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

Ting Liu

3 indexed papers

Recent (6 mo)
3
With code
0
Influential cites
0
Benchmarked
0

Publications per year

3
26

Top categories

AI×2Crypto×2Software Eng.×1NLP×1

Frequent co-authors

Yang He1×
Xiao Ding1×
Bibo Cai1×
Yufei Zhang1×
Kai Xiong1×
Zhouhao Sun1×

Research Timeline

2026
Stealthy Backdoor Attacks against LLMs Based on Natural Style Triggers

The paper introduces BadStyle, a novel backdoor attack framework that generates natural, stealthy poisoned samples using LLMs to compromise various LLMs with high success rates and robust activation.

Usability as a Weapon: Attacking the Safety of LLM-Based Code Generation via Usability Requirements

This paper introduces UPAttack, a novel threat model demonstrating that focusing on explicit usability requirements can cause LLMs to generate insecure code by neglecting implicit security constraints, and proposes U-SPLOIT to automate this attack.

DeepTool: Scaling Interleaved Deliberation in Tool-Integrated Reasoning via Process-Supervised Reinforcement Learning

DeepTool introduces a novel Process-Supervised Reinforcement Learning framework to enhance Tool-Integrated Reasoning by explicitly supervising and rewarding intermediate, interleaved deliberation steps during sequential tool use.

Highlighted terms show continued research focus across papers

Papers

cs.AIRecentMay 28, 2026

DeepTool: Scaling Interleaved Deliberation in Tool-Integrated Reasoning via Process-Supervised Reinforcement Learning

Yang He, Xiao Ding, Bibo Cai, Yufei Zhang +4 more

DeepTool introduces a novel Process-Supervised Reinforcement Learning framework to enhance Tool-Integrated Reasoning by explicitly supervising and rewarding intermediate, interleaved deliberation step…

View →
cs.CRcs.SERecentMay 11, 2026

Usability as a Weapon: Attacking the Safety of LLM-Based Code Generation via Usability Requirements

Yue Li, Xiao Li, Hao Wu, Yue Zhang +4 more

This paper introduces UPAttack, a novel threat model demonstrating that focusing on explicit usability requirements can cause LLMs to generate insecure code by neglecting implicit security constraints…

View →
cs.CRcs.AIcs.CLRecentApr 23, 2026

Stealthy Backdoor Attacks against LLMs Based on Natural Style Triggers

Jiali Wei, Ming Fan, Guoheng Sun, Xicheng Zhang +2 more

The paper introduces BadStyle, a novel backdoor attack framework that generates natural, stealthy poisoned samples using LLMs to compromise various LLMs with high success rates and robust activation.

View →