Research map / Large language models / Alignment and preferences
Benchmarking Overton Pluralism in LLMs
Elinor Poole-Dayan, Jiayi Wu, Taylor Sorensen, Jiaxin Pei, Michiel A. Bakker
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 Gemini 2.5 Pro few-shot free-response judge instead of mean-of-others | rating prediction · Overton rating prediction | MAE: 0.70 -> 0.66 | Improved |
| use Gemini 2.5 Pro few-shot free-response judge instead of semantic similarity | rating prediction · Overton rating prediction | MAE: 0.72 -> 0.66 | Improved |
| use Gemini 2.5 Pro few-shot free-response judge instead of mean-of-others | rating prediction · Overton rating prediction | Spearman rho: 0.64 -> 0.66 | Improved |
| replace human ratings with Gemini predictions in leave-one-model-out | automated benchmark · OvertonBench model ranking | Spearman rho: 0.88 | Improved |
| use Gemini judge for Claude 3.7 Sonnet | automated benchmark · Model Slant adjusted Overton Score | adjusted Overton Score: 0.226 -> 0.329 | Hurt |
| pool the best responses across all eight models | LLMs · OvertonBench human study | Overton Score: 0.433 -> 0.687 | Improved |
| compare DeepSeek V3 against single-largest-viewpoint baseline | DeepSeek V3 · OvertonBench weighted | Weighted Overton Score: 0.524 -> 0.530 | Improved |
| vary coverage threshold from 4.0 (ablation) | model ranking · Full dataset | Kendall tau: 0.64 | No clear effect |
| vary coverage threshold from 4.0 (ablation) | model ranking · Model Slant subset | Kendall tau: 0.84 | No clear effect |
Closest papers by meaning
- Position: A Roadmap to Pluralistic Alignment (ICML 2024 · similarity 0.89)
- Bridging Human and LLM Judgments: Understanding and Narrowing the Gap (NeurIPS 2025 · similarity 0.88)
- Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference (ICML 2024 · similarity 0.87)
- More Is Not More: What Matters for Diversity in LLM Opinions? (NeurIPS 2026 · similarity 0.86)
- On Evaluating LLM Alignment by Evaluating LLMs as Judges (NeurIPS 2025 · similarity 0.86)
- Examining Alignment of Large Language Models through Representative Heuristics: the case of political stereotypes (ICLR 2025 · 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.
