Tag
This paper introduces SWE-Proof, a benchmark of formally verified code patches for real-world software issues, demonstrating that formal verification improves error detection in LLM-generated code and identifies specification synthesis as a key open problem.
An essay by Terence Tao examining how the mathematics community should respond to AI tools capable of research-level tasks, focusing on clarifying the implicit goals and values of mathematical research.
Lawrence Paulson discusses a bogus refutation of the Collatz conjecture caused by a bug in the Lean kernel, reflecting on proof objects and soundness in proof assistants.