AI Paper Atlas

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

LS-Tree: Model Interpretation When the Data Are Linguistic

Jianbo Chen, Michael Jordan

AAAI 2020 · Large language models › Language, safety and interpretability · approach: Linguistic encoding in LLM representations · 16 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 BERT instead of BoWSentiment classification · SST testaccuracy: 0.82% -> 0.93%Improved
Use BERT instead of BoWSentiment classification · IMDB testaccuracy: 0.94% -> 0.93%No clear effect
Use BERT instead of BoWSentiment classification · Yelp testaccuracy: 0.94% -> 0.97%Improved
Use BERT rather than CNN or LSTMModels interpreted by LS-Tree · SST, IMDB, and Yelpaverage correlation with BoW linear coefficients: SST: 0.591, 0.580 -> 0.465; IMDB: 0.442, 0.552 -> 0.321; Yelp: 0.683, 0.684 -> 0.476Mixed
Use BERT rather than CNN or LSTMModels interpreted by LS-Tree · SST and IMDB adversative relationsparent-node interaction score relative to a generic node: SST average score: 1.238; IMDB average score: 1.159Improved
Increase training dataCNN and LSTM adversative-relation modeling · Yelpparent-node interaction score relative to a generic node: Yelp average scores: CNN 0.161; LSTM 0.224; BERT 0.746Improved
Use parent-node interaction scores to distinguish adversative from non-adversative contextsModels interpreting the word while · Yelp reviewsparent-node interaction score relative to a generic node: BERT: 2.167 (adversative) vs 0.082 (non-adversative)Improved
Compare training and test interaction-score varianceOverfitted neural models · Training versus test dataaverage variance of interaction scoresImproved

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