Unverified50% confidenceBenchmarkExact time
On the RoboWorld benchmark, the current strongest frontier models fail on every test task, with an average reward of only 0.06.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/30/2026
First Seen
Valid until: 11/13/2026
Sources
Related Entities
Related Claims
UnverifiedRoboDojo具身智能测评显示，最强模型得分仅12.8分（人类基准为100分），仿真成功率不足9%，真机成功率12.8%77% similarUnverifiedThe Mann-Whitney U test shows that task_score under the naive survival reward falls below the random-policy baseline, with p=0.012.70% similarUnverifiedThe user pointed out that these models achieve high scores on benchmarks yet frequently fail at the simplest tasks.67% similarUnverifiedIf a difficult problem fails across all samples (pass@k ≈ 0), the reward variance is zero, the gradient signal nearly vanishes, and the parameter update approaches zero.67% similarUnverifiedIn the experiment, the task_score under the shaped reward configuration was only 0.125, compared to 0.417 for the naive reward, with an effect size of d=4.47.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1115591API
curl https://kongchang.com/api/v1/knowledge/claims/1115591MCP
get_claim(id=1115591)