Unverified50% confidenceBenchmarkExact time
In real-robot experiments, FLORE's force-reactive policy achieved an 80% success rate on three contact-intensive tasks, whereas a position policy trained on the same demonstration data and executed by the same controller achieved only a 26.7% success rate.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/27/2026
First Seen
Valid until: 1/25/2027
Sources
Related Entities
Related Claims
UnverifiedOn the Push-T task, using 24 different appearances for three independent training runs, Concat IDM raised the planning success rate from 9.1% to 77.2%.70% similarUnverifiedAfter training in simulation, the ASTEROID policy successfully transferred to vision-limited real-world robotic manipulation tasks, improving the success rate by +40% compared with standard distillation methods.70% similarUnverifiedIn 500 automated stress tests, the unconstrained conversational Agent had a task completion rate of 61.2%, while the typed Bot Skill+MCP architecture had 98.4%, a reliability improvement of 60.8%69% similarVerifiedThrough Agent Tuning specialized training, small-parameter models such as 3B and 7B can also acquire strong Agent capabilities.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1102641API
curl https://kongchang.com/api/v1/knowledge/claims/1102641MCP
get_claim(id=1102641)