Corrections to the PeerQA and DeepPlanning scores of WeMM models

#188
No description provided.

We corrected PeerQA and DeepPlanning from per-query local retrieval to task-wide global retrieval and updated the scores for all three WeMM-Embedding models (2B, 4B, and 9B); all other task results remain unchanged.

JUNJIE99 changed pull request status to open
ziyjiang changed pull request status to merged

Sign up or log in to comment