Isabelle Lee @ ICML
206 posts
Isabelle Lee @ ICML
@wordscompute
ml/nlp phding @ usc, currently visiting harvard; training & interpretability & reasoning
boston/la Katılım Temmuz 2017
272 Takip Edilen260 Takipçiler
Isabelle Lee @ ICML retweetledi
Excited to be heading to #ICML2026 to present 3 papers! ✈️
Happy to talk about developmental LM science, continual learning, or LM creativity! See more details below, and send me a message via email if you'd like to chat! (Twitter chat isn't working for me). Also excited for shopping 👜
English
Isabelle Lee @ ICML retweetledi
NLAs look like they "work"---but compared to what? How can we know that they actually tell us anything about model internals?
In my new blogpost (and paper, to be presented at ICML!), we investigate whether activation verbalizers (like NLAs) produce faithful explanations.
GIF
Anthropic@AnthropicAI
New Anthropic research: Natural Language Autoencoders. Models like Claude talk in words but think in numbers. The numbers—called activations—encode Claude’s thoughts, but not in a language we can read. Here, we train Claude to translate its activations into human-readable text.
English
Isabelle Lee @ ICML retweetledi
We don’t always know what problems are hard for LLMs. So devs evaluate on tasks HUMANS find hard or on broad benchmarks. What if we could instead anticipate which scenarios a model will fail on—all without evaluating specific input examples?
🧵NEW PAPER by @jenniferlumeng &al
English
really excited to head home for icml:) and attending the co-located @farairesearch's alignment workshop (for the first time)! would love to meet others interested in training & interpretability
English
work w/ @_emliu @cathy__jiao @BrihiJ @DaniYogatama @FazlBarez @m2saxon
since am headed home for icml, it'll be presented by the amazing @BrihiJ!
this was my first time writing a position paper, which turned into a grant, which i'm turning into multiple projects 🙂 stay tuned
6/6
English
Benchmarks can be superficial, but model explanations and evaluations are fundamentally intertwined. What if we used interpretability as principled, scientific evaluation? If it met scientific standards?
arxiv.org/abs/2605.05508
coming to @evaluatingevals at ACL as oral 🧵
1/6
English
Isabelle Lee @ ICML retweetledi
🗣️ Prediction, Explanation, or Over-interpretation?
Recent work suggests LLMs can verbalize information about latent states and future generations. But training of different verbalization methods varies.
Are they verbalizing, or are we over-interpreting from the explanation?
1/n
English
Isabelle Lee @ ICML retweetledi
In film, "we'll fix it in post" is what you say when something went wrong on set and you don't want to redo it. AI research has made it our entire methodology: train the model, then patch whatever comes out. Our new ICML oral argues this can't be the basis of a science of AI. 🧵
English
check out @_emliu and our work on pretraining interp! we initially asked if we can predict from simple task learning, can we predict a mode complex learning behavior? super excited for follow-ups as well:)
Emmy Liu@_emliu
Copying → morphology/translation → basic arithmetic → complex reasoning & math. Across every model family we tested, LLMs acquire skills in roughly the same order during pretraining. Can we use this to predict what a model will learn next, just from its internals? 🧵
English
Isabelle Lee @ ICML retweetledi
Midtraining is a new part of many training pipelines, but when does it help and can it backfire? 🤔
In our new preprint, we use controlled experiments to pin this down. TL;DR; midtraining helps the most when it “bridges” pretraining and posttraining, and mitigates forgetting after posttraining. Timing is also very important.
🧵
English
Isabelle Lee @ ICML retweetledi
Our report from the Actionable Interpretability workshop is finally public! Some of my favorite scientists argued for hours and this is what they agreed on.
Hadas Orgad @ ICML@OrgadHadas
We’re not saying all interpretability work must be immediately actionable— curiosity-driven research still matters. But actionability is a high bar: understanding that works outside the lab. To make your next project more actionable, use our checklist >>
English
check hadas’s paper on what useful interpretability could enable! ..like effectively predicting deployment failures for example 🙃
Hadas Orgad @ ICML@OrgadHadas
Our ICML 2025 workshop on Actionable Interpretability drew massive interest. But the same questions kept coming up: What does "actionable" mean? Is it achievable? How? We're ready to answer. 🧵
English
Isabelle Lee @ ICML retweetledi
Excited to share our new dataset, FOL-Traces!
We introduce a large-scale dataset of programmatically verified FOL reasoning traces for studying structured logical inference + process fidelity
Happy to hear thoughts from others working on reasoning in LLMs
Check it out here 👇
English