AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

MorphTE: Injecting Morphology in Tensorized Embeddings

Guobing Gan, Peng Zhang, Sunzhu Li, Xiuqing Lu, Benyou Wang

NeurIPS 2022 · Architectures and efficiency › RNNs and normalization · approach: Non-autoregressive translation · 0 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 MorphTE instead of original embeddings at about 20× compressionTransformer · IWSLT'14 De-En and OPUS-100 En-It, En-Es, En-RuBLEU: 34.5 -> 34.9; 32.9 -> 32.9; 39.1 -> 39.1; 31.6 -> 31.9Mixed
Use MorphTE instead of Word2ket at about 20× compressionTransformer · De-En, En-It, En-Es, En-Ru translation datasetsBLEU: +0.7, +0.6, 0.0, +0.6 BLEUImproved
Use MorphTE instead of original embeddingsRE2 · WikiQAMAP: 0.6798 -> 0.6834No clear effect
Use MorphTE instead of original embeddingsRE2 · WikiQAMRR: 0.6970 -> 0.7051Improved
Use MorphTE instead of original embeddingsRE2 · SNLIAccuracy: 0.8492 -> 0.8497No clear effect
Set MorphTE order to 3 rather than 2 or 4 (ablation)MorphTE · IWSLT'14 De-EnBLEU: 34.4 -> 34.9; 34.4 -> 34.9Improved
Replace morphology-based segmentation with randomSeg (ablation)MorphTE · De-En, En-It, En-Es, En-Ru translation datasetsBLEU: 34.9 -> 33.8; 32.9 -> 31.7; 39.1 -> 38.6; 31.9 -> 30.4Hurt
Replace Word2ket's individual vectors with randomly shared vectors rather than morpheme-based sharing (ablation)Word2ket · De-En, En-It, En-Es, En-Ru translation datasetsBLEU: 34.2 -> 34.0; 32.3 -> 32.0; 39.1 -> 38.3; 31.3 -> 30.6Hurt
Increase embedding size while using MorphTE (ablation)Transformer · IWSLT'14 De-En, embedding sizes 216, 512, 1000BLEU: 34.1 -> 34.2; 34.5 -> 34.9; 32.8 -> 33.4Mixed

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