paper-conference

Not All Prefills Are Equal: PPD Disaggregation for Multi-turn LLM Serving
Prefill-Decode (PD) disaggregation has become the standard architecture for modern LLM inference engines, which alleviates the …
Zongze Li, Jingyu Liu, Zach Xu, Yineng Zhang, Tahseen Rabbani, Ce Zhang
PDF Cite Project
Not All Prefills Are Equal: PPD Disaggregation for Multi-turn LLM Serving
Efficient-DLM: From Autoregressive to Diffusion Language Models, and Beyond in Speed
Diffusion language models (dLMs) have emerged as a promising paradigm that enables parallel, non-autoregressive generation, but their …
Yonggan Fu, Lexington Whalen, Zhifan Ye, Xin Dong, Shizhe Diao, Jingyu Liu, Chengyue Wu, Hao Zhang, Enze Xie, Maksim Khadkevich, Jan Kautz, Yingyan Celine Lin, Pavlo
PDF Cite
Efficient-DLM: From Autoregressive to Diffusion Language Models, and Beyond in Speed
TiDAR: Think in Diffusion, Talk in Autoregression
Diffusion language models hold the promise of fast parallel generation, while autoregressive (AR) models typically excel in quality due …
Jingyu Liu, Xin Dong, Zhifan Ye, Rishabh Mehta, Yonggan Fu, Vartika Singh, Jan Kautz, Ce Zhang, Pavlo Molchanov
PDF Cite Website ASAP Seminar Talk Discrete Diffusion Reading Group Talk
TiDAR: Think in Diffusion, Talk in Autoregression
How Far Are We From AGI?
The evolution of artificial intelligence (AI) has profoundly impacted human society, driving significant advancements in multiple …
Tao Feng, Chuanyang Jin, Jingyu Liu, Kunlun Zhu, Haoqin Tu, Zirui Cheng, Guanyu Lin, Jiaxuan You
PDF Cite Project
How Far Are We From AGI?
Scene-LLM: Extending Language Model for 3D Visual Understanding and Reasoning
This paper introduces Scene-LLM, a 3D-visual-language model that enhances embodied agents’ abilities in interactive 3D indoor …
Rao Fu, Jingyu Liu, Xilun Chen, Yixin Nie, Wenhan Xiong
PDF Cite
Scene-LLM: Extending Language Model for 3D Visual Understanding and Reasoning
Effective Long-Context Scaling of Foundation Models
We present a series of long-context LLMs that support effective context windows of up to 32,768 tokens. Our model series are built …
Wenhan Xiong, Jingyu Liu, Igor Molybog, Hejia Zhang, Prajjwal Bhargava, Rui Hou, Louis Martin, Rashi Rungta, Karthik Abinav Sankararaman, Barlas Oğuz, Madian Khabsa, Han Fang, Yashar Mehdad, Sharan Narang, Kshitiz Malik, Angela Fan, Shruti Bhosale, Sergey Edunov, Mike Lewis, Sinong Wang, Hao Ma
PDF Cite Project
Effective Long-Context Scaling of Foundation Models
Text-guided 3D Human Generation from 2D Collections
3D human modeling has been widely used for engaging interaction in gaming, film, and animation. The customization of these characters …
Tsu-Jui Fu, Wenhan Xiong, Yixin Nie, Jingyu Liu, Barlas Oğuz, William Yang Wang
PDF Cite Project
Text-guided 3D Human Generation from 2D Collections