Ref-YouTube-VOS
CodaLab submission: leonnnop 60.7% tops its leaderboard
Not in model cards Models segment and track the object that a sentence describes in YouTube-VOS videos. Scored by J&F on a server-graded validation split.
ECCV 2020 paper (URVOS). Microsoft's Mage-VL card reports it as Ref-YT-VOS.
- Made by
- Seoul National University and Adobe Research (Seo, Lee, Han, URVOS)
- Released
- August 2020
- Kind
- Video understanding
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- Public, plus a private test set
3
Entries on its own leaderboard, 0 models we track
CodaLab submission: leonnnop tops the leaderboard with 60.7%
Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.
- 1 CodaLab submission: leonnnop Testing 60.7%
- 2 CodaLab submission: nowherespyfly Testing 49.4%
- 3 CodaLab submission: Merci1 Testing 41.2%
Source: Official Ref-YouTube-VOS leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | CodaLab submission: leonnnop | 60.7% | Testing | May 2021 | Official leaderboard |
| 2 | CodaLab submission: nowherespyfly | 49.4% | Testing | May 2021 | Official leaderboard |
| 3 | CodaLab submission: Merci1 | 41.2% | Testing | May 2021 | Official leaderboard |
No lab model card reports it yet
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.
More in Audio and video
- Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy.
Qwen 3.8 Max 88.7% - LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization.
Gemini 4 Argon 91.7% - MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy. Qwen 3.8 Max 82.4%
- MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy. Qwen 3.8 Max 90.8%
- FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better.
Phi-4-multimodal 4.00% - VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules. Qwen3.5-Omni-Plus 93.1%