Xavier Thomas (Rohan)

Grad student at BU

[Deepti Ghadiyaram @deeptigp]

Do VLMs truly reason or memorize data? We show
❗even transformations like rotation & scaling still throw them off
❗finetuning fails to generalize across domains
https://t.co/EIfNtDRby5
📅 Stop by our poster on Wed @ 4:30 PM and
chat with @xavierohan #COLM2026 @COLM_conf

[Deepti Ghadiyaram @deeptigp]

Thank you @GoodfireAI for supporting my research group's research on interpretability of multi-modal LLMs. Excited to use Silico! 🚀

[Liam Gallagher @liamgallagher]

SPIRITUAL

[Xavier Thomas (Rohan) @xavierohan]

🥹

[Arsenal @Arsenal]

This belongs to all of us.

[GoonerNick @nickvann1987__]

Arsenal are the Premier League Champions. Every Arsenal fan right now 💪

[pH @pHequals7]

for the folks who know ball

[Arsenal @Arsenal]

The Arsenal. Your Premier League champions.

[Deepti Ghadiyaram @deeptigp]

🔥 [New Paper] Semantic Richness or Geometric Reasoning?
❗️GPT-5.2, Gemini-2.5-Pro struggle on basic geometric tasks when semantic context is sparse.
❗️Performance is tied to data familiarity.
Conclusion: We are far from true visual invariance.
More here: https://t.co/35n0fiPerd

[gokul@CoRL2026 @gokulp01]

Paper: https://t.co/ntR6RchmeL
Website (we have results on real-world real-time tasks on hardware and standard simulated benchmarks): https://t.co/9uG9jdzsWL
Code/models: coming next week!

[Deepti Ghadiyaram @deeptigp]

[1/2] The visual world is diverse—so why are the generative models stuck in a "one-patch-size-fits-all" architecture?

Meet Dynamic-DiT: A simple yet powerful technique that adaptively selects patch sizes based on how the latent manifold evolves at every denoising step.

[Thomas Fel @thomas_fel_]

Thrilled to share that our work investigating recurrence in trained ViTs has been accepted at #ICLR2026! 🎉

It turns out DINOv2 may be secretly recurrent.🦖

A 2-block recurrent transformer can match its activations layer-by-layer and recover 96% of its ImageNet performance.

🧵

[Deepti Ghadiyaram @deeptigp]

Grateful to be interviewed by @RSIPvision on my research journey so far https://t.co/UiP2MtsiHf

[Deepti Ghadiyaram @deeptigp]

[1/4] 🧠 The Great Modality Test: Are MLLMs Blind, Deaf, or Both?

Humans are perception pros: We instinctively know what information to trust.

Blindfolded, you describe the chirping birds.

Headphones on, you focus entirely on the silent visuals.

We robustly use the available and reliable modality.

Can SOTA MLLMs do the same?

We put the best Multimodal LLMs to the ultimate test.

[Deepti Ghadiyaram @deeptigp]

📢📢 Can we automate how humans instantly perceive if a generated video captures a human action done wrong?

Details: https://t.co/vV2T1EG0zX

Paper: https://t.co/18H8t5niJW

Our key idea is to build a manifold of correctly performed actions from several real-world videos. Next, we measure how far off a generated video is from this manifold to get a robust measurement of action correctness.

We perform extensive subjective, quantitative, and quantitative evaluation of different actions, generative models, and MLLMs.

Work done by @xavierohan Youngsun, Ananya, and Audrey

[Xavier Thomas (Rohan) @xavierohan]

@chanwoopark20 Huge congrats!!!

[Deepti Ghadiyaram @deeptigp]

Thrilled to receive BU's Career Development Professorship award Thanks, BU for this very kind acknowledgement! https://t.co/eM16VoxjYm

[Thomas Fel @thomas_fel_]

🕳️🐇Into the Rabbit Hull – Part I (Part II tomorrow)

An interpretability deep dive into DINOv2, one of vision’s most important foundation models.

And today is Part I, buckle up, we're exploring some of its most charming features.

[Xavier Thomas (Rohan) @xavierohan]

@gokulp01 Congrats!!

[Thomas Fel @thomas_fel_]

At ICML for the next 2 days to present multiple works, if you're into interpretability, complexity, or just wanna know how cool @KempnerInst is, hit me up 👋

[Oasis @oasis]

Because we need each other, we believe in one another #oasislive25