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

Natasha Jaques

2 indexed papers

Recent (6 mo)
1
With code
0
Influential cites
0
Benchmarked
0

Publications per year

2
26

Top categories

Architecture×1AI×1Crypto×1

Frequent co-authors

Zhihan Zhang1×
Alexander Le Metzger1×
Jiuyang Lyu1×
Chun-Cheng Chang1×
Jiayi Shao1×
Yujia Liu1×

Research Timeline

2026
FlowSteer: Prompt-Only Workflow Steering Exposes Planning-Time Vulnerabilities in Multi-Agent LLM Systems

The paper introduces FlowSteer, a prompt-only attack that exploits vulnerabilities in how multi-agent LLM systems plan workflows, significantly increasing the success rate of malicious signal propagation.

Embedded Arena: Iterative Optimization via Hardware Feedback

This paper introduces a hardware-in-the-loop agent that optimizes AI models for heterogeneous microcontrollers.

Highlighted terms show continued research focus across papers

Papers

cs.ARcs.AIEmpiricalRecentJun 15, 2026

Embedded Arena: Iterative Optimization via Hardware Feedback

Zhihan Zhang, Alexander Le Metzger, Jiuyang Lyu, Chun-Cheng Chang +9 more

This paper introduces a hardware-in-the-loop agent that optimizes AI models for heterogeneous microcontrollers.

View →
cs.CRMay 12, 2026

FlowSteer: Prompt-Only Workflow Steering Exposes Planning-Time Vulnerabilities in Multi-Agent LLM Systems

Fanxiao Li, Jiaying Wu, Tingchao Fu, Natasha Jaques +2 more

The paper introduces FlowSteer, a prompt-only attack that exploits vulnerabilities in how multi-agent LLM systems plan workflows, significantly increasing the success rate of malicious signal propagat…

View →