Benchmark Almanac

RefAV

HYU_OASIS (exp104_v10_track6_debug) 38.5 tops its leaderboard
Not in model cards

Models find driving scenarios described in natural language. They identify when each scenario occurs and track the objects involved.

Zeekr_UMCV led RefProg and the other CVPR 2025 competition entries with 53.38 HOTA-Temporal in June 2025.

Released
March 2025
Kind
Detection and captioning
Level
University
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Video → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
In views
Vision
14
Entries on its own leaderboard, 0 models we track

HYU_OASIS (exp104_v10_track6_debug) tops the leaderboard with 38.5

Best score per entry from the benchmark's own leaderboard, read on November 20th, 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. 1 HYU_OASIS (exp104_v10_track6_debug) 38.5
  2. 2 MTL (Argonaut) 37.0
  3. 3 MISI (AutoMine) 36.4
  4. 4 Zelos_Agent_Driving 33.5
  5. 5 Lilly 32.5
  6. 6 DataClaw 31.4
  7. 7 Predicate 31.1
  8. 8 blackTea 30.9
  9. 9 lazydogs 30.8
  10. 10 badr 30.6
  11. 11 MICTeam (MICTeam) 27.9
  12. 12 Host_13130_Team (RefProg) 26.3
  13. 13 Host_13130_Team (SM-Agent) 23.3
  14. 14 aliz (SM-Fusion (Qwen2.5-Coder-32B)) 18.8

Source: Official RefAV leaderboard

#EntryScoreSettingSubmitted byDateSource
1 HYU_OASIS (exp104_v10_track6_debug) 38.5 exp104_v10_track6_debug HYU_OASIS May 2026 Official leaderboard
2 MTL (Argonaut) 37.0 Argonaut MTL May 2026 Official leaderboard
3 MISI (AutoMine) 36.4 AutoMine MISI May 2026 Official leaderboard
4 Zelos_Agent_Driving 33.5 Zelos_Agent_Driving Zelos_Agent_Driving May 2026 Official leaderboard
5 Lilly 32.5 Lilly Lilly May 2026 Official leaderboard
6 DataClaw 31.4 DataClaw DataClaw May 2026 Official leaderboard
7 Predicate 31.1 Predicate Predicate May 2026 Official leaderboard
8 blackTea 30.9 blackTea blackTea May 2026 Official leaderboard
9 lazydogs 30.8 lazydogs lazydogs May 2026 Official leaderboard
10 badr 30.6 badr badr May 2026 Official leaderboard
11 MICTeam (MICTeam) 27.9 MICTeam MICTeam May 2026 Official leaderboard
12 Host_13130_Team (RefProg) 26.3 RefProg Host_13130_Team Feb 2026 Official leaderboard
13 Host_13130_Team (SM-Agent) 23.3 SM-Agent Host_13130_Team Apr 2026 Official leaderboard
14 aliz (SM-Fusion (Qwen2.5-Coder-32B)) 18.8 SM-Fusion (Qwen2.5-Coder-32B) aliz May 2026 Official leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Vision

All 636 in Vision
  • MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. Qwen 3.6 Plus 86%
  • MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
  • CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
  • MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy. Seed2.1 Pro 90.7%
  • MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
  • MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy. Qwen 3.8 Max 85.9%

Data audit

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (14 entries).
  • Pass Models scored 14 entries on its leaderboard, none a model we track.
  • Note Newest score No scores yet.
  • Note Frontier models No models scored yet.
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on eval.ai: 14 entries at the last read, none a model we track.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check