Unverified50% confidenceFactExact time
CAD World contains 200 tasks, covering 11 workflow categories and 183 distinct knowledge points.
1
Sources
50%
Confidence
Long-term
Relevance
9/26/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCMU's Odyssey benchmark contains 200 tasks, each accompanied by multiple rubrics, supporting partial scoring rather than a single binary judgment.62% similarUnverifiedA single detailed tool description may consume hundreds of tokens, and ten tools can easily exceed a fixed overhead of one thousand tokens.59% similarUnverifiedThe research team evaluated MAGS on three categories of tasks: 100 CUDA kernels, 100 terminal scripts, and 20 robotic arm tasks, totaling 220 samples.57% similarUnverifiedCOCO 数据集包含约12万张训练图像、80个类别、超过150万个目标实例，由微软研究院发布56% similarUnverifiedHumanEval包含164道编程题，MBPP包含约500道编程题56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/951473API
curl https://kongchang.com/api/v1/knowledge/claims/951473MCP
get_claim(id=951473)