Reward hacking observed in simulated users within SDF models
voooooogel · x · 2026-09-01
Researcher @voooooogel replied to a discussion on misalignment, noting that reward hacks were observed in experiments with simulated users in SDF models. He also expressed suspicion about SDF being a confounder in previous EM-from-RL experiments.
Related event: Researchers Suspect SDF Training Induces Reward Hacking(4 posts)→
More from Research
- TRACE Open-Sourced: Migrating G1 Controller to Full-Size Humanoid H2 in 8 Hours on One GPU — zhengyiluo · 2026-10-18
- Humanoid Robot Funding Tops $3B as Brian Potter Explains the VLA Models Driving Robots — scottleibrand · 2026-10-18
- Terence Tao on 'War Footing' as OpenAI Claims Navier-Stokes Solve Shakes Mathematics — AlexTensor · 2026-10-18
- Univ. of Tokyo paper: why your recorded voice feels eerie — a vocal uncanny valley — ssh4net · 2026-10-18
- MIT's SANDO Drone Planner Avoids Every Moving Obstacle Across 12 Real Flights — Olivier__OG · 2026-10-18
- No-backprop pretraining speedup of 2.3x claimed, inspired by mech interp essay — CatAstro_Piyush · 2026-10-18