Get Started
Home
Topics
Search
Library
Research questionHow can we rigorously improve lower bounds for the Euclidean Steiner ratio?The Gilbert–Pollak conjecture predicts a Steiner ratio of at least √3/2, but the strongest reported lower bound before this work was 0.824. The difficulty is obtaining mathematically certified improvements for every finite point set, rather than evidence from selected geometric instances.
AI
Code Generation & Program Synthesis
Reasoning
Research Paper
Latest papersRecent research connected to this question, newest first.Towards Solving the Gilbert-Pollak Conjecture via Large Language ModelsThe source examines an LLM-based system that generates rule-constrained geometric lemmas as executable code and combines them into verification functions for certified lower bounds. It reports a bound of 0.8559 using thousands of LLM calls, but does not establish the full conjecture.research paper · Sep 2, 2026
Related questions
How can we certify optimality for a long-unsolved maximum independent-set benchmark?When can a random projection preserve JL distance bounds yet lose geometric variation needed for rankings and inference?What convergence guarantees and irreducible errors arise in random-reshuffling optimization for finite sums and composite objectives?Can one squared-loss estimator achieve both minimax and universal exponential rates for finite versus countably infinite hypothesis classes?