#theorem-proving

[ follow ]
#formal-verification #reinforcement-learning #lean #deepseek #lean-4 #ai #mathematics
fromwww.nature.com
3 days ago
Artificial intelligence

Olympiad-level formal mathematical reasoning with reinforcement learning

AlphaProof is an AlphaZero-inspired reinforcement-learning agent that learns rigorous formal mathematical reasoning and proof construction within the Lean interactive theorem-proving environment.
fromInfoQ
6 months ago

DeepSeek Launches Prover-V2 Open-Source LLM for Formal Math Proofs

The DeepSeek research team has expanded their evaluation framework with a new benchmark collection designed specifically for formal theorem proving assessment. This includes ProverBench, a collection of 325 formalized problems.
Scala
[ Load more ]