 -> 408 | Improved |
Closest papers by meaning
- StereoTales: A Multilingual Framework for Open-Ended Stereotype Discovery in LLMs (NeurIPS 2026 · similarity 0.89)
- Bias and Volatility: A Statistical Framework for Evaluating Large Language Model's Stereotypes and the Associated Generation Inconsistency (NeurIPS 2024 · similarity 0.89)
- SocialStigmaQA: A Benchmark to Uncover Stigma Amplification in Generative Language Models (AAAI 2024 · similarity 0.86)
- Position: Rethinking LLM Bias Probing Using Lessons from the Social Sciences (ICML 2025 · similarity 0.86)
- Investigating and Mitigating Undesirable Biases in Large Language Models (AAAI 2025 · similarity 0.86)
- Probing Social Bias in Labor Market Text Generation by ChatGPT: A Masked Language Model Approach (NeurIPS 2024 · similarity 0.86)
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.
