Research map / Large language models / Agents, code and math
Towards Solving the Gilbert-Pollak Conjecture via Large Language Models
Yisi Ke, Tianyu Huang, Yankai Shu, Di He, Jingchu Gai, Liwei Wang
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 bottleneck-region reflection feedback to the LLM (ablation) | LLM lemma system · Steiner ratio lower bound | certified lower bound: no improvement over 10 rounds without it -> 0.8559 with it | Improved |
| use GPT-5 vs Gemini 3 Pro as backbone (ablation) | LLM lemma system · Steiner ratio lower bound | certified lower bound: approximately 0.8559 for both | No clear effect |
| generate rule-constrained lemmas as code instead of end-to-end proofs | LLM lemma system · Steiner ratio lower bound | fraction of effective lemmas: more than 80% effective | Improved |
| run the full LLM-plus-verifier loop | LLM lemma system · Steiner ratio lower bound | certified lower bound: 0.824 -> 0.8559 | Improved |
Closest papers by meaning
- Approximation algorithms for combinatorial optimization with predictions (ICLR 2025 · similarity 0.84)
- Machine Learning meets Algebraic Combinatorics: A Suite of Datasets Capturing Research-level Conjecturing Ability in Pure Mathematics (ICML 2025 · similarity 0.84)
- NN-Steiner: A Mixed Neural-Algorithmic Approach for the Rectilinear Steiner Minimum Tree Problem (AAAI 2024 · similarity 0.84)
- Replicability in Learning: Geometric Partitions and KKM-Sperner Lemma (NeurIPS 2024 · similarity 0.83)
- Solving Inequality Proofs with Large Language Models (NeurIPS 2025 · similarity 0.83)
- A SAT + Computer Algebra System Verification of the Ramsey Problem R(3, 8) (Student Abstract) (AAAI 2024 · similarity 0.82)
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.
