Research map / Learning theory and algorithms / Prediction and decision losses
Robust Human-AI Complementarity under Uncertainty
Yewon Byun, Bryan Wilder
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 |
|---|---|---|---|
| prompt LLM for contrarian reasoning (ablation) | GPT-5 forecasts · ForecastBench | error correlation with humans: 0.963 -> 0.498 | Improved |
| prompt LLM for contrarian reasoning (ablation) | GPT-5 forecasts · TESS studies | error correlation with humans: 0.216 -> -0.338 | Improved |
| prompt LLM with direct error correction (ablation) | GPT-5 forecasts · ForecastBench | error correlation with humans: 0.963 -> 0.995 | Hurt |
| prompt LLM with self-divergent reasoning (ablation) | GPT-5 forecasts · ForecastBench | error correlation with humans: 0.963 -> 0.966 | No clear effect |
| prompt LLM with human-divergent reasoning (ablation) | GPT-5 forecasts · ForecastBench | error correlation with humans: 0.963 -> 0.985 | No clear effect |
| restrict LLM context (different information set) (ablation) | GPT-5 forecasts · ForecastBench | error correlation with humans: 0.963 -> 0.967 | No clear effect |
| combine human and LLM via robust rule instead of calibrated human only | Robust combination · ForecastBench | MSE: 0.154 -> 0.154 | No clear effect |
| combine human and LLM via robust rule with contrarian prompt | Robust combination · TESS studies | MSE: 0.00524 -> 0.00490 | Improved |
| combine human and LLM via robust rule with contrarian prompt | Robust combination · ForecastBench | MSE: 0.154 -> 0.153 | Improved |
| increase human-AI error correlation in simulation (ablation) | Robust estimator · synthetic Gaussian MSE | MSE gain | Hurt |
| increase expert noise under positive error correlation (rho=0.75) (ablation) | Robust estimator · synthetic MSE | MSE gain | Improved |
Closest papers by meaning
- Learning to Decide with AI Assistance under Human-Alignment (NeurIPS 2026 · similarity 0.90)
- Strong Preferences Affect the Robustness of Preference Models and Value Alignment (ICLR 2025 · similarity 0.86)
- A No Free Lunch Theorem for Human-AI Collaboration (AAAI 2025 · similarity 0.86)
- Confounding-Robust Deferral Policy Learning (AAAI 2025 · similarity 0.86)
- Model Alignment as Prospect Theoretic Optimization (ICML 2024 · similarity 0.85)
- Measuring Error Alignment for Decision-Making Systems (AAAI 2025 · similarity 0.85)
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.
