Pau Rodriguez @paurodriguez.bsky.social · 21/10/2025🚀 Excited to share LinEAS, our new activation steering method accepted at NeurIPS 2025! It approximates optimal transport maps e2e to precisely guide 🧭 activations achieving finer control 🎚️ with ✨ less than 32 ✨ prompts! 💻https://github.com/apple/ml-lineas 📄https://arxiv.org/abs/2503.10679 121
Reposted by Pau RodriguezMarco Cuturi @marcocuturi.bsky.social · 03/10/2025Our two phenomenal interns, Alireza Mousavi-Hosseini and Stephen Zhang @syz.bsky.social have been cooking some really cool work with Michal Klein and me over the summer. Relying on optimal transport couplings (to pick noise and data pairs) should, in principle, be helpful to guide flow matching 🧵 2307
Pau Rodriguez @paurodriguez.bsky.social · 11/04/2025Our work on fine-grained control of LLMs and diffusion models via Activation Transport will be presented @iclr_conf as spotlight✨Check out our new blog post machinelearning.apple.com/research/tra...machinelearning.apple.comControlling Language and Diffusion Models by Transporting ActivationsLarge generative models are becoming increasingly capable and more widely deployed to power production applications, but getting these… 071
Reposted by Pau RodriguezDeep Learning Barcelona @dlbcnai.bsky.social · 16/12/2024Què és l’aprenentatge profund ? La @marionamec.bsky.social de @neurofregides.bsky.social ens ho explica en motiu del Deep Learning Barcelona Symposium 2024 (@dlbcn.ai), aquest dijous 19 de desembre. #deeplearning #ciencia #català #barcelona www.youtube.com/shorts/R4u_Z...youtube.comQuè és l'aprenentatge profund ? - La Dimoni de Maxwell #deeplearning #ciencia #català #barcelonaYouTube video by Deep Learning Barcelona 073
Reposted by Pau RodriguezMichael Kirchhof @mkirchhof.bsky.social · 12/12/2024Evaluating your LLM uncertainties with Rougle-L will show clear winners... except that they aren't actually good. We find that Rouge-L spuriously favors some methods over others. 🧵1/4 📄 openreview.net/forum?id=jGt... NeurIPS: Sunday, East Exhibition Hall A, Safe Gen AI workshop 173
Pau Rodriguez @paurodriguez.bsky.social · 10/12/2024Thrilled to share the latest work from our team at @Apple where we achieve interpretable and fine-grained control of LLMs and Diffusion models via Activation Transport 🔥 📄 arxiv.org/abs/2410.23054 🛠️ github.com/apple/ml-act 0/9 🧵 34715
Reposted by Pau RodriguezYoshua Bengio @yoshuabengio.bsky.social · 28/11/2024Thank you to the @neuripsconf.bsky.social for this recognition of the Generative Adversarial Nets paper published ten years ago with @ian-goodfellow.bsky.social, Jean Pouget-Abadie, @memimo.bsky.social, Bing Xu, David Warde-Farley, Sherjil Ozair and Aaron Courville. blog.neurips.cc/2024/11/27/a...blog.neurips.ccAnnouncing the NeurIPS 2024 Test of Time Paper Awards – NeurIPS Blog 419420
Reposted by Pau RodriguezDeep Learning Barcelona @dlbcnai.bsky.social · 22/11/2024Apple will be a platinum sponsor of the Deep Learning Barcelona Symposim 2024. This is the first time that Apple sponsors the event. #DLBCN 011
Reposted by Pau RodriguezDesi R Ivanova @desirivanova.bsky.social · 21/11/2024Bring stats to LM evals!! open.substack.com/pub/desiriva...open.substack.comInbox | Substack 041
Pau Rodriguez @paurodriguez.bsky.social · 20/11/2024Watching Frieren can’t stop thinking that demons are evil LLMs 😅 030