GLIDE has an LLM write failure-filtering guardrails, lifting robot task success from 0% to 70%
stepjamUK · x · 2026-09-23
Some robot tasks are impossible — not just hard — for humans to teleoperate cleanly, like carrying a plate across a table with two independently controlled arms. GLIDE takes a novel approach: an LLM reads the task, predicts where a human demonstrator will fail, and writes a filter between commands and the robot to prevent it. It then watches the data, catches remaining failures, and rewrites itself.
Results: the impossible task jumps from 0% to 70% success; two others go from 10 and 0 up to 90 and 90. Keeping the guardrail at deployment lifts a barely-working policy from 0 to 60-70%. The takeaway: much of "the robot can't learn this" is really "no human could demonstrate it in the first place."
More from Embodied
- NVIDIA open-sources Nemotron 3 Diarization to fix voice agents' speaker-blindness — andimarafioti · 2026-09-23
- Physical AI heats up: Skild AI hits $100M ARR ahead of Encord's 1,000-person summit in SF — lukas_m_ziegler · 2026-09-23
- European humanoid startup Vesoma emerges from stealth, walking its own robot in six months — Jasonio · 2026-09-23
- China Mobile open-sources Open-RAIL, a real-time execution layer for VLA models and humanoids — CyberRobooo · 2026-09-23
- Robotics data startup Eidon shuts down after 2+ years, open-sources its <$30 finger tracker — vanstriendaniel · 2026-09-23
- TUM survey unifies physics-embedded robot learning with a new taxonomy — TUM-AVS · 2026-09-23