Justin Lin @ ICML 2026 🇰🇷

Master’s student in CS@CMU

[Justin Lin @ ICML 2026 🇰🇷 @justin77lin]

@kudareaper This is how ik you don’t know ball. This not kiss of life 🤨🤨

[Justin Lin @ ICML 2026 🇰🇷 @justin77lin]

My #ICML2026 experience😁😁

[Lawrence Jang @JangLawrenceK]

guys I just cancelled my Claude plan I don’t know what happened

[Jerry Huang @jrrhuang]

Happening today at Hall A Poster #2408! 🤩

[Sander Dieleman @sedielem]

📢 Diffusion circle at @icmlconf 2026: join us Thursday July 9 at 3:30PM at the information desk / job board, we'll head out from there and find a spot to sit. No agenda, just get together and talk shop.

Please tell your friends and tag people who might be interested!

[Jerry Huang @jrrhuang]

I’ll be presenting “How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance” at ICML this week! Swing by if you’re curious how to make test-time alignment / reward guidance for flows super fast 😎

👉 Thu, Jul 9, 5:00–6:45 PM KST, Hall A, Poster #2408

Also happy to connect if you're thinking about fast inference, flows / diffusion LLMs, multimodality, and what's next for generative modeling!

[Justin Lin @ ICML 2026 🇰🇷 @justin77lin]

Arrived in Seoul 🇰🇷 #ICML2026 presenting our paper on guidance for Flow Map models(https://t.co/uYlPyuTUPw)
🗓️Thu, Jul 9, 2026 • 5:00 PM – 6:45 PM
📍KSTHALL A #2408
If interested in flows, diffusions, and dynamical systems my DMs are open would love to chat!

[Google DeepMind @GoogleDeepMind]

DiffusionGemma is our new experimental open model with up to 4x faster output on dedicated GPUs.

Instead of predicting word-by-word, it generates entire blocks of text simultaneously. This lets the model self-correct and format complex markdown in real time.

[Jerry Huang @jrrhuang]

Cool to see guidance applied to flow policies! This looks like an offline-RL instantiation of our FMRG-E framework, which also drops the Jacobian, with a learned critic serving as the reward and the denoiser estimate approximating the endpoint lookahead.
https://t.co/7TWZ3v8Tlw

[Sanjit Dandapanthula @sanjitdp]

super excited to share our latest work! are we really tilting? 🤨

tldr: reward guidance for flows and diffusions is supposed to sample from the reward-tilted distribution. we show it doesn’t 😰 and how to (mostly) fix it ✨

plus lots of fun images!! 🖼️

collaboration with the awesome @nmboffi

website: https://t.co/nvOaAiGYq1
paper: https://t.co/EtkeyiuX7s
code: https://t.co/V3Bi4IVPbf

[Jerry Huang @jrrhuang]

Can we guide flow models in just a few steps? 🚀

Flow-based sampling is rapidly moving toward few-step generation. But reward guidance often still requires many steps and costly test-time search.

Excited to introduce Flow Map Reward Guidance (FMRG): a training-free framework for few-step guidance with flow maps.

FMRG matches or surpasses strong baselines on inverse problems and reward-guided text-to-image generation with:
⚡ as few as 3 NFEs
⚡ up to 10× fewer NFEs on inverse problems
⚡ up to 70× fewer NFEs on reward-guided generation

🧵⬇️