SMILE-Temporal - Laughter Detection: leaderboard

Laughter detection on SMILE-Temporal (2026), clips of the SMILE laughter-reasoning video dataset with temporal laughter annotations: a zero-shot model is asked whether laughter occurs in a 5-second video clip and, if so, when it starts and ends. F1 of the laughter or no-laughter decision over all test clips, times 100; higher is better.

Metric: SMILE-Temporal laughter detection F1 (times 100, so 0-100) over all test segments: the model is prompted, zero-shot, to say whether laughter occurs in a 5-second video segment, on the SMILE-Temporal test split (clips of the SMILE laughter-reasoning video dataset); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 3 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Qwen2.5-Omni-7B78.3#1884
2Gemini 3 Flash72.4#175
3MTLLFM Qwen2-Audio-7B (checkpoint unspecified)33.5

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=smile-temporal-laughter-detection · How It Works · Data refreshed daily, snapshot 2026-11-21.