Zobrazeno 1 - 10
of 112,938
for search: '"Liu, Yang"'
Author/Creator:
Liu, Yang
Many grand societal challenges are rooted in the need for new materials, such as those related to energy, health, and the environment. However, the traditional way of discovering new materials is basically trial and error. This time-consuming and exp
Externí odkaz:
https://hdl.handle.net/10919/119128
This open access book focuses on the practical application of Adaptive Dynamic Programming (ADP) in chemotherapy drug delivery, taking into account clinical variables and real-time data. ADP's ability to adapt to changing conditions and make optimal
Externí odkaz:
https://library.oapen.org/handle/20.500.12657/76796
Author/Creator:
Hajimolahoseini, Habib, Hassanpour, Mohammad, Ataiefard, Foozhan, Chen, Boxing, Liu, Yang
This paper introduces a novel method of Progressive Low Rank Decomposition (PLRD) tailored for the compression of large language models. Our approach leverages a pre-trained model, which is then incrementally decompressed to smaller sizes using progr
Externí odkaz:
http://arxiv.org/abs/2406.19995
Author/Creator:
Liu, Yang
In our previous paper [1, 2], we proposed a probabilistic argument to explain the reason why the cosmological constant is very small in 4D. It is natural to ask if the result can be generalized to D-dimension. Moreover, in higher dimensional theory m
Externí odkaz:
http://arxiv.org/abs/2406.19451
By setting the inverse temperature $\beta$ loose to occupy the complex plane, Michael E. Fisher showed that the zeros of the complex partition function $Z$, if approaching the real $\beta$ axis, reveal a thermodynamic phase transition. More recently,
Externí odkaz:
http://arxiv.org/abs/2406.18981
Representation learning in sequential recommendation is critical for accurately modeling user interaction patterns and improving recommendation precision. However, existing approaches predominantly emphasize item-to-item transitions, often neglecting
Externí odkaz:
http://arxiv.org/abs/2406.18470
Recently, the concept of embodied intelligence has been widely accepted and popularized, leading people to naturally consider the potential for commercialization in this field. In this work, we propose a specific commercial scenario simulation, human
Externí odkaz:
http://arxiv.org/abs/2406.17898
Video Moment Retrieval (VMR) aims to localize a specific temporal segment within an untrimmed long video given a natural language query. Existing methods often suffer from inadequate training annotations, i.e., the sentence typically matches with a f
Externí odkaz:
http://arxiv.org/abs/2406.17880
Scaling up neural networks has been a key recipe to the success of large language and vision models. However, in practice, up-scaled models can be disproportionately costly in terms of computations, providing only marginal improvements in performance
Externí odkaz:
http://arxiv.org/abs/2406.17117
When machine learning (ML) models are used in applications that involve humans (e.g., online recommendation, school admission, hiring, lending), the model itself may trigger changes in the distribution of targeted data it aims to predict. Performativ
Externí odkaz:
http://arxiv.org/abs/2406.16756