Antonio Valerio Miceli Barone
@AVMiceliBaroneML / NLP School of Informatics, The University of Edinburgh
Similar User
@p_nawrot
@nikita_moghe
@nsaphra
@sleepinyourhat
@uwnlp
@tomsherborne
@harshitj__
@machelreid
@hardy_qr
@RicoSennrich
@TsingYoga
@agostina_cal
@AnsongNi
@cl_uzh
@danish037
New paper🚨: We introduce POISONBENCH, a benchmark for assessing LLM vulnerabilities to data poisoning during preference learning. Key finding: Even 3% poisoned data can cause up to 80% performance deviation when triggered. 🧵
📢 🎉 New paper with @_clementneo & Shay Cohen! We study how attention heads work with MLP neurons to predict the next token. We find a set of interpretable activity. More in the thread!
🚀🧠 New paper with Michael Lan (My intern) and @philiptorr to appear at #EMNLP2024! We study the question: Do semantically similar tasks share important components (like attention heads and MLPs)? 1/8
Have a question that is challenging for humans and AI? We (@ai_risks + @scale_AI) are launching Humanity's Last Exam, a massive collaboration to create the world's toughest AI benchmark. Submit a hard question and become a co-author. Best questions get part of $500,000 in…
NLP people, what is a good natural language reasoning benchmark that is not already overfit by current-generation LLMs?
Attending #ACL2024? Come hear about our recent work on LLMs unlearning Removed Concepts, I will be at poster session 2, tomorrow, Monday 12th at 2-3:30pm. @michellewmlo & Shay B.Cohen (@InfAtEd )
New Paper 🎉: arxiv.org/pdf/2401.01814… Can language models relearn removed concepts? Model editing aims to eliminate unwanted concepts through neuron pruning. LLMs demonstrate a remarkable capacity to adapt and regain conceptual representations which have been removed 🧵1/8
Spoke with @RyanPGreenblatt from Redwood Research about his impressive GPT4o approach to @fchollet ARC challenge (generating and refining Python programs). We also spoke about his views on AI growth - Ryan was great! youtube.com/watch?v=z9j3wB…
🚨Excited to share our new paper!🚨 We reveal a curious generalization gap in the current refusal training approaches: simply reformulating a harmful request in the past tense (e.g., "How to make a Molotov cocktail?" to "How did people make a Molotov cocktail?") is often…
How well do text-to-SQL parsers handle ambiguous questions? 🤔 Introducing 🌿𝔸𝕄𝔹ℝ𝕆𝕊𝕀𝔸, a new benchmark that tests the limits of text-to-SQL semantic parsers in interpreting ambiguous requests! ambrosia-benchmark.github.io 1/5
New: Read the story of a decade-long propaganda campaign by the Forrest Gump of the internet—a Wikipedia admin who was once Yudkowsky’s strongest soldier—set against the backdrop of the collapse of the semi-unified Internet ethos of the ‘90s and ‘00s tracingwoodgrains.com/p/reliable-sou…
United States Trends
- 1. Mike 1,78 Mn posts
- 2. Serrano 239 B posts
- 3. Canelo 16,4 B posts
- 4. #NetflixFight 71,8 B posts
- 5. Father Time 10,7 B posts
- 6. Logan 77,9 B posts
- 7. #netflixcrash 15,7 B posts
- 8. Rosie Perez 14,8 B posts
- 9. He's 58 25,4 B posts
- 10. ROBBED 101 B posts
- 11. Boxing 297 B posts
- 12. #buffering 10,9 B posts
- 13. Shaq 16,1 B posts
- 14. Ramos 69,6 B posts
- 15. My Netflix 82,8 B posts
- 16. Roy Jones 7.153 posts
- 17. Tori Kelly 5.215 posts
- 18. Cedric 21,8 B posts
- 19. Barrios 50,5 B posts
- 20. Muhammad Ali 18 B posts
Who to follow
-
Piotr Nawrot
@p_nawrot -
Nikita Moghe
@nikita_moghe -
Naomi Saphra (follow elsewhere)
@nsaphra -
Sam Bowman
@sleepinyourhat -
UW NLP
@uwnlp -
Tom Sherborne
@tomsherborne -
Harshit Joshi
@harshitj__ -
Machel Reid
@machelreid -
Fangyu Liu
@hardy_qr -
Rico Sennrich
@RicoSennrich -
Yujia Qin
@TsingYoga -
Agostina Calabrese
@agostina_cal -
Ansong Ni
@AnsongNi -
Zurich Computational Linguistics Group
@cl_uzh -
Danish Pruthi
@danish037
Something went wrong.
Something went wrong.