AI Paper Atlas

Research map / Benchmarks / IWSLT14 De-En

IWSLT14 De-En leaderboard

52 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on IWSLT14 De-En. Best BLEU: 39.49 (EM-Network, ICML 2023).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

IWSLT14 De-En · BLEU

52 papers, each with its best reported number for its own method; higher is better.

#BLEU ↑MethodPaperTopic
139.49EM-Network
BiBERT
EM-Network: Oracle Guided Self-distillation for Sequence Learning
ICML 2023 · Table 4
Knowledge distillation
237Data Diversification
Transformer
Data Diversification: A Simple Strategy For Neural Machine Translation
NeurIPS 2020 · Back-translation table
RNNs and normalization
336.49AB-Net
bert-base-uncased; bert-base-german-cased
Incorporating BERT into Parallel Sequence Decoding with Adapters
NeurIPS 2020 · Table 1
Attention and transformers
436.41Our model + co-teaching (BPE)
Transformer
Sequence Generation with Mixed Representations
ICML 2020 · Table 3
Language, safety and interpretability
536.11BERT-fused model
Transformer
Incorporating BERT into Neural Machine Translation
ICLR 2020 · Table: BLEU of all IWSLT tasks
Recognition and long-tail
636.1GradInit (Adam, no warmup)
Post-LN Transformer
GradInit: Learning to Initialize Neural Networks for Stable and Efficient Training
NeurIPS 2021 · Table 4
RNNs and normalization
736Transformer (GRC-cached)
Pre-Norm Transformer
Cached Transformers: Improving Transformers with Differentiable Memory Cachde
AAAI 2024 · IWSLT14/IWSLT15 translation ta
Attention and transformers
835.96Tree Transformer
Transformer
Tree-Structured Attention with Hierarchical Accumulation
ICLR 2020 · Table 1
Attention and transformers
935.9Pre-Norm PN
Transformer small
PowerNorm: Rethinking Batch Normalization in Transformers
ICML 2020 · Table 1
RNNs and normalization
1035.87DynamicConv IOT (N=4)
DynamicConv
IOT: Instance-wise Layer Reordering for Transformer Structures
ICLR 2021 · Apply on Another Structure (Dy
Attention and transformers
42 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.