AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

A Simple Language Model for Task-Oriented Dialogue

Ehsan Hosseini-Asl, Bryan McCann, Chien-Sheng Wu, Semih Yavuz, Richard Socher

NeurIPS 2020 · Large language models › Language, safety and interpretability · approach: Faithfulness of LLM explanations · 158 citations (OpenAlex)

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 changedWhereMeasuredEffect
single causal LM for DST, action and response (SimpleTOD)GPT-2 · MultiWOZ 2.1 DSTjoint accuracy: 55.3 -> 55.72Improved
SimpleTOD vs DAMD with oracle DBGPT-2 · MultiWOZ 2.0 end-to-endcombined score: 85 -> 87.66Improved
use oracle DB search results as input (ablation)SimpleTOD · MultiWOZ 2.1 end-to-endcombined score: 92.98 -> 87.36Hurt
use dynamic DB search results (ablation)SimpleTOD · MultiWOZ 2.1 end-to-endcombined score: 92.98 -> 90.24Hurt
add end-of-segment and user/system tokens (ablation)SimpleTOD · MultiWOZ 2.1combined score: 26.73 -> 92.98Improved
add user/system tokens (ablation)SimpleTOD with end tokens · MultiWOZ 2.1joint accuracy: 21.5 -> 55.76Improved
add end-of-segment tokens (ablation)SimpleTOD with user/system tokens · MultiWOZ 2.1combined score: 66.87 -> 92.98Improved
initialize from pretrained DistilGPT2 (ablation)6-layer SimpleTOD · MultiWOZ 2.1joint accuracy: 16.45 -> 54.54Improved
initialize from pretrained GPT2 (ablation)12-layer SimpleTOD · MultiWOZ 2.1combined score: 59.65 -> 90.75Improved
train three separate LMs per sub-task instead of one (ablation)SimpleTOD · MultiWOZcombined scoreNo clear effect
nucleus sampling instead of greedy decoding (ablation)SimpleTOD · MultiWOZoverall performanceHurt
12 layers GPT2 vs 6 layers DistilGPT2 (ablation)SimpleTOD · MultiWOZ 2.1combined score: 92.98 -> 90.75Mixed

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy