Unverified50% confidenceFactExact time
The author reviewed approximately 1,500 detection calls, manually mined false positives, adjusted bounding boxes, and then retrained the model.
1
Sources
50%
Confidence
Long-term
Relevance
8/9/2026
First Seen
Sources
自建多模态数据集：从500帧到300万帧的自动化标注实践
Related Claims
UnverifiedMementoGUI的训练数据自动标注管道中，研究人员随机抽取200条轨迹做人工验证，197条被判定完全正确73% similarUnverifiedThe tested task was conducted over four rounds, dispatching a cumulative total of 68 sub-agents to complete batch prompt testing for a digital human writing skills module.70% similarUnverifiedThe experiment completed a total of 1,200 code experiments, with each experiment corresponding to one Git commit and one quantitative metric result.69% similarUnverified2014 年启动的 CommonMark 项目试图通过一份包含 600 多个测试用例的严格规范解决 Markdown 的解析歧义问题67% similarVerifiedSWE-bench Verified is a human-verified subset of approximately 500 samples, excluding tasks with vague descriptions or unreliable tests.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/712771API
curl https://kongchang.com/api/v1/knowledge/claims/712771MCP
get_claim(id=712771)