AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Coherent Dialogue with Attention-Based Language Models

Hongyuan Mei, Mohit Bansal, Matthew Walter

AAAI 2017 · Architectures and efficiency › Attention and transformers · approach: State space and recurrent sequence models · 92 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
Use dynamic attention over the growing dialogue historyRNN language model · MovieTriples testPPL: 27.09 -> 25.52Improved
Use dynamic attention over the growing dialogue historyRNN language model · Ubuntu Troubleshootrecall@1: 11% -> 17%Improved
Model the full dialogue as a language-model sequence rather than a source-target pairDialogue modeling · MovieTriplesPPL: 35.10 -> 27.81Improved
Model the full dialogue as a language-model sequence rather than a source-target pairDialogue modeling · Ubuntu TroubleshootPPL: 104.61 -> 56.16Improved
Add LDA topic features without reranking (ablation)A-RNN · Ubuntu Troubleshoot testBLEU: 0.3713 -> 0.3128Hurt
Rerank continuations using LDA topic matchingA-RNN · Ubuntu Troubleshoot testBLEU: 0.3713 -> 0.4279Improved
Add LDA topic features and rerank using LDA topic matchingA-RNN · Ubuntu Troubleshoot testBLEU: 0.3128 -> 0.3971Improved
Add LDA topic features without reranking (ablation)A-RNN · Ubuntu Troubleshoot devBLEU: 0.2702 -> 0.2908Improved
Use dynamic attention instead of a plain RNNRNN language model · Ubuntu TroubleshootDistinct-1: 0.0007 -> 0.0104Improved
Use dynamic attention instead of a plain RNNRNN language model · MovieTriplesDistinct-1: 0.0004 -> 0.0028Improved

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