Research map / Learning theory and algorithms / Combinatorial and search optimization
BiO-HMC: Dynamic Human-Machine Collaboration for Consensus Decision-Making via Bilevel Optimization
Yinghui Pan, Shuaijie Zhao, Shenbao Yu, Zongyang Liu, Yifeng Zeng, Han Liu, Mingwei Lin
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 BiO-HMC instead of MC-based planners and limited lookahead | consensus decision-making · Sync1-5 (unlimited responses) | # interactions: 450 -> 62 (MC-EVA vs BiO-HMC on Sync1) | Improved |
| use BiO-HMC instead of baselines on complex answer-question relations | consensus decision-making · Sync4 | HSR: 0.810 -> 0.987 (PMC-Cnts vs BiO-HMC) | Improved |
| use BiO-HMC on real-world Galaxy Zoo | consensus decision-making · Galaxy Zoo | # interactions: 12,740 -> 287 (MC-EVA+AS horizon 200 vs BiO-HMC) | Improved |
| increase planning horizon from 100 to 200 (ablation) | MC-EVA+AS · Galaxy Zoo | HSR: 0.709 -> 0.899 (more interactions and longer runtime) | Mixed |
| keep increasing horizon of MC baselines (ablation) | MC-VOI, MC-EVA, MC-EVA+AS, PMC-Cnts · Galaxy Zoo | HSR | No clear effect |
| increase the number of workers (ablation) | BiO-HMC · Sync1, Sync3, Sync5, GZ (limited responses, noise 20%) | HSR: reaches 80% after about 8 workers | Improved |
| add more workers to baselines (ablation) | MC-VOI, MC-EVA, MC-EVA+AS, PMC-Cnts · limited responses (max 20 workers) | HSR | No clear effect |
| vary noise rate from 10% to 50% (ablation) | BiO-HMC · Sync3 (limited responses) | HSR: all above 98% at end of process | No clear effect |
| low noise rate 10% (ablation) | BiO-HMC · Sync3 (limited responses) | HSR | Hurt |
| use learned question prioritization instead of natural, random or greedy selection (ablation) | BiO-HMC · Sync1, Sync3, Sync5, GZ | HSR | Improved |
Closest papers by meaning
- Belief-Calibrated Multi-Agent Consensus Seeking for Complex NLP Tasks (NeurIPS 2025 · similarity 0.84)
- Cascaded Language Models for Cost-Effective Human–AI Decision-Making (NeurIPS 2025 · similarity 0.82)
- Optimal Design for Human Preference Elicitation (NeurIPS 2024 · similarity 0.82)
- IQA-EVAL: Automatic Evaluation of Human-Model Interactive Question Answering (NeurIPS 2024 · similarity 0.82)
- Label Aggregation for Composite Crowd Tasks by Worker Ability Constraint Satisfaction (AAAI 2025 · similarity 0.81)
- AgentHop: A Diagnostic Benchmark for Agentic Multi-Hop Scientific Question Answering (NeurIPS 2026 · similarity 0.81)
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.
