Han Lin

@hanlin_hl

PhD @UNC | MS @Columbia | Research Intern @AIatMeta| Working on generative models, multimodal learning, and LLMs

Chapel Hill, NC

hl-hanlin.github.io

Joined December 2021

1,024 Following

758 Followers

Replies Reposts Media

Pinned
Han Lin
@hanlin_hl
Mar 18
🚀 Excited to share V-Co, a diffusion model that jointly denoises pixels and pretrained semantic features (e.g., DINO). We find a simple but effective recipe: 1️⃣ architecture matters a lot --> fully dual-stream JiT 2️⃣ CFG needs a better unconditional branch -->
3
Han Lin
@hanlin_hl
Oct 15
What if an agent’s failures could shape its own path to self-improvement? 🤔 In WEDGE, we take a step in this direction: turning failures into targeted, learnable training worlds that evolve as the agent improves!
Yue Zhang
@zhan1624
Oct 15
🚨 Excited to share WEDGE: From Failures to Learnable Worlds, our new work on policy-conditioned environment generation for navigation agents! We found that even on the same navigation task, agents tend to fail under different environmental conditions. Yet most environment
00:00
Han Lin
@hanlin_hl
Oct 8
Great and timely analysis for GPT-6 Astra as robotics control policy!
Lin Long
@Kylin_0Long
Oct 8
GPT-6 Astra has shown impressive results in simple robot control tasks. But what if the tasks become more challenging and involve complex geometric constraints, precise contact, or improvised tool use? To find out, we: - Built Ditto-Bench, where simple goals meet challenging
Han Lin
@hanlin_hl
Oct 6
Glad to announce that MetaCanvas has been accepted to #NeurIPS2026! ✨ MetaCanvas lets MLLMs reason and plan directly in spatial and spatiotemporal latent spaces, instead of reducing them to global text encoders for diffusion models. A set of learnable "canvas tokens" carries
Han Lin
@hanlin_hl
Dec 16, 2025
Multimodal LLMs (MLLMs) excel at reasoning, layout understanding, and planning—yet in diffusion-based generation, they are often reduced to simple multimodal encoders. What if MLLMs could reason directly in latent space and guide diffusion generation with fine-grained,
00:00
1
Han Lin
@hanlin_hl
Sep 9
Excited to present 3 papers in person at Malmo #ECCV2026! 🇸🇪✨ Our work spans consistent world generation, error-driven 3D grounding, and visual representation learning. Huge congrats to all collaborators! Feel free to reach out and chat about the work!
4