iiinova — where your digital self keeps evolving

Martin Ziqiao Ma @marstin.bsky.social · 12/05/2026

A glimpse of interaction models: early experiments with time-aligned micro-turns over native audio, video, and text streams, moving toward more seamless human-AI collaboration. thinkingmachines.ai/blog/interac...

thinkingmachines.ai

Interaction Models: A Scalable Approach to Human-AI Collaboration

Interaction models move beyond turn-based AI interfaces by handling multimodal, real-time collaboration natively across audio, video, and text.

021

Martin Ziqiao Ma @marstin.bsky.social · 15/04/2026

Thanks for sharing, LaCET is now open-sourced :)

010

Martin Ziqiao Ma @marstin.bsky.social · 28/03/2026

that’s why I still love my hand typed good old html homepage :(

010

Reposted by Martin Ziqiao Ma

Abdellah Fourtassi @fourtassi.bsky.social · 20/01/2026

Thrilled to announce the 1st Workshop on Computational Developmental Linguistics (CDL) at ACL 2026 🎉 A new venue at the intersection of development linguistics × modern NLP, spearheaded by @fredashi.bsky.social @marstin.bsky.social, and and outstanding team of colleagues! A thread 🧵

3219

Martin Ziqiao Ma @marstin.bsky.social · 20/12/2025

NEPA: Next-Embedding Predictive Autoregression sihanxu.me/nepa/ Key ideas: - One self-supervised signal: cosine-style next-embedding prediction - Autoregression runs directly on the model's native embeddings - No pixel decoder (& loss), no contrastive pairs, no task-specific heads, no random masks

sihanxu.me

NEPA

000

Martin Ziqiao Ma @marstin.bsky.social · 11/12/2025

In my new blog, “Test-Time Training Done Better: From Plastic Adaptation to Elastic Memory Consolidation,” I introduce a long-context modeling architecture that learns to adapt and memorize at test time by updating a subset of the model’s weights during inference. mars-tin.github.io/blogs/posts/...

mars-tin.github.io

Test-Time Training Done Better: From Plastic Adaptation to Elastic Memory

Elastic Test-Time Training (ETTT) that prevents catastrophic forgetting at inference time and overfitting during pretraining LaCT.

000

Martin Ziqiao Ma @marstin.bsky.social · 24/11/2025

Gosh, I’m getting way too emotional writing my thesis acknowledgements...

010

Martin Ziqiao Ma @marstin.bsky.social · 23/11/2025

Will be at #NeurIPS2025 (San Diego) Dec 1-9, then in the Bay Area until the 14th. Hmu if you wanna grab coffee and talk about totally random stuff. Thread with a few things I’m excited about. P.S. 4 NeurIPS papers all started pre-May 2024 and took ~1 year of polishing...so proud of the team!

000

Reposted by Martin Ziqiao Ma

Michael Saxon @saxon.me · 18/11/2025

Trying to decide what to do on the first day of #NeurIPS2025? Check out my, @marstin.bsky.social and @xiangyue96.bsky.social's tutorial, "The Science of Benchmarking: What's Measured, What's Missing, What's Next" on December 2 from 1:30 to 4:00pm. benchmarking.science What will we cover? 1/3

2214

Martin Ziqiao Ma @marstin.bsky.social · 21/10/2025

@fredashi.bsky.social and I wrote a blog for our new mechinterp paper (arxiv.org/abs/2510.13796), including many unpublished and even negative results that we found meaningful to share. An Open-Notebook Exploration of Emergent Grounding in LMs mars-tin.github.io/blogs/posts/...

mars-tin.github.io

An Open-Notebook Exploration of Emergent Grounding in LMs

How We Did This Curiosity-Driven Research? An Open-Notebook Exploration of Emergent Grounding in LMs

070

Martin Ziqiao Ma @marstin.bsky.social · 05/10/2025

Regrettably can’t attend #COLM2025 due to deadlines, but Jane and Joyce will be presenting our work. :) Jane is an exceptional undergraduate researcher and a great collaborator! Go meet her at COLM if you’re curious about her work on mechanistic interpretability, multimodality, & pragmatics!

020

Reposted by Martin Ziqiao Ma

Freda Shi @fredashi.bsky.social · 29/09/2025

🚀 ACL ARR is looking for a Co-CTO to join me lead our amazing tech team and drive the future of our workflow. If you’re interested or know someone who might be, let’s connect! RTs & recommendations appreciated.

143

Martin Ziqiao Ma @marstin.bsky.social · 28/07/2025

Unfortunately, I’ll be missing #ACL2025NLP this year — but here are a few things I’m excited about! 👇

010

Martin Ziqiao Ma @marstin.bsky.social · 14/07/2025

📣 Excited to announce SpaVLE: #NeurIPS2025 Workshop on Space in Vision, Language, and Embodied AI! Join us in San Diego to push the frontiers of spatial understanding and reasoning across CV, NLP, and robotics! 👉 space-in-vision-language-embodied-ai.github.io

120

Reposted by Martin Ziqiao Ma

Hokin @hokin.bsky.social · 30/06/2025

#CoreCognition #LLM #multimodal #GrowAI We spent 3 years to curate 1503 classic experiments spanning 12 core concepts in human cognitive development and evaluated on 230 MLLMs with 11 different prompts for 5 times to get over 3.8 millions inference data points. A thread (1/n) - #ICML2025 ✅

1139

Reposted by Martin Ziqiao Ma

Hokin @hokin.bsky.social · 12/06/2025

New Paper Alert ‼️ Current VLMs completely fail human gaze understanding 🙀 and scaling does NO help ‼️ However, humans, since an extremely age 🧒, are extremely sensitive to other people's gaze 🙄 👀 No mentors, no labs, only pre-doc students, 111 VLMs, and we did it 😎

165

Reposted by Martin Ziqiao Ma

JHU Computer Science @jhucompsci.bsky.social · 10/06/2025

& @tianminshu.bsky.social (+ @marstin.bsky.social, @zhitinghu.bsky.social, ‪@lianhui.bsky.social & more) will present “SimWorld: A World Simulator for Scaling Photorealistic Multi-Agent Interactions,” an @unrealengine.bsky.social-based sim that generates unlimited/diverse urban environments: (13/14)

simworld-cvpr2025.maitrix.org

SimWorld

SimWorld: A World Simulator for Scaling Photorealistic Multi-Agent Interactions

111

Martin Ziqiao Ma @marstin.bsky.social · 30/04/2025

See you at #NAACL2025! I will talk about grounded lexicon acquisition and scaling mechanistically grounded vision language models. Happy to chat if you are around :)

010

Martin Ziqiao Ma @marstin.bsky.social · 23/04/2025

We introduce RefOI, a new dataset of 1.5k objects, each with 3 written and 2 spoken human-produced referring expressions. We also release RefOI-TLHF, a large dataset of token-level human feedback for 10.6k referring expressions. 👀https://vlm-reg.github.io/ 📄https://arxiv.org/abs/2504.16060

vlm-reg.github.io

VLMs Are Not Pragmatically Competent in Referring Expression Generation

VLMs fail to refer like humans. Our study reveals widespread pragmatic issues in GPT-4o, LLaVA, and others, showing how their expressions often violate Gricean maxims.

010

Martin Ziqiao Ma @marstin.bsky.social · 23/04/2025

Vision-Language Models are not yet pragmatically optimal. We identify 3 key failures of pragmatic competence in referring expression generation with VLMs: (1) cannot uniquely refer to the referent, (2) include excessive or irrelevant information, and (3) misalign with human pragmatic preferences.

143

Martin Ziqiao Ma @marstin.bsky.social · 19/04/2025

🔹 Workshop Paper at World Models: Do Vision-Language Models Have Internal World Models? 🗓 Apr 27, 9 p.m. (Peridot 201&206) Paper: openreview.net/forum?id=tpP... Excited for this collaboration with MaitrixOrg, details coming soon :)

openreview.net

Do Vision-Language Models Have Internal World Models? Towards an...

Internal world models (WMs) enable agents to understand the world's state and predict transitions, serving as the basis for advanced deliberative reasoning. Recent large Vision-Language Models...

010

Martin Ziqiao Ma @marstin.bsky.social · 19/04/2025

🔹 ICLR BiAlign Workshop: We’re hosting the Bidirectional Human-AI Alignment Workshop (BiAlign). 🗓 Apr 28, (Garnet 216–214) Website: bialign-workshop.github.io I’ll join remotely — huge thanks to @huashen.bsky.social for leading this!

140

Martin Ziqiao Ma @marstin.bsky.social · 19/04/2025

🔹 ICLR Oral Paper: Do Vision-Language Models Represent Space and How? 🗓 Oral: Apr 25, 3:42–3:54 a.m. (Session 4C) 🗓 Poster: Thu, Apr 24, 10 p.m.–12:30 a.m. (Hall 3 + 2B, #212) Website: spatial-comfort.github.io Big thanks to @fredashi.bsky.social for presenting on site!

100

Martin Ziqiao Ma @marstin.bsky.social · 19/04/2025

I won’t be attending #ICLR2025 in person since #NAACL2025 follows right after, but here are a few things I’m excited about (all time in EDT) ⬇️

100

Martin Ziqiao Ma @marstin.bsky.social · 15/04/2025

🎉 Out of these, 72 papers were accepted, including 5 tiny papers. 10 papers were selected for oral presentations: 2 at CHI and 8 at ICLR. Award winners will be announced during the workshop!

100

Martin Ziqiao Ma @marstin.bsky.social · 15/04/2025

📬 We received over 100 submissions, each reviewed by 2–4 expert reviewers, with ethical assessments included when appropriate. Our program committee features leading researchers in NLP, RL, HCI, ML, and AI/ML Ethics, carefully selected based on scholarly merit and expertise.

100

Martin Ziqiao Ma @marstin.bsky.social · 15/04/2025

🙏 Special thanks to Tammy Masterson, Technical Partnerships Lead at the AI Security Institute, who will be joining us as a panelist.

100

Martin Ziqiao Ma @marstin.bsky.social · 15/04/2025

🙏 We are grateful to our gold sponsors, Prolific and Layer 6 AI of TD Bank Group, for their generous support in funding paper awards and travel grants.

100

Martin Ziqiao Ma @marstin.bsky.social · 15/04/2025

#ICLR2025 and #CHI2025 are just around the corner! We warmly invite you to join us at our ICLR Workshop and CHI SIG on Bidirectional Human-AI Alignment (Bi-Align), a space for rigorous and reflective conversations of alignment research.

130

Reposted by Martin Ziqiao Ma

aclmentorship.bsky.social @aclmentorship.bsky.social · 04/04/2025

📢 Join us for the ACL Mentorship Session @naaclmeeting.bsky.social #NAACL2025 Mentors: • @amuuueller.bsky.social • @fredashi.bsky.social • Jiayuan Mao • @marstin.bsky.social • Oana Ignat • Weijia Shi • @zhijingjin.bsky.social

122

Martin Ziqiao Ma @marstin.bsky.social · 19/03/2025

Meet VEGGIE 🥦 VEGGIE is an instructional video generative model trained solely with diffusion loss, designed for both video concept grounding and instruction-based editing. It effectively handles diverse video editing tasks by pixel-level grounded training in a multi-task learning setup. ⬇️

000

Martin Ziqiao Ma @marstin.bsky.social · 07/03/2025

Let’s conduct another community study by annotating false claim attributions and citations in AI papers.🐸

010

Reposted by Martin Ziqiao Ma

Yoav Artzi @yoavartzi.com · 14/02/2025

I am looking for a postdoc. A serious-looking call coming soon, but this is to get it going. Topics include (but not limited to): LLMs (🫢!), multimodal LLMs, interaction+learning, RL, intersection with cogsci, ... see our work to get an idea: yoavartzi.com/pubs Plz RT 🙏

yoavartzi.com

Publications

12413

Martin Ziqiao Ma @marstin.bsky.social · 07/02/2025

Our #ICLR2025 workshop on Bidirectional Human-AI Alignment now has a sister SIG at #CHI2025! 📅 Submission deadline extended to Feb 15 to welcome both ML & HCI communities. 📍 Authors may opt to present at either ICLR or CHI 2025! More info here: bialign-workshop.github.io

011

Martin Ziqiao Ma @marstin.bsky.social · 28/01/2025

[RT Appreciated] I'm excited to join #ACL @aclmentorship.bsky.social as a mentor! Your ideas can help make mentorship more impactful. Let’s plan together!🚀

000

Martin Ziqiao Ma @marstin.bsky.social · 14/01/2025

Submission #1 is in the bag, but we know your masterpiece is next, don’t keep us waiting. 😎 Also, consider joining the program committee and help shape the future of alignment research. Your reviews might just steer the field! forms.gle/MapXyCZhbcFr...

forms.gle

ICLR 2025 Bi-Align Workshop Reviewer Nomination Form

We strive to expand our reviewing pool by welcoming newer members of the community. We encourage nominations from senior community members as well as self-nominations from individuals who have either ...

020

Reposted by Martin Ziqiao Ma

Freda Shi @fredashi.bsky.social · 27/12/2024

RepL4NLP (collocated with NAACL 2025) is inviting submissions!

192

Martin Ziqiao Ma @marstin.bsky.social · 09/12/2024

Excited to co-organize the #ICLR2025 Workshop on Bidirectional Human-AI Alignment (Bi-Align). Big thanks to an amazing team and speaker lineup! We call for non-archival papers: 2-page (tiny), 4-page (short), or 9-page (long). 👉 openreview.net/group?id=ICLR.… See you in Singapore :)

openreview.net

030

Martin Ziqiao Ma @marstin.bsky.social · 08/12/2024

I first met Jaemin at the Microsoft Accelerate Foundation Models Research (AFMR) dinner during NeurIPS 2023, though I was already familiar with his papers long before that. He’s definitely a strong candidate to consider!

172

Reposted by Martin Ziqiao Ma

Freda Shi @fredashi.bsky.social · 04/12/2024

I'll be in Vancouver for NeurIPS from Dec 10 to 14. Happy to chat with people! Find me at - Dec 10: Future NLP Workshop with UBC NLP - Dec 12: Vector Luncheon - Dec 14: Pluralistic Alignment Workshop, presenting the VLM frame of reference analysis work with the amazing @marstin.bsky.social

093

Martin Ziqiao Ma @marstin.bsky.social · 02/12/2024

A few updates :) 1. Presenting perspective-taking in machine theory of mind at the UMich tomorrow. 2. Visiting the UW on Dec 5 to talk about language grounding, let’s connect if you’re in Seattle. 3. Attending #NeurIPS2024 in Vancouver with a conference and workshop paper, DMs open!

060

Martin Ziqiao Ma @marstin.bsky.social · 29/11/2024

Shizhe presented Hymba (arxiv.org/abs/2411.13676), an innovative family of small language models that utilize a hybrid-head parallel architecture, combining transformer attention mechanisms with state space models (SSMs) for improved efficiency. 👀 cse.engin.umich.edu/event/small-...

cse.engin.umich.edu

Small Hybrid Language Model

010

Martin Ziqiao Ma @marstin.bsky.social · 29/11/2024

Prof Yu Su's talk discusses web agents as a new form of embodied AI powered by LLMs, navigating GUIs and markup languages. It emphasizes multimodality, grounding, and future directions like planning, synthetic data, and safety. 👀 cse.engin.umich.edu/event/yu-su-...

cse.engin.umich.edu

Web Agents: A New Frontier for Embodied Agents

120