Research · Updated 16 Oct, 09:30 am IST
IneqAgent combines LLMs and symbolic computation to prove inequalities in Lean
Why it matters for readers: It matters because it produces machine-checked mathematical proofs by combining AI reasoning with exact computation, not just heuristic guesses.
- IneqAgent coordinates lemma construction, goal reformulation, and backend selection with lemma-level Monte Carlo Tree Search.1
- It uses complementary symbolic backends — sum-of-squares, Bernstein, and successive difference-substitution — to produce exact certificates for Lean proof generation.1
- Across four benchmarks IneqAgent with DeepSeek-V4-Flash solved 175 of 186 problems (94.09%), the highest success rate reported among evaluated methods.1
- IneqAgent achieves higher success rates at low LLM inference cost thresholds and lower mean inference costs on successful runs than a Direct Agent baseline using the same backbones.1
- The system produced Lean-verified strengthenings of Dittert's conjecture in dimensions four and five and certified the Holens–Djokovic (4,4) inequality, implying Merris's conjecture in dimension four.1
Get a brief like this every morning
Uzha reads hundreds of sources and gives you the stories that matter for your work, with every source linked. Free.
Get started