Research map / Label-efficient and robust learning / Recognition and long-tail
Contrast and Generation Make BART a Good Dialogue Emotion Recognizer
Shimin Li, Hang Yan, Xipeng Qiu
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 |
|---|---|---|---|
| Add response generation as an auxiliary task (ablation) | CoG-BART · MELD | Weighted-Avg-F1: 64.26 -> 64.81 | Improved |
| Add response generation as an auxiliary task (ablation) | CoG-BART · IEMOCAP | Weighted-Avg-F1: 64.74 -> 66.18 | Improved |
| Add supervised contrastive loss (ablation) | CoG-BART · MELD | Weighted-Avg-F1: 64.28 -> 64.81 | Improved |
| Add supervised contrastive loss (ablation) | CoG-BART · IEMOCAP | Weighted-Avg-F1: 64.23 -> 66.18 | Improved |
| Prepend speaker labels to utterances (ablation) | CoG-BART · MELD | Weighted-Avg-F1: 64.14 -> 64.81 | Improved |
| Prepend speaker labels to utterances (ablation) | CoG-BART · IEMOCAP | Weighted-Avg-F1: 55.41 -> 66.18 | Improved |
| Add both response generation and supervised contrastive loss (ablation) | CoG-BART · MELD | Weighted-Avg-F1: 63.57 -> 64.81 | Improved |
| Add both response generation and supervised contrastive loss (ablation) | CoG-BART · IEMOCAP | Weighted-Avg-F1: 62.90 -> 66.18 | Improved |
| Add supervised contrastive loss and speaker label prepending (ablation) | CoG-BART · MELD | Weighted-Avg-F1: 63.72 -> 64.81 | Improved |
| Add supervised contrastive loss and speaker label prepending (ablation) | CoG-BART · IEMOCAP | Weighted-Avg-F1: 54.83 -> 66.18 | Improved |
| Add response generation and speaker label prepending (ablation) | CoG-BART · MELD | Weighted-Avg-F1: 64.02 -> 64.81 | Improved |
| Add response generation and speaker label prepending (ablation) | CoG-BART · IEMOCAP | Weighted-Avg-F1: 54.95 -> 66.18 | Improved |
Closest papers by meaning
- Eliciting Positive Emotion through Affect-Sensitive Dialogue Response Generation: A Neural Network Approach (AAAI 2018 · similarity 0.86)
- Knowledge Bridging for Empathetic Dialogue Generation (AAAI 2022 · similarity 0.86)
- SD-Eval: A Benchmark Dataset for Spoken Dialogue Understanding Beyond Words (NeurIPS 2024 · similarity 0.86)
- Emotion Rendering for Conversational Speech Synthesis with Heterogeneous Graph-Based Context Modeling (AAAI 2024 · similarity 0.85)
- Adaptive Graph Learning for Multimodal Conversational Emotion Detection (AAAI 2024 · similarity 0.85)
- Dual-View Learning for Conversational Emotion Recognition Through Context and Emotion-Shift Modeling (AAAI 2025 · similarity 0.84)
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.
