Unverified50% confidenceBenchmarkExact time
In the MicroLens-100k evaluation, JEV-27B-VL using only cover-image scoring reached 0.727 AUC and 0.590 HR@5 among 20 candidates, while collaborative filtering trained on the history of 59,045 users achieved 0.728 AUC and 0.490 HR@5.
1
Sources
50%
Confidence
Long-term
Relevance
10/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn six public text decision benchmarks, the six-group mean of the JEV-27B / JEV-27B-VL rows is 84.07%, higher than the 83.85% of the TypeSafe Jev 1.13 hosted API and the 77.70% of NeoHorse-Jev-4B.65% similarUnverifiedOn the 1,302 expert-annotated trajectories of AgentRewardBench, JEV-27B-VL reports an AUROC of 0.91, with precision 78.4 and recall 70.2 at a threshold of 0.5.64% similarUnverifiedOn the WikiRacing task, Jev scored a perfect 30/30, while CLM-8B scored 26/30.64% similarUnverifiedOn agent tasks, Agens Volundr 32B lags behind: tau2-bench scores 74.2 (competitor 79–80), and SWE-bench Verified (50-question subset) is only 44 (competitor 58–64).64% similarUnverifiedOn Multimodal RewardBench 2, JEV-27B-VL averages 63.8 across four tasks, close to GPT-4.1 and Qwen3-VL-32B but behind GPT-5 and Gemini 3 Pro; its image editing score of 57.8 is a clear weak spot.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1050828API
curl https://kongchang.com/api/v1/knowledge/claims/1050828MCP
get_claim(id=1050828)