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

Zhong Li

10 indexed papers

Recent (6 mo)
9
With code
0
Influential cites
0
Benchmarked
0

Publications per year

10
26

Top categories

AI×6NLP×4Crypto×3Signal Processing×2Info Theory×2ML×2Info Retrieval×2Distributed×1

Frequent co-authors

Zhengzhong Liu2×
Tianhao Guo2×
Xianzhong Li2×
Hao Luo2×
Yuyang Gong2×
Miaokun Chen2×

Research Timeline

2026
Trojan's Whisper: Stealthy Manipulation of OpenClaw through Injected Bootstrapped Guidance

This paper identifies and characterizes 'guidance injection,' a stealthy attack vector that embeds adversarial operational narratives into autonomous coding agents' bootstrap guidance, demonstrating high success rates and evasion capabilities.

PR2: Predictive Routing Replay for MoE-Based LLM Reinforcement Learning

The paper proposes Predictive Routing Replay (PR2) to stabilize reinforcement learning on Mixture of Experts (MoE) LLMs by predicting and incorporating short-horizon router evolution during training and rollout.

D$^3$: Dynamic Directional Graph-Constrained Data Scheduling for LLM Training

The paper proposes $D^3$, a dynamic graph-constrained scheduling framework that optimizes LLM training order by modeling sample interactions as a dynamic influence graph.

Towards Efficient LLMs Annealing with Principled Sample Selection

The paper proposes DiReCT, a novel framework that treats data selection during LLM annealing as a constrained optimization problem based on the spectral geometry of the loss landscape, achieving state-of-the-art performance.

DiscourseFlip: An Oblique Discourse-Level Opinion Manipulation Attack against Black-box Retrieval-Augmented Generation

The paper introduces DiscourseFlip, a novel graph-guided attack that demonstrates how coordinated poisoning across a multi-topic query space can manipulate the overall opinion generated by black-box Retrieval-Augmented Generation (RAG) systems.

DiscourseFlip: An Oblique Discourse-Level Opinion Manipulation Attack against Black-box Retrieval-Augmented Generation

The paper introduces DiscourseFlip, a novel black-box, graph-guided attack that manipulates opinions across an entire multi-topic query network, demonstrating a significant leap in scope and effectiveness over existing RAG attack methods.

Gleam: Adaptive Network-Efficient CUDA API Remoting for Cross-Device GPU Sharing over LANs

This paper proposes Gleam, a framework for efficient GPU sharing across local-area CUDA devices, reducing bandwidth overhead, improving API call latency, and ensuring context consistency.

Negative Results for Reciprocity-Based Over-the-Air Split Learning Under Channel Aging: Stability, Alignment, and Geometry

This paper investigates the performance of various methods in reciprocity-based over-the-air split learning.

Deep Holes and Aperture: Sign Laws for the Empirical Correlation of Lattice-Sampled Random Wave Fields

This paper studies the properties of Gaussian random wave fields on lattices and their implications for signal processing and communication systems.

On Turn-Level Best-of-N Sampling for Scaling LLM Agents

This paper evaluates a simple inference-time scaling approach for large language model agents on complex tasks.

Highlighted terms show continued research focus across papers

Papers

cs.AIEmpiricalRecentNov 4, 2026

On Turn-Level Best-of-N Sampling for Scaling LLM Agents

Jalaj Bhandari, Dhawal Gupta, Hongli Zhan, Zhenting Wang +4 more

This paper evaluates a simple inference-time scaling approach for large language model agents on complex tasks.

View →
eess.SPcs.ITEmpiricalRecent
Oct 27, 2026

Negative Results for Reciprocity-Based Over-the-Air Split Learning Under Channel Aging: Stability, Alignment, and Geometry

Tianhao Guo, Xianzhong Li, Hao Luo

This paper investigates the performance of various methods in reciprocity-based over-the-air split learning.

View →
cs.ITeess.SPEmpiricalRecentOct 27, 2026

Deep Holes and Aperture: Sign Laws for the Empirical Correlation of Lattice-Sampled Random Wave Fields

Tianhao Guo, Hao Luo, Xianzhong Li, Yuzhi Yang

This paper studies the properties of Gaussian random wave fields on lattices and their implications for signal processing and communication systems.

View →
cs.DCcs.LGEmpiricalRecentJul 25, 2026

Gleam: Adaptive Network-Efficient CUDA API Remoting for Cross-Device GPU Sharing over LANs

Zhihao Xu, Hao Zhong, Zeting Zhou, Yuhang Xu +8 more

This paper proposes Gleam, a framework for efficient GPU sharing across local-area CUDA devices, reducing bandwidth overhead, improving API call latency, and ensuring context consistency.

View →
cs.CLcs.AIcs.CRRecentMay 31, 2026

DiscourseFlip: An Oblique Discourse-Level Opinion Manipulation Attack against Black-box Retrieval-Augmented Generation

Yuyang Gong, Miaokun Chen, Jiawei Liu, Zhuo Chen +4 more

The paper introduces DiscourseFlip, a novel graph-guided attack that demonstrates how coordinated poisoning across a multi-topic query space can manipulate the overall opinion generated by black-box R…

View →
cs.CLcs.AIcs.CRRecentMay 31, 2026

DiscourseFlip: An Oblique Discourse-Level Opinion Manipulation Attack against Black-box Retrieval-Augmented Generation

Yuyang Gong, Miaokun Chen, Jiawei Liu, Zhuo Chen +4 more

The paper introduces DiscourseFlip, a novel black-box, graph-guided attack that manipulates opinions across an entire multi-topic query network, demonstrating a significant leap in scope and effective…

View →
cs.LGcs.AIRecentMay 29, 2026

PR2: Predictive Routing Replay for MoE-Based LLM Reinforcement Learning

Daize Dong, Junlin Chen, Haolong Jia, Jiawei Wu +8 more

The paper proposes Predictive Routing Replay (PR2) to stabilize reinforcement learning on Mixture of Experts (MoE) LLMs by predicting and incorporating short-horizon router evolution during training a…

View →
cs.CLcs.AIRecentMay 29, 2026

D$^3$: Dynamic Directional Graph-Constrained Data Scheduling for LLM Training

Yuanjian Xu, Jianing Hao, Guang Zhang, Zhong Li

The paper proposes $D^3$, a dynamic graph-constrained scheduling framework that optimizes LLM training order by modeling sample interactions as a dynamic influence graph.

View →
cs.CLRecentMay 29, 2026

Towards Efficient LLMs Annealing with Principled Sample Selection

Yuanjian Xu, Jianing Hao, Wanbo Zhang, Zhong Li +1 more

The paper proposes DiReCT, a novel framework that treats data selection during LLM annealing as a constrained optimization problem based on the spectral geometry of the loss landscape, achieving state…

View →
cs.CRcs.AIMar 20, 2026

Trojan's Whisper: Stealthy Manipulation of OpenClaw through Injected Bootstrapped Guidance

Fazhong Liu, Zhuoyan Chen, Tu Lan, Haozhen Tan +5 more

This paper identifies and characterizes 'guidance injection,' a stealthy attack vector that embeds adversarial operational narratives into autonomous coding agents' bootstrap guidance, demonstrating h…

View →