WMT26
PulseAugur coverage of WMT26 — every cluster mentioning WMT26 across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
UvA-MT leverages LLMs for WMT26 translation task with novel sampling and voting
Researchers from UvA-MT have detailed their approach for the WMT26 General Machine Translation shared task, focusing on the unconstrained track across 23 translation directions. Their method heavily utilizes Large Langu…
-
Cohere releases North Small Translate model, benchmarked on WMT26
Cohere has released a new translation model called North Small Translate, which was benchmarked using the WMT26 dataset. The company highlighted that the WMT26 benchmarks were released after the model's creation, preven…
-
University of Melbourne tackles low-resource Pacific creole translation
Researchers from the University of Melbourne have developed a machine translation system for Pacific creole languages, including Tok Pisin, Bislama, and Solomon Pijin. Their approach involves pre-training on a large, do…
-
New method creates pseudo-references for machine translation evaluation
Researchers have developed a novel method for creating pseudo-references for machine translation evaluation, particularly for language pairs lacking human-generated references. This approach involves using multiple MT m…
-
ESTS details WMT26 model compression using GPT-OSS-20B and GPT-5.1
Researchers from ESTS have detailed their submissions to the WMT26 Model Compression Shared Task, focusing on English-to-Simplified Chinese and English-to-Egyptian Arabic translation. Their approach involved pruning exp…
-
Cohere releases 218B MoE translation model, North Small Translate
Cohere has quietly released North Small Translate, a 218-billion-parameter Mixture-of-Experts (MoE) model specifically designed for machine translation. This sparse model, with 25 billion active parameters per token, su…
-
AI model improves translation accuracy by focusing on difficult glossary terms
Researchers have developed a novel approach for terminology-aware machine translation, focusing on "hard examples" where a model's output differs from a prescribed glossary. By filtering training data to include only th…
-
University of Florida Gators use retrieval-augmented translation for low-resource languages
Researchers from the University of Florida Gators have developed a novel approach for low-resource machine translation, specifically targeting North-Eastern Indian languages. Their system, submitted to the WMT26 shared …