Siyuan Song

Figure 1 showing alignment pipeline using CLIP models on BabyView data.

Figure 2: human judgments are correlated with CLIP scores.

Can we use VLMs to quantify multimodal alignment in children's experiences? We analyze a large corpus of headcam videos to find out!

New preprint from our BabyView project, led by @alvinwmtan.bsky.social and Jane Yang: arxiv.org/abs/2511.18824

Looking forward to #NeurIPS25 this week 🏝️! I'll be presenting at Poster Session 3 (11-2 on Thursday). Feel free to reach out!

I’m excited to present SimpleStories at EurIPS!

Also if anyone at #EurIPS is interested in chatting about LLM data efficiency, interpretability, model inconsistency or other topics feel free to DM me.

Dataset and models: lnkd.in/e_VGWqhP
Code: lnkd.in/eEidmv74
Paper: lnkd.in/eH6jS9uY

String probability might be the best tool for assessing LMs' grammatical knowledge, yet it does not directly tell you 'how grammatical' a string is. Here's why and how we should use string probability and minimal pairs:
Excited to see this out - it's my great honor to be part of this amazing team!

Oh cool! Excited this LM + construction paper was SAC-Highlighted! Check it out to see how LM-derived measures of statistical affinity separate out constructions with similar words like "I was so happy I saw you" vs "It was so big it fell over".

Delighted Sasha's (first year PhD!) work using mech interp to study complex syntax constructions won an Outstanding Paper Award at EMNLP!

Also delighted the ACL community continues to recognize unabashedly linguistic topics like filler-gaps... and the huge potential for LMs to inform such topics!

Interested in doing a PhD at the intersection of human and machine cognition? ✨ I'm recruiting students for Fall 2026! ✨

Topics of interest include pragmatics, metacognition, reasoning, & interpretability (in humans and AI).

Check out JHU's mentoring program (due 11/15) for help with your SoP 👇

🧠 New at #NeurIPS2025!
🎵 We're far from the shallow now🎵
TL;DR: We introduce the first "reasoning embedding" and uncover its unique spatio-temporal pattern in the brain.

🔗 arxiv.org/abs/2510.228...

Introducing Global PIQA, a new multilingual benchmark for 100+ languages. This benchmark is the outcome of this year’s MRL shared task, in collaboration with 300+ researchers from 65 countries. This dataset evaluates physical commonsense reasoning in culturally relevant contexts.

Very excited to be going to Chicago for
@agnescallard.bsky.social's famous Night Owls next week! I'll be discussing my essay "ChatGPT and the Meaning of Life". Hope to see you there if you're local!

Title of our paper: “Hey, wait a minute: on at-issue sensitivity in Language Models” by Sanghee Kim and Kanishka Misra. Below: A person says “Sue, Max’s girlfriend, was a tennis champ!”; a second person responds with “What racket does she use?” (which targets at-issue content); a third person replies with “They’re dating?” (which targets not at-issue content)

If I spill the tea—“Did you know Sue, Max’s gf, was a tennis champ?”—but then if you reply “They’re dating?!” I’d be a bit puzzled, since that’s not the main point! Humans can track what’s ‘at issue’ in conversation. How sensitive are LMs to this distinction?

New paper w/ @sangheekim.bsky.social!

I will be recruiting PhD students via Georgetown Linguistics this application cycle! Come join us in the PICoL (pronounced “pickle”) lab. We focus on psycholinguistics and cognitive modeling using LLMs. See the linked flyer for more details: bit.ly/3L3vcyA

Title page of the paper: WUGNECTIVES: Novel Entity Inferences of Language Models from Discourse Connectives, with two figures at the bottom Left: Our figure 1 -- comparing previous work, which usually predicted the connective given the arguments (grounded in the world); our work flips this premise by getting models to use their knowledge of connectives to predict something about the world. Right: Our main results across 7 types of connective senses. Models are especially bad at Concession connectives.

"Although I hate leafy vegetables, I prefer daxes to blickets." Can you tell if daxes are leafy vegetables? LM's can't seem to! 📷

We investigate if LMs capture these inferences from connectives when they cannot rely on world knowledge.

New paper w/ Daniel, Will, @jessyjli.bsky.social

Gonna keep updating it regularly and have some fun with the resources we’ve got to grow & test Chinese BabyLMs🐣stay tuned!

Honored to get the chance to contribute to the Chinese dataset! And had a great time working with all the awesome collaborators!

Excited to present this at COLM tomorrow! (Tuesday, 11:00 AM poster session)

I will be giving a short talk on this work at the COLM Interplay workshop on Friday (also to appear at EMNLP)!

Will be in Montreal all week and excited to chat about LM interpretability + its interaction with human cognition and ling theory.

Both Direct and Indirect Evidence Contribute to Dative Alternation Preferences in Language Models Language models (LMs) tend to show human-like preferences on a number of syntactic phenomena, but the extent to which these are attributable to direct exposure to the phenomena or more general propert...

Traveling to my first @colmweb.org🍁

Not presenting anything but here are two posters you should visit:

1. @qyao.bsky.social on Controlled rearing for direct and indirect evidence for datives (w/ me, @weissweiler.bsky.social and @kmahowald.bsky.social), W morning

Paper: arxiv.org/abs/2503.20850

On my way to #COLM2025 🍁

Check out jessyli.com/colm2025

QUDsim: Discourse templates in LLM stories arxiv.org/abs/2504.09373

EvalAgent: retrieval-based eval targeting implicit criteria arxiv.org/abs/2504.15219

RoboInstruct: code generation for robotics with simulators arxiv.org/abs/2405.20179

Language Models Fail to Introspect About Their Knowledge of Language There has been recent interest in whether large language models (LLMs) can introspect about their own internal states. Such abilities would make LLMs more interpretable, and also validate the use of s...

I’m at #COLM2025 from Wed with:

@siyuansong.bsky.social Tue am introspection arxiv.org/abs/2503.07513

@qyao.bsky.social Wed am controlled rearing: arxiv.org/abs/2503.20850

@sashaboguraev.bsky.social INTERPLAY ling interp: arxiv.org/abs/2505.16002

I’ll talk at INTERPLAY too. Come say hi!

Heading to #COLM2025 to present my first paper w/ @jennhu.bsky.social @kmahowald.bsky.social !

When: Tuesday, 11 AM – 1 PM
Where: Poster #75

Happy to chat about my work and topics in computational linguistics & cogsci!

Also, I'm on the PhD application journey this cycle!

Paper info 👇:

Illustration of the blog post's main argument, summarized as: "Theory of Mind as a Central Skill for Researchers: Research involves many skills.If each skill is viewed separately, each one takes a long time to learn. These skills can instead be connected via theory of mind – the ability to reason about the mental states of others. This allows you to transfer your abilities across areas, making it easier to gain new skills."

🤖 🧠 NEW BLOG POST 🧠 🤖

What skills do you need to be a successful researcher?

The list seems long: collaborating, writing, presenting, reviewing, etc

But I argue that many of these skills can be unified under a single overarching ability: theory of mind

rtmccoy.com/posts/theory...

Picture of the UT Tower with "UT Austin Computational Linguistics" written in bigger font, and "Humans processing computers processing human processing language" in smaller font

The compling group at UT Austin (sites.utexas.edu/compling/) is looking for PhD students!

Come join me, @kmahowald.bsky.social, and @jessyjli.bsky.social as we tackle interesting research questions at the intersection of ling, cogsci, and ai!

Some topics I am particularly interested in:

Can AI aid scientists amidst their own workflows, when they do not know step-by-step workflows and may not know, in advance, the kinds of scientific utility a visualization would bring?

Check out @sebajoe.bsky.social’s feature on ✨AstroVisBench:

Simon Goldstein & Harvey Lederman, What Does ChatGPT Want? An Interpretationist Guide - PhilPapers This paper investigates LLMs from the perspective of interpretationism, a theory of belief and desire in the philosophy of mind. We argue for three conclusions. First, the right object of study ...

Simon Goldstein and I have a new paper, “What does ChatGPT want? An interpretationist guide”.

The paper argues for three main claims.

philpapers.org/rec/GOLWDC-2 1/7

I did a QA with Quanta about interpretability and training dynamics! I got to talk about a bunch of research hobby horses and how I got into them.

Latent learning: episodic memory complements parametric learning by enabling flexible reuse of experiences When do machine learning systems fail to generalize, and what mechanisms could improve their generalization? Here, we draw inspiration from cognitive science to argue that one weakness of machine lear...

Why does AI sometimes fail to generalize, and what might help? In a new paper (arxiv.org/abs/2509.16189), we highlight the latent learning gap — which unifies findings from language modeling to agent navigation — and suggest that episodic memory complements parametric learning to bridge it. Thread:

Announcing the first (and perhaps only) Multilingual Minds and Machines Meeting! Come join us in Nijmegen, June 22-23, 2026, if you are interested in computational models of human multilingualism: mmmm2026.github.io

Did you know?

❌77% of language models on @hf.co are not tagged for any language
📈For 95% of languages, most models are multilingual
🚨88% of models with tags are trained on English

In a new blog post, @tylerachang.bsky.social and I dig into these trends and why they matter! 👇

Our new lab for Human & Machine Intelligence is officially open at Princeton University!

Consider applying for a PhD or Postdoc position, either through Computer Science or Psychology. You can register interest on our new website lake-lab.github.io (1/2)