Understanding the Difficulty of Training Transformers
-
Updated
May 31, 2022 - Python
#
wmt
Here are 15 public repositories matching this topic...
Tensorflow Implementation of "Theory and Experiments on Vector Quantized Autoencoders"
-
Updated
Feb 27, 2019 - Python
From-scratch Transformer (Vaswani 2017) for zh-en translation. Pure PyTorch, SentencePiece BPE, token-based batching, FP16 training, beam search. Trained on WMT17 (~25M pairs).
nlp deep-learning machine-translation pytorch transformer seq2seq attention-is-all-you-need chinese-english wmt
-
Updated
May 11, 2026 - Python
The ALLIES Lifelong Learning Machine Translation baseline system's repository
machine-translation neural-machine-translation nmt interactive-learning active-learning incremental-learning lifelong-learning continual-learning wmt lifelong-machine-learning
-
Updated
Oct 7, 2022 - Jupyter Notebook
Hunyuan Machine Translation — 🥇 1st in 30/31 WMT25 language pairs
-
Updated
Sep 8, 2025 - Python
Lara 3 translation outputs on selected WMT25 General MT content
-
Updated
Jul 31, 2026 - Python
Repo for our attempt at the Shared Task (Quality Estimation) WMT2020. http://www.statmt.org/wmt20/quality-estimation-task.html
-
Updated
Mar 7, 2021 - Jupyter Notebook
Drawer Material Design UI in React Native.
-
Updated
Jan 22, 2018 - JavaScript
The Suboptimal Quality of WMT Test Sets and Their Impact on HumanParity
-
Updated
May 13, 2024 - Python
React Native Starter Design for application starter design using react native.
-
Updated
Jan 19, 2018 - JavaScript
Pseudo-reference construction for the WMT26 General MT task: generate-score-select-repair pipeline, per-document provenance, and the analysis behind 'In the Blind' (WMT 2026).
-
Updated
Sep 12, 2026 - Python
Digests map servers. Point it at an ArcGIS REST, WMS, WFS or WMTS endpoint and get the whole layer out as GeoParquet - natively, on macOS, read-only, resumable.
macos arcgis geospatial wms gis geoserver data-extraction parquet esri wfs ogs arcgis-server ows featureserver wmt arcgis-rest-api duckdb geoparquet
-
Updated
Oct 2, 2026 - Swift
TamilLingBench: English→Tamil machine translation benchmark for obligatory verb agreement (rationality, gender, number). Tests whether MT evaluation metrics such as COMET, xCOMET, MetricX and CometKiwi catch wrong Tamil agreement. Code, data and model outputs for the WMT 2026 paper "Obligatory Slots".
nlp benchmark translation morphology machine-translation comet computational-linguistics tamil quality-estimation finite-state-transducer evaluation-metrics interpretability wmt mt-evaluation dravidian-languages llm-evaluation activation-patching metricx challenge-set english-tamil
-
Updated
Sep 12, 2026 - Python
Add this topic to your repo
To associate your repository with the wmt topic, visit your repo's landing page and select "manage topics."