joschkastrueber.bsky.social
PhD student at the University of Tübingen, member of @bethgelab.bsky.social, @uni_tue and @MPI_IS (IMPRS-IS). LLM multi-turn post-training and evaluations.
👤 Joined on 07 February 2025
Followers
Following
Total Posts
@joschkastrueber.bsky.socialDavon kann Manuel Hagel noch seinen Enkeln erzählen: tazeins zur Wahl in Baden-Württemberg @tschlze.bsky.social @taz.de
Show PostBlogpost: bethgelab.github.io/delta-belief... Paper: alphaxiv.org/abs/intrinsi... Code: github.com/bethgelab/de... A massive thanks to my collaborators Ilze Amanda Auzina, Sergio Hernández Gutiérrez, Shashwat Goel, @bayesiankitten.bsky.social and Matthias Bethge (8/8) @bethgelab.bsky.social
Show PostThis suggests a shift in how we train agents: Instead of external critics or verifiers, 👉 Let agents learn by tracking their own uncertainty reduction. A step toward agents that reason about what they don’t know. (7/8)
Show Post👉 Learns information-seeking strategies that generalise to OOD (6/8) Despite being trained solely on 20 Questions, the agent skills transfer to new OOD tasks, such as customer service and user personalisation 👥
Show Post👉 Continues to seek information beyond the training horizon The results suggest that ∆Belief rewards generalize to longer horizons; they teach general information-seeking strategies that continue to resolve uncertainty as more evidence becomes available. 🔎 (5/8)
Show Post