Research map / Large language models / Language, safety and interpretability
Filling the Gap of Utterance-aware and Speaker-aware Representation for Multi-turn Dialogue
Longxiang Liu, Zhuosheng Zhang, Hai Zhao, Xi Zhou, Xiang Zhou
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 utterance-aware and speaker-aware decoupling and fusion | ELECTRA · MuTual | R4@1: 0.900 -> 0.916 | Improved |
| add utterance-aware and speaker-aware decoupling and fusion | ELECTRA · Ubuntu | R10@1: 0.845 -> 0.866 | Improved |
| add utterance-aware and speaker-aware decoupling and fusion | ELECTRA · Douban | R10@1: 0.287 -> 0.325 | Improved |
| add utterance-aware and speaker-aware decoupling and fusion | Chinese ELECTRA · ECD | R10@1: 0.607 -> 0.639 | Improved |
| add utterance-aware masking only (ablation) | ELECTRA · MuTual dev | R@1: 0.903 -> 0.909 | Improved |
| add speaker-aware masking only (ablation) | ELECTRA · MuTual dev | R@1: 0.903 -> 0.913 | Improved |
| use BiGRU over utterance representations instead of max-pooling (ablation) | MDFN · MuTual dev | R@1: 0.909 -> 0.923 | Improved |
| use one decoupling block instead of two (ablation) | MDFN · MuTual dev | R@1: 0.909 -> 0.923 | Improved |
| add gating to fuse complementary channel information (ablation) | MDFN · MuTual dev | R@1: 0.914 -> 0.923 | Improved |
| include original representations in gate calculation (ablation) | MDFN · MuTual dev | R@1: 0.918 -> 0.923 | Improved |
| use max-pooling instead of mean-pooling for within-utterance aggregation (ablation) | MDFN · MuTual dev | R@1: 0.911 -> 0.923 | Improved |
| use max-pooling instead of CNN aggregation (ablation) | MDFN · MuTual dev | R@1: 0.916 -> 0.923 | Improved |
Closest papers by meaning
- Aligning Spoken Dialogue Models from User Interactions (ICML 2025 · similarity 0.87)
- F2RVLM: Boosting Fine-grained Fragment Retrieval for Multi-Modal Long-form Dialogue with Vision Language Model (AAAI 2026 · similarity 0.87)
- SD-Eval: A Benchmark Dataset for Spoken Dialogue Understanding Beyond Words (NeurIPS 2024 · similarity 0.86)
- TA&AT: Enhancing Task-Oriented Dialog with Turn-Level Auxiliary Tasks and Action-Tree Based Scheduled Sampling (AAAI 2024 · similarity 0.86)
- NTPP: Generative Speech Language Modeling for Dual-Channel Spoken Dialogue via Next-Token-Pair Prediction (ICML 2025 · similarity 0.85)
- An Efficient Self-Learning Framework For Interactive Spoken Dialog Systems (ICML 2024 · similarity 0.85)
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.
