AI Paper Atlas

Research map / Benchmarks / IWSLT14 En-De

IWSLT14 En-De leaderboard

14 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on IWSLT14 En-De. Best BLEU: 31.88 (HiCTL Base, ICLR 2021).

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 En-De · BLEU

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

#BLEU ↑MethodPaperTopic
131.88HiCTL Base
XLM-R
On Learning Universal Representations Across Languages
ICLR 2021 · Low-resource MT table
Recognition and long-tail
231.8EM-Network
BiBERT
EM-Network: Oracle Guided Self-distillation for Sequence Learning
ICML 2023 · Table 4
Knowledge distillation
330.6Data Diversification
Transformer
Data Diversification: A Simple Strategy For Neural Machine Translation
NeurIPS 2020 · Back-translation table
RNNs and normalization
430.45BERT-fused model
Transformer
Incorporating BERT into Neural Machine Translation
ICLR 2020 · Table: BLEU of all IWSLT tasks
Recognition and long-tail
529.93Our model + co-teaching (BPE)
Transformer
Sequence Generation with Mixed Representations
ICML 2020 · Table 3
Language, safety and interpretability
629.52IOT (N=4)
Transformer
IOT: Instance-wise Layer Reordering for Transformer Structures
ICLR 2021 · Table: BLEU scores of IOT on e
Attention and transformers
729.47Tree Transformer
Transformer
Tree-Structured Attention with Hierarchical Accumulation
ICLR 2020 · Table 1
Attention and transformers
829.4NPRF enc-dec with RPEStable, Fast and Accurate: Kernelized Attention with Relative Positional Encoding
NeurIPS 2021 · Machine translation BLEU table
Attention and transformers
929.07Tied TransformerTied Transformers: Neural Machine Translation with Shared Encoder and Decoder
AAAI 2019 · Table 2
Attention and transformers
1028.91MoDEMoDE: A Mixture-of-Experts Model with Mutual Distillation among the Experts
AAAI 2024 · Table: Comparisons on the benc
Mixture of experts
4 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.