Research map / Vision-language and multimodal / Human and embodied modeling
Learning Emotion Representations From Verbal and Nonverbal Communication
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| add SAP to the vanilla model (ablation) | EmotionCLIP · BoLD | mAP: 21.97 -> 22.28 | Helped |
| add SAP to the vanilla model (ablation) | EmotionCLIP · BoLD | AUC: 68.85 -> 69.06 | Helped |
| add SAAM to the vanilla model (ablation) | EmotionCLIP · BoLD | mAP: 21.97 -> 21.53 | Hurt |
| add SAAM to the vanilla model (ablation) | EmotionCLIP · BoLD | R²: 0.130 -> 0.137 | Helped |
| add SNCE alongside SAP (ablation) | EmotionCLIP · BoLD | mAP: 22.28 -> 22.51 | Helped |
| add SNCE alongside SAP (ablation) | EmotionCLIP · BoLD | AUC: 69.06 -> 69.30 | Helped |
| replace the temporal encoder with mean pooling (ablation) | EmotionCLIP · BoLD | mAP: 22.51 -> 21.17 | Hurt |
| use random initialization rather than pre-trained weights for both encoders (ablation) | EmotionCLIP · BoLD | mAP: 22.51 -> 11.02 | Hurt |
| freeze both encoders initialized with pre-training weights (ablation) | EmotionCLIP · BoLD | mAP: 22.51 -> 18.43 | Hurt |
| use EmotionCLIP rather than the strongest supervised visual baseline | BoLD · BoLD | mAP: 19.29 -> 22.51 | Helped |
| use EmotionCLIP rather than Affect2MM | MovieGraphs · MovieGraphs test | accuracy: 30.58 -> 32.35 | Helped |
| use EmotionCLIP rather than the strongest visual supervised baseline | Emotic · Emotic | mAP: 35.48 -> 32.91 | Hurt |
Closest papers by meaning
- AffectGPT: A New Dataset, Model, and Benchmark for Emotion Understanding with Multimodal Large Language Models (ICML 2025 · similarity 0.91)
- AffectGPT-RL: Revealing Roles of Reinforcement Learning in Open-Vocabulary Emotion Recognition (NeurIPS 2026 · similarity 0.89)
- Customizing Visual Emotion Evaluation for MLLMs: An Open-vocabulary, Multifaceted, and Scalable Approach (ICLR 2026 · similarity 0.88)
- OV-MER: Towards Open-Vocabulary Multimodal Emotion Recognition (ICML 2025 · similarity 0.88)
- Emotion-Coherent Reasoning for Multimodal LLMs via Emotional Rationale Verifier (AAAI 2026 · similarity 0.88)
- $E^3$: Exploring Embodied Emotion Through A Large-Scale Egocentric Video Dataset (NeurIPS 2024 · 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.