[KongchangAI]
BenchmarkMicroLens-100k

MicroLens-100k

A dataset benchmark in the short-video recommendation domain, used to evaluate tasks such as cover ranking and prediction of user viewing behavior.

Timeline (last 90 days)

Oct 17

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.

Unverified50%

All Facts (1)

Unverified

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.

50%

Source Articles

Getting Started with the JEV-27B-VL Visual Decision ModelOct 17