Pruning Attention Layers Slashes Action Expert FLOPs by 94% in Robotics
mathildepapillo · x · 2026-08-05
Based on deep insights into the model interface, developers fine-tuned the MolmoBOT and Pi05 models, retaining only specific attention layers (e.g., L24 and Pi05 L14-16). This drastically reduced the action expert's computation (cutting FLOPs by 94% and 15%, respectively), with no statistically significant difference found in success rates or execution times.
More from Embodied
- URXR One Spatial Display Glasses Launch on Kickstarter Weighing Only 93g — OwariDa · 2026-08-05
- NVIDIA Open-Sources Alpamayo 2 Reasoning Model for Autonomous Driving — CodeByPoonam · 2026-08-05
- Impressive Progress on NIST Robot Benchmark: Models Master Complex Contact-Rich Tasks — chris_j_paxton · 2026-08-05
- NTU Releases ACE-Data-0: Large-Scale Synchronized Multimodal Robotics Dataset — liuziwei7 · 2026-08-05
- Boston Dynamics Founder: Home Robots Won't Be Too Long — LexSokolin · 2026-08-05
- Tesla's Planned 'Terafab' to Span 10x Giga Texas, Target 1TW AI Compute — XFreeze · 2026-08-05