Research map / Vision-language and multimodal / Cross-modal learning
SpEmoC: A Balanced Speaker-Segment Multimodal Emotion Benchmark
Sania Bano, Shahzad Ahmad, Santosh Kumar Vipparthi, Sukalpa Chanda, Subrahmanyam Murala
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| train and evaluate on balanced SpEmoC instead of MELD | EMOE · in-domain | Macro-F1: 38.19 -> 70.25 | Improved |
| train and evaluate on balanced SpEmoC instead of CAER | EMOE · in-domain | Macro-F1: 15.54 -> 70.25 | Improved |
| train and evaluate on SpEmoC instead of MELD/CAER | TCL-MAP · in-domain | Macro-F1: 45.46 -> 73.34 | Improved |
| train and evaluate on SpEmoC instead of MELD | EMOE · in-domain | imbalance gap: 20.85 -> 0.05 | Improved |
| train and evaluate on SpEmoC instead of MELD | MulT · in-domain | imbalance gap: 20.21 -> 0.65 | Improved |
| train on SpEmoC and test on MELD vs train on MELD and test on CAER | TCL-MAP · cross-dataset | W-F1: 64.21 -> 50.73 | Hurt |
| add SpEmoC pretraining before MELD fine-tuning (ablation) | EMOE · MELD | W-F1: 59.04 -> 63.26 | Improved |
| add SpEmoC pretraining before MELD fine-tuning (ablation) | EMOE · MELD Fear F1 | F1: 6.06 -> 16.33 | Improved |
| add SpEmoC pretraining before MELD fine-tuning (ablation) | EMOE · MELD Disgust F1 | F1: 6.38 -> 17.65 | Improved |
| add SpEmoC pretraining before CAER fine-tuning (ablation) | EMOE · CAER | W-F1: 36.52 -> 38.16 | Improved |
| add MELD pretraining before SpEmoC fine-tuning (ablation) | EMOE · SpEmoC | W-F1: 70.30 -> 71.47 | Improved |
| add CAER pretraining before SpEmoC fine-tuning (ablation) | EMOE · SpEmoC | W-F1: 70.30 -> 71.84 | Improved |
Closest papers by meaning
- AffectGPT: A New Dataset, Model, and Benchmark for Emotion Understanding with Multimodal Large Language Models (ICML 2025 · similarity 0.89)
- Emotion-LLaMA: Multimodal Emotion Recognition and Reasoning with Instruction Tuning (NeurIPS 2024 · similarity 0.88)
- MME-Emotion: A Holistic Evaluation Benchmark for Emotional Intelligence in Multimodal Large Language Models (ICLR 2026 · similarity 0.87)
- Emotion-Coherent Reasoning for Multimodal LLMs via Emotional Rationale Verifier (AAAI 2026 · similarity 0.87)
- Adaptive Graph Learning for Multimodal Conversational Emotion Detection (AAAI 2024 · similarity 0.87)
- OV-MER: Towards Open-Vocabulary Multimodal Emotion Recognition (ICML 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
