lyhisme commited on
Commit
b45776c
·
verified ·
1 Parent(s): ad76e1b

Add OraRL project video

Browse files
Files changed (1) hide show
  1. README.md +4 -0
README.md CHANGED
@@ -15,6 +15,10 @@ We build **OraRL**, a reinforcement-learning framework for **unified video multi
15
 
16
  Our goal is to make broad video RL both **reliable** and **efficient** by treating task annotations as positive rollouts while retaining policy samples for the on-policy baseline. The resulting **Video-ORA** family handles seven task families without chain-of-thought decoding.
17
 
 
 
 
 
18
  ## What we release
19
 
20
  - **Video-ORA-9B**: a unified video MLLM for temporal and spatial grounding, segmentation, tracking, video QA, and spatial intelligence.
 
15
 
16
  Our goal is to make broad video RL both **reliable** and **efficient** by treating task annotations as positive rollouts while retaining policy samples for the on-policy baseline. The resulting **Video-ORA** family handles seven task families without chain-of-thought decoding.
17
 
18
+ <video controls playsinline preload="metadata" width="100%" src="https://orarl.github.io/assets/orarl-teaser.mp4"></video>
19
+
20
+ <p align="center"><strong>▶ OraRL project overview (1:38)</strong></p>
21
+
22
  ## What we release
23
 
24
  - **Video-ORA-9B**: a unified video MLLM for temporal and spatial grounding, segmentation, tracking, video QA, and spatial intelligence.