12 papers
One mechanism for many mental spaces: a shared router over a value slot in language models
Oliver Steele, Jiangtao Wen, Yuxing Han
Language builds discourse contexts other than the actual: a painting, a belief, a memory, a hypothetical. Each is a mental space in which the same entity can take a different value…
Belief-reality separation lives in routing over a shared value slot in language models
Oliver Steele, Jiangtao Wen, Yuxing Han
Capable language models hold what a character believes apart from what is true: told "Anna believes the cup is blue; in reality it is red," they answer blue about Anna and red abou…
RAL2M: Retrieval Augmented Learning-To-Match Against Hallucination in Compliance-Guaranteed Service Systems
Mengze Hong, Di Jiang, Jiangtao Wen +5
Hallucination is a major concern in LLM-driven service systems, necessitating explicit knowledge grounding for compliance-guaranteed responses. In this paper, we introduce Retrieva…
ReSeFlow: Rectifying SE(3)-Equivariant Policy Learning Flows
Zhitao Wang, Yanke Wang, Jiangtao Wen +2
Robotic manipulation in unstructured environments requires the generation of robust and long-horizon trajectory-level policy with conditions of perceptual observations and benefits…
Efficient Bayer-Domain Video Computer Vision with Fast Motion Estimation and Learned Perception Residual
Haichao Wang, Jiangtao Wen, Yuxing Han
Video computer vision systems face substantial computational burdens arising from two fundamental challenges: eliminating unnecessary processing and reducing temporal redundancy in…
It Takes a Good Model to Train a Good Model: Generalized Gaussian Priors for Optimized LLMs
Jun Wu, Patrick Huang, Jiangtao Wen +1
Despite rapid progress in large language models (LLMs), the statistical structure of their weights, activations, and gradients-and its implications for initialization, training dyn…
BackSlash: Rate Constrained Optimized Training of Large Language Models
Jun Wu, Jiangtao Wen, Yuxing Han
The rapid advancement of large-language models (LLMs) has driven extensive research into parameter compression after training has been completed, yet compression during the trainin…
Vision without Images: End-to-End Computer Vision from Single Compressive Measurements
Fengpu Pan, Heting Gao, Jiangtao Wen +1
Snapshot Compressed Imaging (SCI) offers high-speed, low-bandwidth, and energy-efficient image acquisition, but remains challenged by low-light and low signal-to-noise ratio (SNR)…
Leveraging Motion Estimation for Efficient Bayer-Domain Video Convolutional Networks
Haichao Wang, Xinyue Xi, Jiangtao Wen +1
Existing computer vision pipelines suffer from two efficiency bottlenecks: image signal processors (ISPs) first convert Bayer pixel data to RGB pixel by pixel, and video models rep…
Leveraging Compressed Frame Sizes For Ultra-Fast Video Classification
Yuxing Han, Yunan Ding, Chen Ye Gan +1
Classifying videos into distinct categories, such as Sport and Music Video, is crucial for multimedia understanding and retrieval, especially when an immense volume of video conten…
Judging a video by its bitstream cover
Yuxing Han, Yunan Ding, Jiangtao Wen +1
Classifying videos into distinct categories, such as Sport and Music Video, is crucial for multimedia understanding and retrieval, especially in an age where an immense volume of v…
Phocus: Picking Valuable Research from a Sea of Citations
Xinrong Zhang, Zihou Ren, Xi Li +5
The deluge of new papers has significantly blocked the development of academics, which is mainly caused by author-level and publication-level evaluation metrics that only focus on…