Preprints|Short CommunicationNot peer reviewedCheap Reward Models for Small-Scale RLHF ExperimentsJonas Berg, Anders HolmVersion 1 · 29 October 2026 · 10.xxxx/ctm.pp.2026.0008Abstract PDF 1,480 5120 cited
Proceedings|Full paperSandboxing Agent Tool Calls with Capability TokensMarcus Tan, Jonas BergRTUA 2026 · pp. 10–18 · 3 July 2026 · 10.xxxx/ctm.pr.2026.0002Abstract PDF 1,190 4301 cited