Research map / Large language models / AI and society
Align When They Want, Complement When They Need! Human-Centered Ensembles for Adaptive Human-AI Collaboration
Hasan Amin, Ming Yin, Rajiv Khanna
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 |
|---|---|---|---|
| route between aligned and complementary specialists (oracle) instead of single standard AI | Standard AI · WoofNette | team accuracy (%): 69.13 -> 74.75 | Improved |
| use Rational Routing Shortcut (confidence-based routing) instead of oracle routing | Adaptive AI · WoofNette | team accuracy (%): 74.75 -> 75.13 | No clear effect |
| use adaptive ensemble instead of behavior-aware single model | Behavior-Aware AI · WoofNette | team accuracy (%): 70.90 -> 75.13 | Improved |
| train a complementary-only specialist instead of standard AI | Standard AI · WoofNette | team accuracy (%): 69.13 -> 69.96 | No clear effect |
| train an aligned-only specialist instead of standard AI | Standard AI · WoofNette | team accuracy (%): 69.13 -> 60.73 | Hurt |
| use behavior-aware training instead of standard AI training | Standard AI · WoofNette | team accuracy (%): 69.13 -> 70.90 | Improved |
| increase specialist divergence (delta) (ablation) | Adaptive AI · College Admissions (simulated) | team accuracy (%) | Improved |
| increase human accuracy in alignment region (ablation) | Adaptive AI · College Admissions (simulated) | team accuracy (%) | Improved |
| make aligned fraction unbalanced away from 0.5 (ablation) | Adaptive AI · College Admissions (simulated) | team accuracy (%) | Hurt |
| add noise to group (region) labels used for routing (ablation) | Adaptive AI · College Admissions (simulated) | team accuracy (%) | Hurt |
Closest papers by meaning
- Position: Towards Bidirectional Human-AI Alignment (NeurIPS 2025 · similarity 0.90)
- Learning to Decide with AI Assistance under Human-Alignment (NeurIPS 2026 · similarity 0.89)
- An Evolutionary Perspective on AI Alignment (Student Abstract) (AAAI 2025 · similarity 0.89)
- A No Free Lunch Theorem for Human-AI Collaboration (AAAI 2025 · similarity 0.87)
- NestRL: A Nested Training Regime for Mutual Adaptation in Human–AI Teaming (NeurIPS 2026 · similarity 0.87)
- Human-AI Teaming Through the Lens of Calibration (NeurIPS 2026 · 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.
