Unverified50% confidenceBenchmarkExact time
In the title-abstract screening stage, when researchers provide explicit inclusion and exclusion criteria, model F2 scores can improve by 28.8%
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIncluding specific evaluation dimensions in prompts (such as format, word count, key points to cover, and items to exclude) can improve model output quality by an average of 30%–50% compared to vague prompts.64% similarUnverifiedIn the full-text screening stage, rationale statements written by researchers can improve model performance by approximately 15%62% similarUnverifiedIn a blind test of real-world workflows involving 270 interdisciplinary experts, the Hunyuan model scored 2.67/4, outperforming the GLM series' score of 2.51/4.62% similarUnverifiedResearch shows that on certain open-source benchmarks, the n-gram overlap rate between the training data of leading large models and the evaluation data can be as high as 15%–30%.62% similarUnverified金融风控中可以使用F2分数(β=2)来更强调查全率，垃圾邮件过滤中可以使用F0.5分数(β=0.5)来更强调查准率61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/893421API
curl https://kongchang.com/api/v1/knowledge/claims/893421MCP
get_claim(id=893421)