EventHallusion
Models answer questions about video events that can prompt false descriptions. The evaluation tests whether answers stay grounded in the clip.
- Made by
- EventHallusion authors
- Released
- September 2024
- Kind
- Video understanding
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- All public
- Measures
- Propensity
- Behavior
- Hallucination
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Last model added
- November 2023
- Scores refreshed
- November 20th, 2026
GPT-4o (snapshot unspecified) tops the leaderboard with 91.9% GPT-4V leads with 76.5%
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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. The best model we track is GPT-4V, at 76.5%. This view shows only models we track, each at its best.
- 1 GPT-4o (snapshot unspecified) Default 91.9%
- 2 Gemini-1.5-Pro (snapshot unspecified) Default 80.4%
- 3 GPT-4V Default 76.5%
- 4 LLaVA-NeXT-Video TCD 67.5%
- 5 VILA TCD 66.5%
- 6 PLLaVA Default 60.6%
- 7 LLaMA-VID Default 54%
- 8 ShareGPT4Video Default 49.1%
- 9 Video-LLaVA Default 46%
- 10 VideoChatGPT Default 36.4%
- 11 VideoChat2 TCD 34.7%
Source: The EventHallusion paper
- 1 GPT-4V Default 76.5%
Source: The EventHallusion paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4o (snapshot unspecified) | 91.9% | Default | Official leaderboard | |
| 2 | Gemini-1.5-Pro (snapshot unspecified) | 80.4% | Default | Official leaderboard | |
| 3 | GPT-4V | 76.5% | Default | Nov 2023 | Official leaderboard |
| 4 | LLaVA-NeXT-Video | 67.5% | TCD | Official leaderboard | |
| 5 | VILA | 66.5% | TCD | Official leaderboard | |
| 6 | PLLaVA | 60.6% | Default | Official leaderboard | |
| 7 | LLaMA-VID | 54% | Default | Official leaderboard | |
| 8 | ShareGPT4Video | 49.1% | Default | Official leaderboard | |
| 9 | Video-LLaVA | 46% | Default | Official leaderboard | |
| 10 | VideoChatGPT | 36.4% | Default | Official leaderboard | |
| 11 | VideoChat2 | 34.7% | TCD | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4V | 76.5% | Default | Nov 2023 | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official EventHallusion 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 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%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 3 years ago, in November 2023, for GPT-4V. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (11 entries).
- Pass Models scored 1 model we track, plus 10 other entries on its leaderboard.
- Warning Newest score The newest came 3 years ago, in November 2023, for GPT-4V.
- Note Frontier models None of its models are on the capabilities index.
- Pass Headroom The best is 76.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 1 row for models we track and 10 other entries at the last read.
- 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.