Unverified50% confidenceFactExact time
FDE-Gym's experiments show that even when the Agent grasps all key facts, the conversion success rate is still less than half, indicating that current models have systematic shortcomings in planning and execution capability on long-chain complex tasks.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn Agent systems, context (rather than model capability) is the true performance bottleneck.72% similarUnverified轻量模型在高难度任务上的失败往往是结构性的，切换模型比优化提示词更务实71% similarUnverifiedEven on leading models, precise temporal and spatial localization remains a core unsolved challenge; models perform acceptably on coarse-grained semantic understanding, but have clear gaps on precise localization tasks.70% similarUnverifiedWhen evaluating Agent capabilities, task completion rate is more indicative of a model's reliability than reasoning fluency.70% similarUnverifiedOne run fell below the most naive diff method due to a missing normalization step, demonstrating that preprocessing steps have a decisive impact on this type of task.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1041825API
curl https://kongchang.com/api/v1/knowledge/claims/1041825MCP
get_claim(id=1041825)