Skip to content

Latest commit

 

History

History
38 lines (22 loc) · 1.7 KB

File metadata and controls

38 lines (22 loc) · 1.7 KB

Changelog

All notable changes to this project will be documented in this file.

bug-fix 2026-01-11

Released v2.1.0.

  • Bug fixed: Resolved several issues from previous versions.

bug-fix 2025-11-03

  • Fixed the issue in ViDoSeek-page and MMLongBench-page. More details in this issue.

new-release 2025-08-08

  • Release the raw video files here. Please note that raw videos are not required for MMEB evaluation — video frames are all you need. We provide the raw videos only in case they are useful for your own purposes. Please refer to our main data repository for more instructions.

bug-fix 2025-07-31

  • Updated the MomentSeeker task by deduplicating some test cases. More details in this issue.

new-release 2025-06-03

Released v2.0.0.

  • Expanded Modality Support: VLM2Vec now supports unified training and evaluation on three modalities: images, videos, and visual documents.
  • VLM2Vec v2: Initial release of the V2 model and framework.
  • MMEB-v2 Benchmark: Introduced a new comprehensive benchmark for evaluating performance across all supported modalities.

new-release 2025-02-11

Released v1.1.0.

  • Refactored the sub-batch splitting logic within GradCache and the VLM processor to simplify future extensions.