PhD supervised by Tim Rocktäschel and Ed Grefenstette, part time at Cohere. Language and LLMs. Spent time at FAIR, Google, and NYU (with Brenden Lake). She/her.https://bsky.app/profile/lauraruis.bsky.socialhttps://bsky.app/profile/lauraruis.bsky.social/post/3ldvbfjwna22a"Rather than being animals that *think*, we are *animals* that think"; the last sentence of Tom Griffiths's characterisation of human intelligence through limited time, compute, and communication hits different today than it did 4 years ago.22 Dec 2024 11:04 +0000at://did:plc:sofu42aaa45my3kutq22wt2v/app.bsky.feed.post/3ldvbfjwna22ahttps://bsky.app/profile/lauraruis.bsky.social/post/3ldee5snlak2ySometimes o1's thinking time almost feels like a slight. o1 is like "oh I thought about this uninvolved question of yours for 7 seconds and here is my 20 page essay on it"15 Dec 2024 17:38 +0000at://did:plc:sofu42aaa45my3kutq22wt2v/app.bsky.feed.post/3ldee5snlak2yhttps://bsky.app/profile/lauraruis.bsky.social/post/3lcsobhwpa22jI'll be at NeurIPS tues-sun, send me a message if you'd like to chat!08 Dec 2024 16:51 +0000at://did:plc:sofu42aaa45my3kutq22wt2v/app.bsky.feed.post/3lcsobhwpa22jhttps://bsky.app/profile/lauraruis.bsky.social/post/3lbx2zlwl322uDo you know what rating you’ll give after reading the intro? Are your confidence scores 4 or higher? Do you not respond in rebuttal phases? Are you worried how it will look if your rating is the only 8 among 3’s? This thread is for you.27 Nov 2024 17:25 +0000at://did:plc:sofu42aaa45my3kutq22wt2v/app.bsky.feed.post/3lbx2zlwl322uhttps://bsky.app/profile/lauraruis.bsky.social/post/3lbfeqqpkts22How do LLMs learn to reason from data? Are they ~retrieving the answers from parametric knowledge🦜? In our new preprint, we look at the pretraining data and find evidence against this: Procedural knowledge in pretraining drives LLM reasoning ⚙️🔢 🧵⬇️20 Nov 2024 16:31 +0000at://did:plc:sofu42aaa45my3kutq22wt2v/app.bsky.feed.post/3lbfeqqpkts22