Papers with Code · CC BY-SA 4.0

Dialogue

319 평가 기록 72 datasets

Dialogue is notoriously hard to evaluate. Past approaches have used human evaluation.

하위 태스크

VD Visual Dialog 8 데이터셋 · 98 rows DU Dialogue Understanding 19 데이터셋 · 88 rows DS Dialogue State Tracking 7 데이터셋 · 39 rows DG Dialogue Generation 14 데이터셋 · 24 rows DA Dialogue Act Classification 5 데이터셋 · 22 rows TD Task-Oriented Dialogue Systems 6 데이터셋 · 16 rows CD Conversation Disentanglement 3 데이터셋 · 11 rows ED End-To-End Dialogue Modelling 2 데이터셋 · 10 rows PD Problem-Solving Deliberation 1 데이터셋 · 3 rows ER Empathetic Response Generation 1 데이터셋 · 2 rows GD Goal-Oriented Dialog 1 데이터셋 · 1 rows IE Interactive Evaluation of Dialog 1 데이터셋 · 1 rows MG Multimodal GIF Dialog 1 데이터셋 · 1 rows PD Persona Dialogue in Story 1 데이터셋 · 1 rows WD Workflow Discovery 1 데이터셋 · 1 rows DL Dialog Learning 0 데이터셋 · 0 rows DI Dialogue Interpretation 0 데이터셋 · 0 rows DM Dialogue Management 0 데이터셋 · 0 rows GD Goal-Oriented Dialogue Systems 0 데이터셋 · 0 rows SC Short-Text Conversation 0 데이터셋 · 0 rows TD Task-Completion Dialogue Policy Learning 0 데이터셋 · 0 rows