atlas

Research map / Architectures and efficiency / RNNs and normalization

Sawtooth Factorial Topic Embeddings Guided Gamma Belief Network

ICML 2021 · Architectures and efficiency › RNNs and normalization · approach: Mechanistic circuit interpretability · 10 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
Use Sawtooth Connection to couple topics across layers, rather than independent learned loadings (ablation)SawETM · 20NGAC: 49.25 -> 51.25Helped
Use Sawtooth Connection to couple topics across layers, rather than independent learned loadings (ablation)SawETM · R8AC: 59.93 -> 63.82Helped
Share topic embeddings between adjacent layers through Sawtooth Connection (ablation)SawETM · 20NGNMI: 48.87 -> 50.77Helped
Share topic embeddings between adjacent layers through Sawtooth Connection (ablation)SawETM · R8NMI: 44.12 -> 45.90Helped
Increase the number of layers beyond threeWHAI · Per-heldout-word perplexityper-heldout-word perplexityNo clear effect
Increase the number of layers beyond threeDETM · Per-heldout-word perplexityper-heldout-word perplexityNo clear effect
Increase the number of layersSawETM · Per-heldout-word perplexityper-heldout-word perplexityHelped
Increase the number of layersDETM · Topic qualitytopic qualityHurt

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