Zhenmei Shi

Featured in: Favicon

arxiv.org

Articles

[2503.11108] Limits of KV Cache Compression for Tensor Attention based Autoregressive Transformers

1 month ago | arxiv.org | Xiaoyu Li |Yingyu Liang |Zhenmei Shi |Zhao Song
[2503.09069] Theoretical Guarantees for High Order Trajectory Refinement in Generative Flows

1 month ago | arxiv.org | Xiaoyu Li |Yingyu Liang |Jiangxuan Long |Zhenmei Shi

Today Hugging Face is TRIPLING your donation to arXiv with a 2:1 match! Give today to keep science open for all.
Dissecting Submission Limit in Desk-Rejections: A Mathematical Analysis of Fairness in AI Conference Policies

2 months ago | arxiv.org | Xiaoyu Li |Yingyu Liang |Zhizhou Sha |Zhenmei Shi
On the Computational Capability of Graph Neural Networks: A Circuit Complexity Bound Perspective

Jan 11, 2025 | arxiv.org | Yingyu Liang |Zhenmei Shi |Zhao Song |Wei Wang
On Computational Limits and Provably Efficient Criteria of Visual Autoregressive Models: A Fine-Grained Complexity Analysis

Jan 7, 2025 | arxiv.org | Xiaoyu Li |Yingyu Liang |Zhizhou Sha |Zhenmei Shi
[2501.04299] Circuit Complexity Bounds for Visual Autoregressive Model

Jan 7, 2025 | arxiv.org | Xiaoyu Li |Yingyu Liang |Zhenmei Shi |Zhao Song

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
[2412.06148] The Computational Limits of State-Space Models and Mamba via the Lens of Circuit Complexity

Dec 9, 2024 | arxiv.org | Xiaoyu Li |Yingyu Liang |Zhenmei Shi |Zhao Song
[2411.07602] Circuit Complexity Bounds for RoPE-based Transformer Architecture

Nov 12, 2024 | arxiv.org | Xiaoyu Li |Yingyu Liang |Zhenmei Shi |Jiangxuan Long
Bypassing the Exponential Dependency: Looped Transformers Efficiently Learn In-context by Multi-step Gradient Descent

Oct 15, 2024 | arxiv.org | Xiaoyu Li |Yingyu Liang |Zhenmei Shi |Zhao Song
Advancing the Understanding of Fixed Point Iterations in Deep Neural Networks: A Detailed Analytical Study

Oct 15, 2024 | arxiv.org | Xiaoyu Li |Yingyu Liang |Zhenmei Shi |Zhao Song

Contact details

Emails

[email protected]

Socials & Sites

Try JournoFinder For Free

Search and contact over 1M+ journalist profiles, browse 100M+ articles, and unlock powerful PR tools.

Start Your 7-Day Free Trial →

Zhenmei Shi

Articles

[2503.11108] Limits of KV Cache Compression for Tensor Attention based Autoregressive Transformers

[2503.09069] Theoretical Guarantees for High Order Trajectory Refinement in Generative Flows

Dissecting Submission Limit in Desk-Rejections: A Mathematical Analysis of Fairness in AI Conference Policies

On the Computational Capability of Graph Neural Networks: A Circuit Complexity Bound Perspective

On Computational Limits and Provably Efficient Criteria of Visual Autoregressive Models: A Fine-Grained Complexity Analysis

[2501.04299] Circuit Complexity Bounds for Visual Autoregressive Model

[2412.06148] The Computational Limits of State-Space Models and Mamba via the Lens of Circuit Complexity

[2411.07602] Circuit Complexity Bounds for RoPE-based Transformer Architecture

Bypassing the Exponential Dependency: Looped Transformers Efficiently Learn In-context by Multi-step Gradient Descent

Advancing the Understanding of Fixed Point Iterations in Deep Neural Networks: A Detailed Analytical Study

Contact details

Emails

Socials & Sites