Research map / Probabilistic and scientific modeling / Molecules and proteins
Inferring Multi-Dimensional Ideal Points for US Supreme Court Justices
Mohammad Islam, K. S. M. Hossain, Siddharth Krishnan, Naren Ramakrishnan
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 |
|---|---|---|---|
| raise error threshold from 0.15 to 0.30 (ablation) | SCIPM, K=5 · Supreme Court votes, xi=80% | average recall: 59.14 -> 95.91 | Improved |
| raise accuracy threshold from 60% to 90% (stricter) (ablation) | SCIPM, K=5, eps 0.20 · Supreme Court votes | average recall: 93.55 -> 58.71 | Hurt |
| increase number of topics from 5 to 20 (ablation) | SCIPM, eps 0.15, xi 80% · Supreme Court votes | average recall: 59.14 -> 43.66 | Hurt |
| increase number of topics from 5 to 20 (ablation) | SCIPM, eps 0.30, xi 60% · Supreme Court votes | average recall: 99.35 -> 97.63 | No clear effect |
| use SCIPM instead of Lauderdale and Clark (2014) | vote prediction · Supreme Court votes | average recall: 48.13% -> 79.46% | Improved |
| use SCIPM instead of ATM at eps 0.15, xi 80%, K=5 | vote prediction · Supreme Court votes | average recall: 12.50 -> 59.14 | Improved |
| use SCIPM instead of ATM at eps 0.15, xi 90%, K=20 | vote prediction · Supreme Court votes | average recall: 35.00 -> 24.73 | Hurt |
| use SCIPM instead of naive Bayes classifier | vote prediction · Supreme Court votes | average recall: 27.95% -> 79.46% | Improved |
| increase number of topics from 5 to 20 at strict threshold (ablation) | SCIPM, eps 0.15, xi 90% · Supreme Court votes | average recall: 37.42 -> 24.73 | Hurt |
Closest papers by meaning
- Can an LLM Reason Like a Lawyer? Benchmarking the ability of LLMs to map the facts of a case to the elements of the applicable legal rule (NeurIPS 2026 · similarity 0.81)
- Representative Ranking for Deliberation in the Public Sphere (ICML 2025 · similarity 0.79)
- Heterogeneous Judge-Aware Ranking with Sensitivity, Disagreement, and Confidence (NeurIPS 2026 · similarity 0.79)
- Proportional Aggregation of Preferences for Sequential Decision Making (AAAI 2024 · similarity 0.79)
- Political Actor Agent: Simulating Legislative System for Roll Call Votes Prediction with Large Language Models (AAAI 2025 · similarity 0.78)
- When Neutral Summaries Are Not That Neutral: Quantifying Political Neutrality in LLM-Generated News Summaries (Student Abstract) (AAAI 2025 · similarity 0.78)
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.
