List your company

Dialogue State Tracking on SGD

80.68Joint GAReacTOD
23.188838.114453.0467.9656Nov 18, 2020Oct 18, 2021Sep 18, 2022Aug 18, 2023Jul 18, 2024Jun 17, 2025May 18, 2026

Evaluation results

MethodLinks
ReacTODModel=Claude-Opus-4.6, predicts domains end-to-end=true, impractical for production latency=true2026.0580.68—
ReacTODModel=Qwen3-32B, predicts domains end-to-end=true2026.0564.09—
ReacTODModel=gpt-oss-20B, predicts domains end-to-end=true2026.0562.92—
ReacTODModel=Qwen3-8B, predicts domains end-to-end=true2026.0557.31—
ReacTODModel=Gemma3-12B, predicts domains end-to-end=true2026.0555.58—
SRPModel=Claude-Opus-4.6, gold domain labels=true, per-domain isolated sessions=true, reproduced=true2026.0545.2—
Seq2Seq-DUBackbone=BERT-Base, Uncased2020.1130.191
FastSGT2020.1129.290.3
SGD-baseline2020.1125.490.6