# Textual Emotion Analysis on DFEC internally curated 2025 (test) Source: [Textual Emotion Analysis on DFEC internally curated 2025 (test)]() Collection index: [llms.txt]() Data updated: 2026-04-06T07:15:01.827Z Evaluation of Textual Emotion Analysis performance on the test split of the DFEC internally curated 2025 dataset, reporting metrics including Correctness, Context, CIDEr, and Rouge-L. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Textual Emotion Analysis | | Dataset (including any stated split) | DFEC internally curated 2025 (test) | | Metric | Test Score (2000 samples) | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 17 of 17 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Test Score (2000 samples)](): Higher is better; unit: SCALAR. - [Correctness Score](): Higher is better; unit: SCALAR. - [Detail Score](): Higher is better; unit: SCALAR. - [Context Score](): Higher is better; unit: SCALAR. - [Temporal Score](): Higher is better; unit: SCALAR. - [CIDEr](): Higher is better; unit: PERCENT. - [Rouge-L](): Higher is better; unit: PERCENT. - [AutoDQ Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Test Score (2000 samples) | Result date | Report paper | | --- | --- | --- | --- | | FaVChat (Qwen3) | 9.01 | Not recorded | 68fa77611e209131cea72d93 | | FaVChat (Qwen2.5) | 8.89 | Not recorded | 68fa77611e209131cea72d93 | | Qwen3-VL-8B-Face | 8.87 | Not recorded | 68fa77611e209131cea72d93 | | Qwen2.5-VL-7B-Face | 8.64 | Not recorded | 68fa77611e209131cea72d93 | | Qwen2.5-VL-72B | 8.47 | Not recorded | 68fa77611e209131cea72d93 | | Qwen3-VL-30B-A3B | 8.44 | Not recorded | 68fa77611e209131cea72d93 | | Qwen3-VL-8B | 7.99 | Not recorded | 68fa77611e209131cea72d93 | | Qwen2.5-VL-7B | 7.84 | Not recorded | 68fa77611e209131cea72d93 | | Claude4-Sonnet | 7.84 | Not recorded | 68fa77611e209131cea72d93 | | GPT-4o | 7.67 | Not recorded | 68fa77611e209131cea72d93 | | Omni-Emotion | 7.53 | Not recorded | 68fa77611e209131cea72d93 | | HumanOmni | 7.51 | Not recorded | 68fa77611e209131cea72d93 | | PLLaVA | 7.47 | Not recorded | 68fa77611e209131cea72d93 | | FaceTrack-MM | 7.47 | Not recorded | 68fa77611e209131cea72d93 | | Emotion-LLaMA | 7.42 | Not recorded | 68fa77611e209131cea72d93 | | VideoLLaMA2 | 7.41 | Not recorded | 68fa77611e209131cea72d93 | | ShareGPT4Video | 7.33 | Not recorded | 68fa77611e209131cea72d93 | ### FaVChat (Qwen3) Test Score (2000 samples): 9.01. Result ID: 69d35d2b47f34ce924bab715. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Size":"8B","Fine-tuned on FaVChat-170K":"true"}. This record does not assert independent reproduction. ### FaVChat (Qwen2.5) Test Score (2000 samples): 8.89. Result ID: 69d35d2b47f34ce924bab714. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Size":"7B","Fine-tuned on FaVChat-170K":"true"}. This record does not assert independent reproduction. ### Qwen3-VL-8B-Face Test Score (2000 samples): 8.87. Result ID: 69d35d2b47f34ce924bab710. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Size":"8B","Fine-tuned on FaVChat-170K":"true"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B-Face Test Score (2000 samples): 8.64. Result ID: 69d35d2b47f34ce924bab70f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Size":"7B","Fine-tuned on FaVChat-170K":"true"}. This record does not assert independent reproduction. ### Qwen2.5-VL-72B Test Score (2000 samples): 8.47. Result ID: 69d35d2b47f34ce924bab709. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Size":"72B"}. This record does not assert independent reproduction. ### Qwen3-VL-30B-A3B Test Score (2000 samples): 8.44. Result ID: 69d35d2b47f34ce924bab70b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Size":"30B"}. This record does not assert independent reproduction. ### Qwen3-VL-8B Test Score (2000 samples): 7.99. Result ID: 69d35d2b47f34ce924bab70a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Size":"8B"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B Test Score (2000 samples): 7.84. Result ID: 69d35d2b47f34ce924bab708. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Size":"7B"}. This record does not assert independent reproduction. ### Claude4-Sonnet Test Score (2000 samples): 7.84. Result ID: 69d35d2b47f34ce924bab70c. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT-4o Test Score (2000 samples): 7.67. Result ID: 69d35d2b47f34ce924bab70d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Omni-Emotion Test Score (2000 samples): 7.53. Result ID: 69d35d2b47f34ce924bab713. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Size":"7B","Visual and speech modalities training":"true"}. This record does not assert independent reproduction. ### HumanOmni Test Score (2000 samples): 7.51. Result ID: 69d35d2b47f34ce924bab712. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Size":"7B","Visual and speech modalities training":"true"}. This record does not assert independent reproduction. ### PLLaVA Test Score (2000 samples): 7.47. Result ID: 69d35d2b47f34ce924bab707. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM Size":"7B"}. This record does not assert independent reproduction. ### FaceTrack-MM Test Score (2000 samples): 7.47. Result ID: 69d35d2b47f34ce924bab70e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Size":"7B"}. This record does not assert independent reproduction. ### Emotion-LLaMA Test Score (2000 samples): 7.42. Result ID: 69d35d2b47f34ce924bab711. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM Size":"7B","Visual and speech modalities training":"true"}. This record does not assert independent reproduction. ### VideoLLaMA2 Test Score (2000 samples): 7.41. Result ID: 69d35d2b47f34ce924bab706. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM Size":"7B"}. This record does not assert independent reproduction. ### ShareGPT4Video Test Score (2000 samples): 7.33. Result ID: 69d35d2b47f34ce924bab705. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM Size":"7B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.