iiinova — where your digital self keeps evolving

Najoung Kim @najoung.bsky.social · 02/10/2026

sadly I won't be at #COLM2026 this year but tinlab students Audrey and Jing will be there so talk to them! A little bit about what they'll be doing:

130

Najoung Kim @najoung.bsky.social · 01/10/2026

briefly emerging to wish happy october to my friends

0101

Najoung Kim @najoung.bsky.social · 22/09/2026

Thanks @goodfire.bsky.social for supporting my lab with this grant! Excited to see how Silico could support projects on compositionality I've been meaning to explore for a while, as well as our lab's ongoing project on the role of people in ideation with research agents.

goodfire.com

Announcing Goodfire Research Grants - Goodfire

$1M in free Silico usage for academic and nonprofit researchers working on AI interpretability, alignment, and life sciences

090

Najoung Kim @najoung.bsky.social · 17/09/2026

Excited about finding grammaticality as an emergent property encoded in LM representations, deconfounded from factors affecting general probability! Hope this helps revisit debates in this space with a more nuanced answer to the prerequisite "do LMs even show sensitivity" Q

1150

Reposted by Najoung Kim

Jane Li 🦖 @janeli.bsky.social · 17/09/2026

🦀New preprint! (w/ @najoung.bsky.social)🦞 Is grammaticality a major organizing principle of NLM representations? We show that many NLMs exhibit abstract rep. separation for grammaticality. We believe this work addresses debates about confounds in measuring model gram. knowledge. [1/10]

12010

Najoung Kim @najoung.bsky.social · 02/07/2026

...check out our poster on mechanistic interpretability of entity tracking on Thursday! Peter, Qiao, & Gabriel will all be there in person (or so I hope) Paper link here: arxiv.org/abs/2605.302...

030

Najoung Kim @najoung.bsky.social · 02/07/2026

i will be at #ICML2026! first time in my career (!) that I'm attending a major conf in my home town and super excited about it 💖🇰🇷 (but also... clash of worlds 😬) I'll be doing hallway track for the most part so DM or email me if you'd like to hang! + don't forget to 👇

160

Preprint link here: osf.io/preprints/ps... 💖💖

osf.io

OSF

030

I'd love to hear your feedback! I don't think any claim is particularly controversial, and in retrospect, maybe the takeaway is even obvious. But I think this is a good summary of the beliefs I formed over the past few years, and was definitely useful for me to articulate them.

The position piece argues that behavior has to be constitutive for compositionality and mechanism is neither necessary nor sufficient. This doesn't mean that mechanism has no role to play: I lay out three roles it can play (interface question, boostrap, hypothesis generation).

For a while, I've been thinking "maybe interp will save us from the evaluation crisis we've been facing RE: compositionality" and really wanted to believe that, but actually trying to cache this out made me less sure.

130

Najoung Kim @najoung.bsky.social · 17/04/2026

🐸 New position paper on compositionality! 🐸 I synthesize my thoughts about the proper role/interpretation of behavior and mechanism in asking the question "Is this system (mind or machine) exhibiting compositionality?"

No Escape from Behavior in Evaluating Compositionality by Najoung Kim

1274

OG hyp gen paper update! Even if you're not interested in the specific phenomenon we look at, the general framing should be of interest to a much broader audience

040

Reposted by Najoung Kim

Kanishka Misra @kanishka.bsky.social · 27/01/2026

“All bears have a property”, “Some bears have a property”, “Bears have a property” are different in terms of how the property is generalized to a specific bear – a great example of how language constrains thought! This holds for kids, adults, and according to our new work, (V)LMs! 🧵

Title page of our paper: "Bears, all bears, and some bears. Language Constraints on Language Models' Inductive Inferences"

1269

Najoung Kim @najoung.bsky.social · 08/12/2025

the slides are kanishka's thread but if you are interested in the non-cut-down version of the first half here it is: bsky.app/profile/najo...

Najoung Kim @najoung.bsky.social · 08/12/2025

happy to have finally made a pilgrimage to ucsd, a special place for some of us :)

You can get more info on my research interests on my website: najoung.kim And some of the bigger picture thoughts that might not be evident from individual publications here: bsky.app/profile/najo...

I also want to mention that the lang x computation research community at BU is growing in an exciting direction, especially with new faculty like @amuuueller.bsky.social, @anthonyyacovone.bsky.social, @nsaphra.bsky.social, & @profsophie.bsky.social! Also, Boston is quite nice :)

192

I'm recruiting through Linguistics (Deadline: Jan 7), CS & CDS (Deadlines: Dec 15). Find more info here: najoung.kim/students/

najoung.kim

For Students

BU Linguistics / CS

110

Najoung Kim @najoung.bsky.social · 18/11/2025

My lab at BU is recruiting PhD students and possibly a postdoc this year! We study humans & machines, centered around topics like meaning, generalization, evaluation methods and design, and the nature of computation and representation that underlie language and cognition. 🫴🫴

1133

Reposted by Najoung Kim

Here are the slides: docs.google.com/presentation... For context the intended audience is Lang Dev researchers who are attending a session on "what can LLMs tell us about human language". If you have any thoughts I'd love to hear them!

docs.google.com

SLD Plenary - Najoung Kim [external share]

Whence insights? The value of delineating human and machine CogSci Najoung Kim (Boston University) Society for Language Development Annual Symposium November 6, 2025 1

082

Najoung Kim @najoung.bsky.social · 18/11/2025

docs.google.com

SLD Plenary - Najoung Kim [external share]

Whence insights? The value of delineating human and machine CogSci Najoung Kim (Boston University) Society for Language Development Annual Symposium November 6, 2025 1

082

Najoung Kim @najoung.bsky.social · 10/11/2025

for something to be considered creative, it has to be semantically/pragmatically felicitous!!!

081

Najoung Kim @najoung.bsky.social · 09/11/2025

i was a bit nervous about it but so happy that it was generally well received, even from folks who are usually skeptics! it meant a lot. will share slides soon and hopefully a position paper too, watch this space 👁

Najoung Kim @najoung.bsky.social · 09/11/2025

so awesome to have spoken alongside Virginia Valian and Ethan Wilcox, and a pre-talk conversation with them was really helpful! people should do this more

130

Najoung Kim @najoung.bsky.social · 09/11/2025

honored to have given a plenary address at the Society for Language Development annual symposium titled "Whence insights? The value of delineating human and machine CogSci". It's a synthesis of a few years of thoughts, recently concretized with @aditya-yedetore.bsky.social and @kanishka.bsky.social

screenshot of a slide titled "Whence insights? The value of delineating human and machine CogSci"

1132

Najoung Kim @najoung.bsky.social · 25/07/2025

Required qualifications: BA, hands-on engineering experience, and interest in research. Please help spread the word and share with people who might be a good fit! 🪄 Unfortunately only open to US citizens/permanent residents/independent work auth holders.

000

Najoung Kim @najoung.bsky.social · 25/07/2025

The first part of this effort was RExBench: bsky.app/profile/najo... There will be some freedom of scope for the second part of the project within the theme of research agent evaluation - the RA will contribute to scoping the project along with the team as well.

100

Najoung Kim @najoung.bsky.social · 25/07/2025

👾 Full-time research assistant position (1 year) with @sebschu.bsky.social and me! 👾 We're looking for someone to join the research agent evaluation team, starting Fall 2025. Application link to be available soon, but feel free to send us your CV and/or come talk to us at #ACL2025. 🧵

1114

Reposted by Najoung Kim

Lizzie Coppock @lizzieloo.bsky.social · 24/07/2025

I have a sabbatical coming up and I'm going to Nepal! Why Nepal? You can read about it at my first blog post about this trip. TL;DR: linguistic diversity, writing systems (and pretty scripts), classifiers, and a school for Newar kids in Kathmandu. sites.bu.edu/lislab/2025/...

sites.bu.edu

Field trip to Nepal! | Linguistic Semantics Lab (LiSLab)

071

Najoung Kim @najoung.bsky.social · 22/07/2025

ever since VLMs were a thing i've been interested in how the additional visual modality changes language in meaningful ways. after negative findings after negative findings, excited to report this result! proud of our junior authors for digging into this 🐸

0123

Najoung Kim @najoung.bsky.social · 08/07/2025

green carded finally 💚💚

080

Najoung Kim @najoung.bsky.social · 02/07/2025

Excited this is finally out and ❤️ to our team @nedwards99.bsky.social @yukyunglee.bsky.social Audrey Mao and Yulu Qin! of course @sebschu great as always 🙏🙏

Najoung Kim @najoung.bsky.social · 02/07/2025

Seeing an experiment and thinking "but have they tried X? what if we do Y?" is a key part of research and a start to new discoveries. RExBench tests if coding agents can implement new extensions. It complements recent evals (eg PaperBench from OpenAI ) on replication! See 👇 for details

Koyena Pal @koyena.bsky.social · 30/06/2025

Reposted by Najoung Kim

🚨 Registration is live! 🚨 The New England Mechanistic Interpretability (NEMI) Workshop is happening Aug 22nd 2025 at Northeastern University! A chance for the mech interp community to nerd out on how models really work 🧠🤖 🌐 Info: nemiconf.github.io/summer25/ 📝 Register: forms.gle/v4kJCweE3UUH...

0108

Najoung Kim @najoung.bsky.social · 18/06/2025

i'll be in copenhagen for a few days, lmk if you want to get coffee! will be around most of Thurs and early Sat. alternatively you can also come see me give a talk (at a museum apparently) on Fri: cphnlp.github.io

cphnlp.github.io

Copenhagen NLP Symposium 2025

symposium website

070

Reposted by Najoung Kim

Anna Rogers @annarogers.bsky.social · 26/05/2025

📢 The Copenhagen NLP Symposium on June 20th! - Invited talks by @loubnabnl.hf.co (HF) @mziizm.bsky.social (Cohere) @najoung.bsky.social (BU) @kylelo.bsky.social (AI2) Yohei Oseki (UTokyo) - Exciting posters by other participants Register to attend and/or present your poster at cphnlp.github.io /1

cphnlp.github.io

Copenhagen NLP Symposium 2025

symposium website

13412

Reposted by Najoung Kim

Matt Goldrick @mattgoldrick.bsky.social · 29/05/2025

Very sorry to learn of the passing of LouAnn Gerken, who had such an impact on our understanding of the acquisition of speech sounds and sound patterns obits.mlive.com/us/obituarie...

obits.mlive.com

LouAnn Gerken Obituary (1959 - 2025) - Tucson, AZ - Arizona Daily Star

View LouAnn A. Gerken's obituary, send flowers and sign the guestbook.

24826

Najoung Kim @najoung.bsky.social · 04/05/2025

thanks Matt, means a lot!! 😊

Najoung Kim @najoung.bsky.social · 04/05/2025

hello NAACL friends I'm giving a keynote today at RepL4NLP at 1:30PM local time, come say hi! I'll mostly be musing about things with light research discussions

Screenshot of a slide that says "what does it take to convince ourselves that a system is exhibiting compositionality?" with a side comment "mostly AI, but humans too!!" for the word system

1171

Najoung Kim @najoung.bsky.social · 29/03/2025

we are scheming to make everyone's lab name a recursive acronym!