Ves Stoyanov
@vesko_stHead of AI at @magicaltome. Ex-Language Researcher at @FacebookAI. Large LMs and multilingual NLP. @JHUCLSP and @Cornell alumn. https://t.co/WTSCasqDI6
Similar User
@LukeZettlemoyer
@ml_perception
@uwnlp
@HannaHajishirzi
@jaseweston
@PSH_Lewis
@riedelcastro
@MohitIyyer
@mohitban47
@royschwartzNLP
@hllo_wrld
@swabhz
@sjmielke
@seo_minjoon
@eunsolc
We're rebooting Tome to be a different company. @magicaltome is now an AI assistant for breaking into new enterprise accounts. Here's a bit on the journey we've been on…
🤔 Can LLMs truly plan ahead or are they just predicting the next word? I explore this question by testing AI models with the classic shortest path problem. Discover what this reveals about their planning abilities! 🧠 #AI #MachineLearning #LLM linkedin.com/posts/ves-sto_…
Is OpenAI string matching and filtering its outputs? Are the world's most powerful Natural Language Processing models are still gated by good old-fashioned string matching to filter copyrighted content from its output? #OpenAI linkedin.com/posts/ves-sto_…
I love perplexity.ai -- the best search engine out there! Big time rooting for @denisyarats and @perplexity_ai Give it a try if you haven't yet.
I recently joined @mbzuai as an Adjunct Professor in the NLP department. I am enjoying working with the amazing people there. The department is hiring -- would love to have more amazing colleagues, please apply!
Our NLP dept @mbzuai is hiring multiple open rank faculty positions in Speech/NLP. Some facts: * Ranked #18 in the world for NLP (csrankings.org) * 1st AI university * Friendly & supportive work environment Apply Now! apply.interfolio.com/137880
It's been an exciting 2023! Looking forward to an even more productive 2024.
200M prompts submitted. 30B words authored with AI. This year, we evolved Tome’s AI and built a complete expression toolkit to take your ideas from 0 to 1. The top 10 features of 2023 ↓
It is great to see @magicaltome on LinkedIn's Top 50 startup list! I am really excited at the opportunity to build a great product that helps people be more creative and expressive! linkedin.com/pulse/linkedin…
Playing with our most advanced AI feature @magicaltome: page generation. It uses few-shot learning to produce different ways to express my idea, the hierarchy of the content, and the design layout. I continue to be in awe of how easily my ideas can come to life on a page!
What a milestone for @magicaltome! 10 million creators!!! At Meta 10 million didn't seem like a huge number, but I have come to appreciate the dedication to customers and building something useful to get this far.
. ⋆⁺₊⋆ 10 million creators. Limitless imagination. ⋆⁺₊⋆ You energize us to keep improving Tome every day. Here’s to continuing to explore, experiment, and stay inspired.
Thanks to all the former colleagues and close collaborators that made Llama 2 a reality. The startup community really appreciates the model being available for free even for commercial use. Very excited to try it at @magicaltome!
This is huge: Llama-v2 is open source, with a license that authorizes commercial use! This is going to change the landscape of the LLM market. Llama-v2 is available on Microsoft Azure and will be available on AWS, Hugging Face and other providers Pretrained and fine-tuned…
Unfortunately, I can't be in Toronto for ACL this year. But our work on bgGLUE will be there. Please go meet @mhardalov and the rest of our co-author and hear about this interesting new resource for NLP in Bulgarian!
Hi #ACL2023NLP Toronto 🇨🇦 Tomorrow we are presenting "bgGLUE: A Bulgarian General Language Understanding Evaluation Benchmark" during the first poster session at 9am ET. Leaderboard: bgglue.github.io 📜 aclanthology.org/2023.acl-long.… 💿 github.com/bgGLUE/bgglue #NLProc 🧵⤵️ 1/6
I’m super excited to join Tome to help build product-grade AI to power storytelling. tome.app/tome/why-i-joi…
We’re thrilled to welcome ⋆⁺₊⋆ Ves Stoyanov ⋆⁺₊⋆ to Tome as our head of AI. @vesko_st spent nearly a decade as an AI leader at Meta, where he used LLMs to make online experiences safer. Ves took our questions on why he joined, his favorite podcasts, and more 🎙:
We’re thrilled to welcome ⋆⁺₊⋆ Ves Stoyanov ⋆⁺₊⋆ to Tome as our head of AI. @vesko_st spent nearly a decade as an AI leader at Meta, where he used LLMs to make online experiences safer. Ves took our questions on why he joined, his favorite podcasts, and more 🎙:
Apparently, GPT-4 turned @GaryMarcus from a "skeptic" into an alarmist. From "not impressed by large language models’ shallow, unreliable, confabulating intelligence" to "AI systems with human-competitive intelligence" in just a few months.
a big deal: @elonmusk, Y. Bengio, S. Russell, @tegmark, V. Kraknova, P. Maes, @Grady_Booch, @AndrewYang, @tristanharris & over 1,000 others, including me, have called for a temporary pause on training systems exceeding GPT-4 futureoflife.org/open-letter/pa…
United States Trends
- 1. Bengals 81,2 B posts
- 2. Chargers 70,7 B posts
- 3. McPherson 12,5 B posts
- 4. Joe Burrow 22,3 B posts
- 5. Herbert 36 B posts
- 6. #BaddiesMidwest 22,7 B posts
- 7. #MondayMotivation 10 B posts
- 8. Zac Taylor 4.466 posts
- 9. #ช็อตฟีลMV 187 B posts
- 10. MILKLOVE SHOT FEEL 224 B posts
- 11. #BoltUp 5.724 posts
- 12. #CINvsLAC 10,6 B posts
- 13. WWIII 187 B posts
- 14. 60 Minutes 34,7 B posts
- 15. TAEYEON 45,4 B posts
- 16. Carti 16,8 B posts
- 17. Harbaugh 16,5 B posts
- 18. Money Mac N/A
- 19. Scotty 10,3 B posts
- 20. Myron 11,9 B posts
Who to follow
-
Luke Zettlemoyer
@LukeZettlemoyer -
Mike Lewis
@ml_perception -
UW NLP
@uwnlp -
Hanna Hajishirzi
@HannaHajishirzi -
Jason Weston
@jaseweston -
Patrick Lewis
@PSH_Lewis -
Sebastian Riedel (@[email protected])
@riedelcastro -
Mohit Iyyer
@MohitIyyer -
Mohit Bansal
@mohitban47 -
Roy Schwartz
@royschwartzNLP -
Victor Zhong
@hllo_wrld -
Swabha Swayamdipta ✈️ EMNLP'24 🏖️
@swabhz -
Sabrina J. Mielke @ EMNLP2024
@sjmielke -
Minjoon Seo
@seo_minjoon -
Eunsol Choi
@eunsolc
Something went wrong.
Something went wrong.