Laura O'Mahony
@_lauraaisling
Followers
251
Following
1K
Media
4
Statuses
118
Maths + ML PhD Candidate at SFI CRT in foundations of data science in UL 📚🇮🇪
Cork, Ireland
Joined November 2018
RT @richardcsuwandi: Most AI systems today follow the same predictable pattern: they're built for specific tasks and optimized for objectiv….
0
25
0
RT @MLStreetTalk: AI is so smart, why are its internals 'spaghetti'? We spoke with @kenneth0stanley and @akarshkumar0101 (MIT) about their….
0
65
0
"A stylish woman in a beige coat walking confidently past a moving train at a modern train station.". Create images and videos in seconds with Grok Imagine.
758
722
4K
RT @andrewdai99: 📣 New 📝!. Under Alex’s great leadership, we identified a unification under QDC measures that exist to bridge synthetic dat….
0
6
0
It was a pleasure to have been part of this project led by @Dahoas1 . With synthetic data being so important in training LLMs these days, this survey on the impacts of QDC of synthetic data for LLM performance is timely.
How important is the quality, diversity, and complexity (QDC) of synthetic data for LLM performance? What effect does QDC data composition have on self-improvement?. We just released a comprehensive survey discussing these questions (and many more) 🧵
0
1
9
I love this paper as it finally tackles something I’ve been confused about for the last few years since I started working on interpretability!.
Have you ever wondered what ✨mechanistic interpretability✨ is, & how it differs from other NLP interpretability research? @nsaphra and I have the paper for you!. Check out our paper (which I'll present @BlackboxNLP @emnlpmeeting in Miami next month!).
0
0
3
RT @CohereForAI: This clip from The Journey of Aya documentary features @DeividasMat, @singhshiviii, @luisa_moura_, @muhaksim, and @_lauraa….
cohere.com
Cohere’s research lab, Cohere Labs, released the Aya model, a state-of-the-art, open source, massively multilingual, research LLM covering 101 languages – including more than 50 previously underser...
0
2
0
RT @ziruichen44: Why do varied DNN designs yield equally good models of human vision? Our preprint with @michaelfbonner shows that diverse….
arxiv.org
Do neural network models of vision learn brain-aligned representations because they share architectural constraints and task objectives with biological vision or because they learn universal...
0
43
0
Fascinating paper!.
𝗧𝗵𝗲 𝗚𝗲𝗻𝗼𝗺𝗶𝗰 𝗖𝗼𝗱𝗲 - 𝘁𝗵𝗲 𝗴𝗲𝗻𝗼𝗺𝗲 𝗶𝗻𝘀𝘁𝗮𝗻𝘁𝗶𝗮𝘁𝗲𝘀 𝗮 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗺𝗼𝗱𝗲𝗹 𝗼𝗳 𝘁𝗵𝗲 𝗼𝗿𝗴𝗮𝗻𝗶𝘀𝗺 🧬.very excited to share this new preprint from me and Nick Cheney 😀🧵
0
1
2
RT @karpathy: To help explain the weirdness of LLM Tokenization I thought it could be amusing to translate every token to a unique emoji. T….
0
1K
0
RT @sarahookr: Is bigger always better? 🐘 The idea that scaling more than any other ingredient has driven progress has become formalized as….
0
85
0
RT @davidbau: Time to study #llama3 405b, but gosh it's big!. Please retweet: if you have a great experiment but not enough GPU, here is an….
ndif.us
NDIF is a research computing project that enables researchers and students to crack open the mysteries inside large-scale AI systems.
0
37
0
RT @maksym_andr: Perhaps my favorite jailbreak: making a harmful request in the past tense (How to create Y? →How did people create Y?). W….
0
10
0
RT @yong_zhengxin: 🔥New work on multilinguality + safety + mech interp!. We show that DPO training in only English can detoxify LLM in many….
0
40
0
RT @sarahookr: Aya took 14 months involving 3000 + collaborators and was as much a protest about how research is done as it was a movement….
0
24
0
RT @CohereForAI: 🌱 We’re very excited that our work "Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning" was also a….
0
21
0
RT @singhshiviii: 2/2! Yay! First ever acceptance at a conference! And it's ACL! 🎉. Huge congrats to all co-authors! .It's been a such a jo….
0
7
0
RT @andrewdai99: How can LLMs automatically generate meaningfully diverse, high-quality text responses in creative domains (like generating….
0
31
0