News

  1. Thinking with Anchors, our work on grounded and efficient document reasoning with ADOPD 2026, is now available on arXiv.
  2. The interactive demo for Context Learning is now available.
  3. Chimera, our hybrid visual diffusion transformer with a Chinchilla-style scaling recipe, is now available.
  4. FLARE, our framework for converting hybrid autoregressive models into diffusion language models, is now available.
  5. LaViDa-O, our unified diffusion model for multimodal understanding and generation, appears at ICLR 2026.
  6. LaViDa-R1, our reasoning model for unified multimodal diffusion, appears at ICML 2026.
  7. MENTOR, an efficient framework for controllable multimodal image generation, appears in Findings of ACL 2026.
  8. Doc-AGround, our work on visual grounding for text-rich images, appears in Findings of ACL 2026.
  9. UniTemp, a framework for video generation in any temporal order, is now available.
  10. Sparse-LaViDa, our efficient sparse multimodal diffusion model, appears at CVPR 2026.
  11. FastCar, our work on efficient autoregressive video generation for edge devices, appears at ICLR 2026.
  12. OIDA-QA, our multimodal benchmark for opioid-industry document analysis, appears at AAAI 2026.
  13. Two papers get accepted to ICLR 2025.
  14. One paper gets accepted to NAACL 2025.
  15. Two papers get accepted to AAAI 2025.
  16. Two papers get accepted to WACV 2025.
  17. Two papers get accepted to EMNLP 2024.
  18. One paper gets accepted to ACL Findings 2024.
  19. One paper gets accepted to ICML 2024.
  20. One paper gets accepted to NAACL 2024.
  21. One paper get accepted to ICLR (Mathematical and Empirical Understanding of Foundation Models) 2024.
  22. Two papers get accepted to CVPR 2024.
  23. Three papers (ADoPD, LRM, SOHES) get accepted to ICLR 2024 (1 Oral, 2 Poster).
  24. Two papers (Document OOD Detection and Document CLIP) get accepted by EMNLP 2023.
  25. One paper on entity segmentation gets accepted to NeurIPS 2023 (Spotlight).
  26. One paper gets accepted to ICCV 2023 (Oral).
  27. I will be serving as an Area Chair of WACV 2024.
  28. One paper gets accepted to AAAI 2023.
  29. One paper (Entity Segmentation) gets accepted to PAMI.
  30. One paper gets accepted to AAAI 2023.
  31. One paper (Differential Privacy in Learning Language Models) gets accepted to BigData 2022.
  32. One paper gets accepted to WACV 2023.
  33. Our most recent Doc-series paper (SelfDoc->UniDoc->MGDoc) gets accepted to EMNLP 2022.
  34. One paper (OOD Detection) gets accepted to NeurIPS 2022.
  35. Three papers get accepted to ECCV 2022.
  36. One paper gets accepted to Interspeech 2022.
  37. One paper gets accepted to NAACL 2022.
  38. Three papers get accepted to CVPR 2022.
  39. One paper (Adaptive Axis Attentions) gets accepted to ACL 2022 Findings.
  40. One paper gets accepted to WWW 2022.
  41. Two papers get accepted to AAAI 2022.
  42. The second Doc-series paper (SelfDoc->UniDoc) gets accepted to NeurIPS 2021.
  43. One paper gets accepted to NAACL 2021.
  44. Three papers (SelfDoc is our first Doc-series paper) get accepted to CVPR 2021.
  45. My FIRST summer intern Peizhao’s work on “Document Representation Learning” got accepted by CVPR 2021.
  46. One paper gets accepted to NeurIPS 2020.
  47. One paper on video captioning gets accepted to Neurcomputing.
  48. One paper gets accepted to ECCV 2020.
  49. Started work at Adobe Research.
  50. Successfully passed my oral defense!
  51. One paper gets accepted to ICCV 2019.
  52. One paper gets accepted to ACM MM 2019.
  53. One paper gets accepted to CVPR 2019.