Unverified50% confidenceBenchmarkExact time
In the multi-stage Depot study of EmbodiedIntentBench, a configuration that achieved 87.5% full-depot target recall on visible target observations completed only 1 of 16 tasks, whereas a configuration with lower recall completed 14.
1
Sources
50%
Confidence
Long-term
Relevance
11/10/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn cross-task transfer experiments, the research states inherited from three source tasks raised the average base value of 12 held-out NatureBench tasks from 12.01 under cold start to 21.89, and the proportion matching SOTA jumped from 41.7% to 86.1%.69% similarUnverifiedAutomationBench-AA's 68% completion rate indicates that models still have about one-third of failure space in automated tasks, and there is still a gap to reliable production-grade agents.67% similarUnverifiedIn task B, after reweighting, about 17% of checks would be routed to the managed fallback due to verification failure, actually increasing the load that was originally intended to be reduced.66% similarUnverifiedIn 23 tasks, the best configuration successfully obtained all key facts from stakeholders, but among them only 10 tasks achieved a normalized business improvement of 0.8 or above.66% similarUnverifiedBy improving the sampling process, while keeping the same weights, the nominal maximum layer-evaluation budget required for about 96% accuracy dropped from 60,000 (100 attempts × 50 steps × 12 layers) to 1,200 (5 attempts × 20 steps × 12 layers), roughly a 50-fold reduction.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1153689API
curl https://kongchang.com/api/v1/knowledge/claims/1153689MCP
get_claim(id=1153689)