AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

ProtoryNet - Interpretable Text Classification Via Prototype Trajectories

Dat Hong, Tong Wang, Stephen Baek

JMLR 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Prototype and concept dictionary models

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
allow fine-tuning of the sentence encoder (ablation)ProtoryNet · IMDB, Amazon, Yelp, Rotten Tomatoes, Hotel, Steamaccuracy: IMDB 0.871 -> 0.914; Rotten Tomatoes 0.771 -> 0.881Improved
use ProtoryNet instead of ProSeNet on long textsProtoryNet (not fine-tuned) · Yelp Reviews long textsaccuracy: 0.931 -> 0.949Improved
use ProtoryNet instead of ProSeNet on short textsProtoryNet (not fine-tuned) · Yelp Reviews short textsaccuracy: 0.943 -> 0.863Hurt
prune prototypes with a 0.5% threshold from K=200 (ablation)ProtoryNet · Amazon Reviewsaccuracy: 0.918 -> 0.918 with 14 of 200 prototypesNo clear effect
replace LSTM with logistic regression (ablation)ProtoryNet · IMDBaccuracy: 0.859 -> 0.914Improved
replace LSTM with a decision tree (ablation)ProtoryNet · Rotten Tomatoesaccuracy: 0.796 -> 0.881Improved
replace LSTM with average of sentiments (ablation)ProtoryNet · IMDBaccuracy: 0.602 -> 0.914Improved
apply sparsity transformation (one active prototype) instead of dense similarity (ablation)ProtoryNet (fine-tuned) · Rotten Tomatoesaccuracy: 0.896 -> 0.881Hurt
apply sparsity transformation instead of dense similarity (ablation)ProtoryNet (fine-tuned) · Yelp Reviewsaccuracy: 0.956 -> 0.954Hurt
add diversity and prototypicality terms with small nonzero weights (ablation)ProtoryNet · Hotel Reviews, K=100accuracyImproved
increase initial number of prototypes K (ablation)ProtoryNet · Hotel ReviewsaccuracyNo clear effect
use ProtoryNet instead of ProSeNet on multi-classProtoryNet · Consumer complaints (4-class)accuracy: 0.878 -> 0.927Improved

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