Show HN: TheoremSmith: Mass-Producing Hard Proof Problems to Train LLMs
0 comments
- Points10
- Comments0
0 comments
Making models better at formal verification will help push the boundaries of math research and significantly reduce the number of bugs in our code. Recently, training models on RL environments (tasks paired with verifiers that reward correct solutions) has driven big gains in model performance, and proofs are unusually easy to verify. Here we introduce TheoremSmith, a pipeline for generating synthetic formal verification environments. It reliably creates multi-hour, frontier-difficulty tasks…
Story details from Hacker News ↗ · linked article on charleshuang.me.
Show HN: TheoremSmith: Mass-Producing Hard Proof Problems to Train LLMs was first seen on Hacker News · Show HN on Nov 10, 2026. Since then it has appeared in 47 checks on 2 different days. Its best position was #5 of 30, reached on Nov 11, 01:21 UTC. It dropped off the list after Nov 11, 18:21 UTC.
Position on Hacker News · Show HN at every check (lower is better).
Best rank per day (UTC).
Platform metric as reported by Hacker News.
Archived daily, weekly and monthly lists that include this entry.
Not at the moment. It was last on the list on Nov 11, 18:21 UTC, after 2 days of trending.
#5 on Hacker News · Show HN, on Nov 11, 01:21 UTC. Trendoday checks this list every 30 minutes.
It was first seen on Nov 10, 2026 and has been on the list on 2 days (47 checks) since.
10 points and 0 comments; submitted by charles6huang on Nov 10, 2026.