←
Andreas Grivas564 posts
Andreas Grivas avatar

Andreas Grivas

@andreasgrv

Interested in Bottlenecks in Neural Networks; Unargmaxable Outputs. Postdoc in ML/NLP at the University of Edinburgh.

◷ Joined August 2015▧ 58 Photos and videos
Andreas Grivas avatar
Andreas Grivas@andreasgrv·16 Feb 2024

How expressive is your deep multi-label classifier? Can it represent all outputs of interest? 🤔 SPOILER🚨: Your model can have *test set* outputs that are impossible to predict! 🚫 Check out our paper! arxiv.org/abs/2310.10443 🧵(1/7)

Post attachment
Nathan Godey @COLM 🌉 avatar
⟳ Andreas Grivas retweeted
Nathan Godey @COLM 🌉@nthngdy·Aug 19

🧵New paper: Matryoshka LM Suites Instead of training all models of a suite independently, we nest them within a single architecture and train the suite in a single run => much less compute, same performance Also allows free distillation, better speculative decoding & more 👇

Post attachment
Edoardo Ponti avatar
⟳ Andreas Grivas retweeted
Edoardo Ponti@PontiEdoardo·Jun 29

I'll be in Seoul for #ICML2026 to present - fast and expressive MTP - blending SFT and RLVR with prefix sampling - faster differentiable sparse attention - self-improving world models with latent actions Dm me if you'd like to meet and chat (incl. PhD/postdoc openings in my lab)!

Piotr Nawrot avatar
⟳ Andreas Grivas retweeted
Piotr Nawrot@p_nawrot·May 13

The best supervisor I could have asked for is building something really exciting at Imperial London. If you're considering a PhD in Efficient NLP — seriously, look into this.

I am moving to @ICComputing at @imperialcollege as an associate professor, where I will be expanding my lab! I am looking for PhDs and postdocs to join me on my quest to build foundation models with adaptive tokenisation and memory (AToM FMs, funded by @ERC_Research)

Post attachment
Desmond Elliott avatar
⟳ Andreas Grivas retweeted
Desmond Elliott@delliott·16 Dec 2025

I am grateful that the Carlsberg Foundation is supporting our basic research on tokenization-free language models at the University of Copenhagen. I will be hiring Ph.D students to start in September 2026. Feel free to reach out early if you want to express informal interest.

Fra politologi til arkæologi. Fra astrofysik til marinbiologi og glaciologi. 159 forskere modtager i dag en bevilling fra Carlsbergfondet til vidt forskellige grundvidenskabelige initiativer. Se hvilke projekter, der har fået støtte 👉bit.ly/4iK2fV2 #dkforsk

Post attachment
Benjamin Minixhofer avatar
⟳ Andreas Grivas retweeted
Benjamin Minixhofer@bminixhofer·15 Dec 2025

We are releasing Bolmo today! Bolmo is the best byte-level model so far. It comes close to and sometimes surpasses Olmo 3. Bolmo also performs competitively in terms of speed & is fully open. I was skeptical of byte-level models for a long time but I finally switched camps🧵

Post attachment
Ai2@allen_ai15 Dec 2025

Introducing Bolmo, a new family of byte-level language models built by "byteifying" our open Olmo 3—and to our knowledge, the first fully open byte-level LM to match or surpass SOTA subword models across a wide range of tasks. 🧵

Post attachmentPost attachment
Edoardo Ponti avatar
⟳ Andreas Grivas retweeted
Edoardo Ponti@PontiEdoardo·15 Dec 2025

Finally, you can count the r's in strawberry and check if 3.11 is higher than 3.9 without tokenisation interfering: Here's Bolmo, a fully open byte-level LLM with latent tokenisation, derived from a SOTA LLM (Olmo 3). Promising on coding and char-level understanding!

Ai2@allen_ai15 Dec 2025

Introducing Bolmo, a new family of byte-level language models built by "byteifying" our open Olmo 3—and to our knowledge, the first fully open byte-level LM to match or surpass SOTA subword models across a wide range of tasks. 🧵

Post attachmentPost attachment
Ivan Titov avatar
⟳ Andreas Grivas retweeted
Ivan Titov@iatitov·1 Dec 2025

Happy to announce one (or more) postdoctoral positions at the U Amsterdam! There’s a lot of flexibility in research direction, including continual learning, memory in LLMs, AI safety, unlearning/editing, reasoning, and interpretability - areas our group is currently focused on.

Post attachment
Piotr Nawrot avatar
⟳ Andreas Grivas retweeted
Piotr Nawrot@p_nawrot·1 Dec 2025

We'll present "Inference-Time Hyper-Scaling with KV Cache Compression", both at NeurIPS and EurIPS. We believe that future advances in AI will require model efficiency, and this work is another step in this direction. Save the date! -San Diego, Thur 11:00 -Copenhagen, Thur 10:30

Post attachment
Ivan Titov avatar
⟳ Andreas Grivas retweeted
Ivan Titov@iatitov·16 Nov 2025

Excited about the collaboration with Kolya @FelineAutomaton . We’re offering a fully funded PhD at @EdinburghNLP (start Sept 2026), working on language-based state representations for time series, comes with a generous budget for travel and experiments.

This tweet is unavailable