Justin Barton (@all_your_bayes) 's Twitter Profile
Justin Barton

@all_your_bayes

Machine learning ∩ drug discovery, protein engineering. distributed systems, modern datamancy

ID: 3249427495

calendar_today19-06-2015 02:23:10

29 Tweet

112 Followers

350 Following

NASA (@nasa) 's Twitter Profile Photo

Our Mars Odyssey orbiter phoned home, relaying news from NASA InSight indicating its solar panels are open & collecting sunlight on the Martian surface. Also in the dispatch: this snapshot from the lander's arm showing the instruments in their new home: go.nasa.gov/2FIr1pU

Our Mars Odyssey orbiter phoned home, relaying news from <a href="/NASAInSight/">NASA InSight</a> indicating its solar panels are open &amp; collecting sunlight on the Martian surface. Also in the dispatch: this snapshot from the lander's arm showing the instruments in their new home: go.nasa.gov/2FIr1pU
Jinwoo Leem (@ideasbyjin) 's Twitter Profile Photo

Alchemab Therapeutics's tech team are delighted to present our latest pre-print on an antibody-specific language model, AntiBERTa, on bioRxiv! #antibodies #transformers #biotech (1/5) biorxiv.org/content/10.110…

Pieter Meysman (@chevaliersf) 's Twitter Profile Photo

The IMMREP23 Kaggle #TCR-#epitope specificity #MachineLearning #prediction challenge is live now! Spread the word! airr_community @iRecRepPapers kaggle.com/competitions/t…

Pieter Meysman (@chevaliersf) 's Twitter Profile Photo

The IMMREP23 challenge has its final leader board! It was a thrilling competition with many upsets and passionate participants. Huge thanks to Justin Barton who oversaw the kaggle competition! kaggle.com/competitions/t…

Jinwoo Leem (@ideasbyjin) 's Twitter Profile Photo

We'll be at #NeurIPS2023 on Friday & Saturday! Check out our work and models on🤗 Enhancing antibody language models with structural information mlsb.io/papers_2023/En… huggingface.co/alchemab Great team effort Justin Barton Jake Galson #antibodies #multimodal #huggingface

Katie Link (@katieelink) 's Twitter Profile Photo

2️⃣ AntiBERTa2: enhancing antibody language models with structural information 💉🧪 by Jinwoo Leem, Justin Barton, Jake Galson et al Alchemab Therapeutics! Paper: mlsb.io/papers_2023/En… Model: huggingface.co/alchemab/antib…

Jinwoo Leem (@ideasbyjin) 's Twitter Profile Photo

🦅 the FAbCon has landed 🦅 Our #generative #antibody #LLM available on huggingface.co/alchemab 🐣 FAbCon-small (144M) 🐥 FAbCon-medium (297M) 🦅 FAbCon-large (2.4B) Team effort Alchemab Therapeutics Justin Barton Aretas Gaspariūnas Henry Farmery Jake Galson @JorgeNDias 🎉 biorxiv.org/content/10.110…

Biology+AI Daily (@biologyaidaily) 's Twitter Profile Photo

A generative foundation model for antibody sequence understanding 1. FAbCon is a groundbreaking generative antibody-specific language model with 2.4 billion parameters, showcasing superior performance in antigen-binding prediction and antibody design tasks compared to previous

A generative foundation model for antibody sequence understanding

1. FAbCon is a groundbreaking generative antibody-specific language model with 2.4 billion parameters, showcasing superior performance in antigen-binding prediction and antibody design tasks compared to previous
Jay Alammar (@jayalammar) 's Twitter Profile Photo

أرحِب! Command R7B Arabic نموذج لغة عربي جديد متفوق، بالحجم الخفيف (7B) يتميز في دعمه لتطبيقات الإنشاء المعزز بالإسترجاع (Retrieval-Augmented Generation RAG) سياق المدخلات يتسع لمئات الصفحات (128K Context Size) مفتوح الأوزان ومتاح للتحميل الآن هذه الفئة من النماذج الصغيرة مهمة

أرحِب!
Command R7B Arabic
 نموذج لغة عربي جديد متفوق، بالحجم الخفيف (7B)
يتميز في دعمه لتطبيقات الإنشاء المعزز بالإسترجاع (Retrieval-Augmented Generation RAG)
 سياق المدخلات يتسع لمئات الصفحات (128K Context Size)
مفتوح الأوزان ومتاح للتحميل الآن

هذه الفئة من النماذج الصغيرة مهمة
Graham Neubig (@gneubig) 's Twitter Profile Photo

Some people have said that OpenAI achieved state of the art results on the SWE-Bench Verified leaderboard with their codex model, but that's actually not quite correct, no matter how you measure it. A quick 🧵

Some people have said that OpenAI achieved state of the art results on the SWE-Bench Verified leaderboard with their codex model, but that's actually not quite correct, no matter how you measure it.

A quick 🧵