AI Paper Atlas

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

Non-Linear Similarity Learning for Compositionality

Masashi Tsubaki, Kevin Duh, Masashi Shimbo, Yuji Matsumoto

AAAI 2016 · Label-efficient and robust learning › Recognition and long-tail · approach: Multilingual language representation pretraining · 15 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
replace cosine similarity with polynomial kernel (ablation)SUM + STR SEM · SICK test (SemEval 2014 Task 1)Pearson r: 0.7647 -> 0.8480Improved
replace cosine similarity with RBF kernel (ablation)SUM (bag of words) · SICK test (SemEval 2014 Task 1)Pearson r: 0.7588 -> 0.8339Improved
add STR SEM structure features to SUM (ablation)Poly kernel · SICK test (SemEval 2014 Task 1)Pearson r: 0.8332 -> 0.8480Improved
add STR POS structure features to SUM (ablation)Poly kernel · SICK test (SemEval 2014 Task 1)Pearson r: 0.8332 -> 0.8301No clear effect
add STR POS structure features to SUM (ablation)Cosine kernel · SICK test (SemEval 2014 Task 1)Pearson r: 0.7588 -> 0.7510No clear effect
use tanh non-linear composition instead of linear (ablation)Poly (SUM + STR SEM) · SICK test (SemEval 2014 Task 1)Pearson r: 0.8480 -> 0.8392Hurt
use tanh non-linear composition instead of linear (ablation)RBF (SUM + STR SEM) · SICK test (SemEval 2014 Task 1)Pearson r: 0.8447 -> 0.8393Hurt
use tanh non-linear composition instead of linear (ablation)Cosine (SUM + STR SEM) · SICK test (SemEval 2014 Task 1)Pearson r: 0.7647 -> 0.7717Improved
polynomial kernel instead of cosine on MSE (ablation)SUM · SICK test (SemEval 2014 Task 1)MSE: 0.4820 -> 0.3205Improved
initialize word vectors with LSA instead of GloVe or NLM (ablation)Poly (SUM) · SICK test (SemEval 2014 Task 1)Pearson r (Figure 2)Hurt
random initialization instead of pre-trained GloVe/NLM (ablation)Poly (SUM) · SICK test (SemEval 2014 Task 1)Pearson r (Figure 2)No clear effect
proposed model vs feature-engineered SemEval systemsPoly (SUM + STR SEM) · SICK test (SemEval 2014 Task 1)Pearson r: 0.8280 -> 0.8480Improved

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