Unverified50% confidenceEventExact time
In GitHub issue #14, the OmniDocBench team reran Youtu-Parsing on v1.5 and got an overall score of 93.37, while Tencent's internal number is 93.22; the formula CDM differs by about 2 points, and the table TEDS differs by about 2 points in the opposite direction.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
Related Entities
Related Claims
UnverifiedGitHub issue #7714 calculated that injecting the same state into the context window consumes 5,764 tokens, while the semantic content itself requires only 3,587 tokens, meaning a 37.8% token overhead per LLM call.68% similarUnverifiedOn the SWE-bench benchmark (2,294 real GitHub issues), Claude 2 scored 1.96%.61% similarUnverifiedAfter GitHub Copilot streamlined the Task tool prompt, approximately 1,300 Task prompt tokens were reduced per round.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1009504API
curl https://kongchang.com/api/v1/knowledge/claims/1009504MCP
get_claim(id=1009504)