Marcin Junczys-Dowmunt (Marian NMT)
marian-nmt.bsky.social
NLP. NMT. Main author of Marian NMT. Research Scientist at Microsoft Translator. https://marian-nmt.github.io
👤 Joined on 25 November 2024
167
Followers
134
Following
16
Total Posts
📜 Kullanıcı Geçmişi
- 25 Kas 2024, 06:12
🎉 Bluesky'e katıldı - İlk kullanıcı adı:
@marian-nmt.bsky.social
🚫 Blokladığı Kullanıcılar
❤️ Beğendiği Gönderiler
Distillation is violation of Terms of Service. That is not Intellectual Property theft. It’s up to the company to enforce TOS, not the government. Outputs of AI stems are not copyrightable either. Oh, and OpenAI might be violating copyright anyway.
Show Post
Just 10 days after o1's public debut, we’re thrilled to unveil the open-source version of the technique behind its success: scaling test-time compute By giving models more "time to think," Llama 1B outperforms Llama 8B in math—beating a model 8x its size. The full recipe is open-source!
Show Post
Lol that is a nice word. It sadly does not exist. I meant "tool", but the predictive keyboard thought that would be too easy
Show PostIs it that bad? I like knowing these blobs are immutable (aside from deletion), it forces me to think more before posting (not really)
Show Post
Wrote up some notes on Microsoft's new Phi-4 LLM. They trained it on a LOT of synthetic data, and the details of how and why they did that are really interesting. https://simonwillison.net/2024/Dec/15/phi-4-technical-report/
Show Post