AI Paper Atlas

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

Sequence Generation with Mixed Representations

Lijun Wu, Shufang Xie, Yingce Xia, Yang Fan, Jian-Huang Lai, Tao Qin, Tieyan Liu

ICML 2020 · Large language models › Language, safety and interpretability · approach: Controlled text generation

Read the paper

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
mixed representations from BPE and SP tokenizers (two encoders with mix-attention, two decoders)Transformer 256 backbone · IWSLT14 translation (12 directions)BLEU: more than 1.0 BLEU on De→En, En→Ro, Pt-br→En vs Transformer 256Improved
mixed representations (no co-teaching) vs Transformer 512 baselineTransformer · IWSLT14 En→De and De→EnBLEU: about 0.5 points for most tasksImproved
co-teaching algorithm added on top of mixed representationsmixed-representation model · IWSLT14 translation (12 directions)BLEU: more than 1.0 on 11 of 12 tasks; over 1.5 on En→TrImproved
stacked representations instead of additive mixing of self-attention and mix-attention (ablation)mixed-representation encoder · IWSLT14 En↔DeBLEU: worse than additive mixed representations on all four En↔De entries (e.g., En→De BPE 28.53 vs 28.96)Hurt
co-teaching with a single tokenizer (forward-translated data, BPE only) (ablation)Transformer 512 · IWSLT14 En↔DeBLEU: En→De BPE 28.80 to 29.67; De→En BPE 34.84 to 36.10Improved
two-tokenizer co-teaching vs single-tokenizer co-teaching (ablation)mixed-representation model · IWSLT14 En→De SPBLEU: 29.93 (two tokenizers) vs 29.28 (one tokenizer)Improved
replace SentencePiece with WordPiece in the mixed model (BPE+WP) (ablation)mixed-representation model · IWSLT14 En↔DeBLEU: En→De 28.98 (BPE) / 29.15 (WP) vs 28.80 baseline; De→En 35.23/35.24 vs 34.84Improved
mixed representations + co-teaching on abstractive summarizationTransformer 256 backbone · GigawordROUGE-1: 36.0 vs 35.5 (Transformer 512)Improved

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