AI Paper Atlas

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

Analyzing Compositionality-Sensitivity of NLI Models

Yixin Nie, Yicheng Wang, Mohit Bansal

AAAI 2019 · Large language models › Language, safety and interpretability · approach: Faithfulness of LLM explanations

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 compositionality-sensitivity examples selected by a confidently wrong bag-of-words modelNLI evaluation · SNLI and MNLI CS0.7accuracyMixed
Use syntactic TreeLSTM instead of sequential BiLSTMESIM / S-TLSTM · SNLI CS0.7accuracy: 55.28 -> 57.51Improved
Use syntactic TreeLSTM instead of sequential BiLSTMESIM / S-TLSTM · MNLI Matched CS0.7accuracy: 48.31 -> 48.90Improved
Use syntactic TreeLSTM instead of sequential BiLSTMESIM / S-TLSTM · MNLI Mismatched CS0.7accuracy: 48.32 -> 50.61Improved
Replace RNN layers with fully-connected layers (ablation)RSE · SNLI CS0.7accuracy: 52.73 -> 43.60Hurt
Replace RNN layers with fully-connected layers (ablation)ESIM · SNLI CS0.7accuracy: 55.28 -> 40.49Hurt
Replace RNN layers with fully-connected layers (ablation)DR-BiLSTM · SNLI CS0.7accuracy: 55.28 -> 41.38Hurt
Train with shuffled input words (ablation)ESIM · SNLI CS0.7accuracy: 55.28 -> 45.49Hurt
Train with shuffled input words (ablation)DR-BiLSTM · SNLI CS0.7accuracy: 55.28 -> 44.77Hurt
Train with subject-object-swap adversarial examples (ablation)ESIM · subject-object-swap and adjective-modifier-swap adversariescorrect-label prediction rate: 2.1 -> 99.1; 9.6 -> 26.9Mixed
Train with adjective-modifier-swap adversarial examples (ablation)ESIM · adjective-modifier-swap and subject-object-swap adversariescorrect-label prediction rate: 4.8 -> 99.6; 2.1 -> 1.0Mixed

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