Anna Hedström

Almost forgot to share — last month, I defended my thesis, with distinction! Feeling deeply grateful for the learnings, collaborations and friendships along the way. New chapter at @ETH_AI_Center 🚀

PRISM got accepted at @NeurIPSConf 2025! Congrats to the team ✨ @lkopf_ml @nfelnlp @kirill_bykov @BommerPhiline @Marina_MCV @EberleOliver

@MatthewKowal9 Amazing work!

This was a really fun project to work on - and huge shoutouts to my amazing collaborators who made the project such a delight!! 🎉💪

Couldn’t be more proud and happy for my labmate @kirill_bykov who made it to the other side! Congrats again to the fantastic body of work produced!

My brilliant co-author @salim_amk0 is presenting our work on Mechanistic Error Reduction with Abstention (MERA) now at ICML in Vancouver! 🚀 If you’re at ICML, come by East Exhibition Hall A-B, E-2605 at 4:30 pm (Vancouver, BC). We’d love to hear what you think!

🚀 I'll be presenting our #ICML paper this afternoon! You’ve probably heard of Mechanistic Steering, the idea of modifying internal activations of a language model at inference-time (e.g., adding a vector) to influence its behaviour, often for alignment. But we take a different angle: 👉 We use it for error reduction. If you've explored this space, you know it’s full of heuristics: Which vector to use? How long should it be? When to steer at all? 🎯 In our work, we bring principled answers to these questions, with provable guarantees. We introduce MERA (Mechanistic Error Reduction with Abstention for Language Models), a method for reducing errors in LLMs at inference-time by: ✅ Steering only when necessary ✅ Adapting how much to steer ✅ Abstaining unless confident improvement And the best part? MERA is modular. You can plug it into any existing steering method to make it more effective and safer. 📍Catch me at @icmlconf 📌 Poster Location: East Exhibition Hall A-B, E-2605 at 4:30 pm. 🧠 Paper: openreview.net/pdf?id=fUCPq5R… Big thanks to my amazing co-authors: @anna_hedstroem, @tom_bewley, Saumitra Mishra, and Manuela Veloso. #ICML2025 #LLMs #MechanisticSteering #InferenceTime #LLMSafety #ResponsibleAI #TrustworthyAI #AIResearch

Endless gratitude to brilliant @salim_amk0, @tom_bewley, and our fantastic collaborators at JP Morgan #ICML25

4/ What’s fascinating is not just the outcome but how concepts like "error" show up inside LLMs. This opens the door to more general forms of lightweight, post-training control — we're curious where else MERA may help. Paper openreview.net/pdf?id=fUCPq5R… github.com/annahedstroem/…

Couldn’t be more excited to share our latest paper — accepted to ICML 2025 @icmlconf — with JP Morgan AI Research. It explores a simple question: To safely and effectively mitigate errors post-training, when (and how much) should we steer large language models? 🧵

@lkopf_ml @kirill_bykov @nfelnlp @BommerPhiline @Marina_MCV @EberleOliver

Very excited to share this preprint on labelling polysemantic neurons! Have a read arxiv.org/pdf/2506.15538 And happy midsummer!

@spacecadet_kels You’re amazing!! Congrats

Very proud to share my TEDx talk, “AI won’t save us, but it can help us” - recorded at the QueensU 2025 TEDx event (the largest in Canada!): youtu.be/9V2UnKapYsI?si…

@saprmarks Agreed!

In a new post, I argue that interpretability researchers should demo downstream applications of their research as a means of validation.

If you're at #AAAI2025 don't miss our poster today (alignment track)! Paper 📘: arxiv.org/pdf/2502.15403 Code 👩‍💻: github.com/annahedstroem/… Team work with @eirasf and @Marina_MCV

I couldn’t be more proud and happy to share that our paper also got awarded survey certification for "exceptionally thorough/ insightful survey” of interpretability evaluation Grateful to my brilliant co-authors @BommerPhiline @tfburns @SLapuschkin @WojciechSamek @Marina_MCV

Our new paper is out! "Evaluating Interpretable Methods via Geometric Alignment of Functional Distortions" 📖 Read: openreview.net/pdf?id=ukLxqA8… 💻 Code: github.com/annahedstroem/… Thanks to my best collaborators @BommerPhiline @tfburns @SLapuschkin @WojciechSamek @Marina_MCV

@SatyaScribbles so well deserved!!!!

Great to see our work 'More RLHF, More Trust?' receive an oral presentation at #ICLR2025. Nice to see one can receive these opportunities even while being GPU poor. 🎉