On Friday October 17, this site was moved to a new server, https://mw.hh.se. The original address will continue to work. Whithin a week or two this site will return to the original address. /Peo HH IT-dep

WG211/M25Barke: Revision history

From WG 2.11
Jump to navigationJump to search

Diff selection: Mark the radio buttons of the revisions to compare and hit enter or the button at the bottom.
Legend: (cur) = difference with latest revision, (prev) = difference with preceding revision, m = minor edit.

21 November 2025

  • curprev 20:1320:13, 21 November 2025 Jacques talk contribs 1,280 bytes +1,280 Created page with "Toward Trustworthy LLM Generation: From Verifiable Proofs to Reliable Agents Verifiable programming provides a concrete foundation for trustworthy LLM generation. This talk introduces DeepProof-32B, a 32-billion-parameter reasoning model designed for program verification and trained with reinforcement learning from verifiable rewards, achieving state-of-the-art performance among open models. The first part distills lessons from building models that can reason about thei..."
Retrieved from "http://mw.hh.se/wg211/index.php?title=WG211/M25Barke"