atlas

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

Recurrent Hierarchical Topic-Guided RNN for Language Generation

ICML 2020 · Large language models › Language, safety and interpretability · approach: Controlled text generation · 13 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
add recurrent sentence-context topic transitions to the hierarchical topic model (ablation)GBN-RNN · APNEWS, IMDB, and BNC testperplexity: 44.35 -> 42.71; 54.53 -> 51.36; 80.25 -> 79.13Helped
swap the locations of topic vectors at layers 3 and 1 (ablation)rGBN-RNN · APNEWS, IMDB, and BNC testperplexity: 42.71 -> 43.55; 51.36 -> 53.28; 79.13 -> 81.12Hurt
increase the recurrent topic-guided model from one to three layers (ablation)rGBN-RNN · APNEWS testperplexity: 46.35 -> 43.26 -> 42.71Helped
increase the recurrent topic-guided model from one to three layers (ablation)rGBN-RNN · IMDB testperplexity: 55.76 -> 53.82 -> 51.36Helped
increase the recurrent topic-guided model from one to three layers (ablation)rGBN-RNN · BNC testperplexity: 81.94 -> 80.25 -> 79.13Helped

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