Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Hao Shi

Hao Shi

2 indexed papers

Recent (6 mo)
2
With code
0
Influential cites
0
Benchmarked
0

Publications per year

2
26

Top categories

AI×1ML×1Crypto×1

Frequent co-authors

Wenhao Liu1×
Yunhe Li1×
Weizhi Fei1×
Xiangyuan Wang1×
Mengzhe Ruan1×
Hanxu Hou1×

Research Timeline

2026
Low Rank Adaptation for Adversarial Perturbation

This paper demonstrates that adversarial perturbations possess a low-rank structure, and proposes a two-step method to leverage this property to significantly improve the efficiency and effectiveness of black-box adversarial attacks.

ReasonAlloc: Hierarchical Decoding-Time KV Cache Budget Allocation for Reasoning Models

This paper proposes a training-free framework called ReasonAlloc to mitigate inference bottlenecks in large language models by recasting decoding-time key-value compression as a hierarchical budget allocation problem.

Highlighted terms show continued research focus across papers

Papers

cs.AIEmpiricalRecentJun 9, 2026

ReasonAlloc: Hierarchical Decoding-Time KV Cache Budget Allocation for Reasoning Models

Wenhao Liu, Hao Shi, Yunhe Li, Weizhi Fei +6 more

This paper proposes a training-free framework called ReasonAlloc to mitigate inference bottlenecks in large language models by recasting decoding-time key-value compression as a hierarchical budget al…

View →
cs.LGcs.CRRecentApr 30, 2026

Low Rank Adaptation for Adversarial Perturbation

Han Liu, Shanghao Shi, Yevgeniy Vorobeychik, Chongjie Zhang +1 more

This paper demonstrates that adversarial perturbations possess a low-rank structure, and proposes a two-step method to leverage this property to significantly improve the efficiency and effectiveness…

View →