Act3D trains VLMs to explore 3D with outcome-only RL, hitting 53.2% across six benchmarks
_akhaliq · x · 2026-10-20
Act3D teaches a VLM to think and explore in 3D: given a single image, it uses outcome-only RL to decide when and how to move the camera and gather evidence.
- 53.2% average across six 3D benchmarks
- +22% over its Qwen base model
- Scores above Gemini 3.6 Flash
Related event: ACT3D: RL with Outcome Rewards Teaches VLMs to Actively Explore 3D Worlds(2 posts)→
More from Research
- RFD4-Proteina's 'underrated' docs detail partial diffusion for protein redesign — AllThingsApx · 2026-10-23
- MPI-Sec PhD Thesis on Vulnerability Detection Benchmarks Includes Pre-LLM Classic "Top Score on the Wrong Exam" — mboehme_ · 2026-10-23
- LiveNerf's 29-day data finds no nerfing of Claude Opus 5.5 — TheOnlyVibemaster · 2026-10-23
- New paper makes 9 falsifiable predictions toward a major AI-enabled terror attack by 2027-2028 — StephenLCasper · 2026-10-23
- HumanoidPose recovers humanoid robots' full-body joint states from RGB video alone — cholz · 2026-10-23
- Sophont's Holon-S1 predicts genes and biomarkers from routine pathology slides — iScienceLuvr · 2026-10-23