Research map / Multi-agent systems and games / Games and equilibria
Agentmandering: A Game-Theoretic Framework for Fair Redistricting via Large Language Model Agents
Hao Li, Haotian Chen, Ruoyuan Gong, Juanjuan Wang, Hao Jiang
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 |
|---|---|---|---|
| add Choose-and-Freeze LLM agents on top of Recom candidate generation (vs plain Recom sampling) | Recom · Arizona redistricting | Unfairness (x10^-1): 4.27 -> 4.03 | Improved |
| add Choose-and-Freeze LLM agents on top of Recom candidate generation | Recom · Wisconsin redistricting | Unfairness (x10^-1): 4.21 -> 3.93 | Improved |
| add Choose-and-Freeze agents (variance of metrics) | ensemble sampling baselines · seven swing states | standard deviation: standard deviations at least two orders of magnitude smaller | Improved |
| add Choose-and-Freeze agents (compactness) | Recom · Arizona redistricting | PPS (x10^-2): 6.77 -> 4.77 | Hurt |
| add Choose-and-Freeze agents (Nevada unfairness vs enacted plan) | CD-2020 · Nevada redistricting | Unfairness (x10^-1): 4.29 -> 4.30 | No clear effect |
| use LLM agents instead of rule-based choose/freeze (partisan bias, population deviation, compactness rules) (ablation) | Agentmandering · swing states | Unfairness | Improved |
| increase candidate maps per round from 10 to 100 (ablation) | Agentmandering · Arizona | Unfairness (x10^-1): 5.02 -> 4.03 | Improved |
| increase candidate maps per round from 100 to 500 (ablation) | Agentmandering · Arizona | Unfairness (x10^-1): 4.03 -> 4.03 | No clear effect |
| increase candidate maps per round from 10 to 100 (ablation) | Agentmandering · Pennsylvania | Unfairness (x10^-1): 4.29 -> 3.48 | Improved |
| let Republican vs Democratic agent choose first (ablation) | Agentmandering · swing states | fairness metrics | No clear effect |
| swap candidate generator from Recom to Flip (ablation) | Agentmandering · swing states | Bias, Unfairness, PPS | No clear effect |
| swap candidate generator from Recom to Flip (ablation) | Agentmandering · swing states | Population Deviation | Improved |
Closest papers by meaning
- Implications of Distance over Redistricting Maps: Central and Outlier Maps (AAAI 2024 · similarity 0.85)
- Political Actor Agent: Simulating Legislative System for Roll Call Votes Prediction with Large Language Models (AAAI 2025 · similarity 0.84)
- Strategic Manipulation in Temporal Voting with Undesirable Candidates (Student Abstract) (AAAI 2025 · similarity 0.83)
- Unravelling Expressive Delegations: Complexity and Normative Analysis (AAAI 2024 · similarity 0.83)
- Temporal Fairness in Multiwinner Voting (AAAI 2024 · similarity 0.83)
- Replicating Electoral Success (AAAI 2025 · similarity 0.83)
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.
