Unverified50% confidenceFactExact time
The AskStream benchmark contains four task categories and eight subtasks, covers 438 videos, and includes manual annotation of 2,714 frame-level time windows.
1
Sources
50%
Confidence
Long-term
Relevance
10/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe GraftSet-Video dataset constructed by the research team contains 2,397 prompts corresponding to 9,588 format-matched video examples, with an average of approximately 4 examples per prompt.67% similarUnverifiedThe CapMem benchmark dataset contains 75 videos with a total duration of 33.7 hours, 1,000 manually annotated multiple-choice questions, covering 16 different scenarios.67% similarUnverifiedIn actual testing, using an Apify actor to find the 30 most-viewed videos under the TikTok #AI tools hashtag scanned over 300 videos, and the Agent output a structured CSV after reading the transcripts.63% similarUnverifiedVideo rendering is completed through RunningHub workflows, with consumer-grade API supporting up to 5 parallel tasks and enterprise-grade API supporting 100 parallel tasks.62% similarUnverifiedA Bilibili creator released a 177-episode, 14-hour-45-minute Agent development interview question video series.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1114220API
curl https://kongchang.com/api/v1/knowledge/claims/1114220MCP
get_claim(id=1114220)