SOTAVerified

Intent Discovery

Given a set of labelled and unlabelled utterances, the idea is to identify existing (known) intents and potential (new intents) intents. This method can be utilised in conversational system setting.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 31–40 of 42 papers

TitleStatusHype
Towards Real-world Scenario: Imbalanced New Intent DiscoveryCode0
Decoupling Pseudo Label Disambiguation and Representation Learning for Generalized Intent DiscoveryCode0
Integration of Old and New Knowledge for Generalized Intent Discovery: A Consistency-driven Prototype-Prompting FrameworkCode0
Watch the Neighbors: A Unified K-Nearest Neighbor Contrastive Learning Framework for OOD Intent DiscoveryCode0
Benchmark Data and Evaluation Framework for Intent Discovery Around COVID-19 Vaccine HesitancyCode0
One Stone, Four Birds: A Comprehensive Solution for QA System Using Supervised Contrastive LearningCode0
Open Intent Discovery through Unsupervised Semantic Clustering and Dependency ParsingCode0
Pseudo-Label Enhanced Prototypical Contrastive Learning for Uniformed Intent DiscoveryCode0
Continual Generalized Intent Discovery: Marching Towards Dynamic and Open-world Intent RecognitionCode0
A Diffusion Weighted Graph Framework for New Intent DiscoveryCode0
Show:102550
← PrevPage 4 of 5Next →
All datasetsATISPersian-ATISSNIPS

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1k-PCA + HDBSCANARI74.94—Unverified
#ModelMetricClaimedVerifiedStatus
1k-PCA + HDBSCANARI11.97—Unverified
#ModelMetricClaimedVerifiedStatus
1k-PCA + HDBSCANARI59.23—Unverified