Benchmark
Meld Benchmark
What is Meld?
MELD (Multimodal EmotionLines Dataset) is a multimodal multi-party dataset for emotion recognition in conversations. Contains approximately 13,000 utterances…
- Released
- 2018-10-05
- Modality
- multimodal
- Evaluates
- multimodal, psychology, creativity
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 1
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/meld?top_n=500
Meld results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Qwen2.5-Omni-7B | Qwen | 0.57 | 2025-03-27 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.