On-policy distillation (OPD) has recently emerged as a promising approach for transferring the capabilities of large agents to compact students. In deep search, where agents iteratively retrieve external evidence, a teacher's preference among plausible queries before retrieval may not predict which queries will return...
Chen-Xi Li, Xue-Yang Feng, Song-Yang Liu et al.· 0 citations
Novels tell rich stories, but their textual form cannot directly present the sights and sounds of the worlds they describe. Film adaptations offer an audiovisual experience but require substantial production effort and viewing time; film recaps are shorter but depend on existing footage. Neither route directly provides...
Kyrie You Wu, Chao-Zhuo Li, Song-Yang Liu et al.· 0 citations
This work is the first to formalize Role Grounding Attack (RGA), an attack model that exploits a vulnerability in role-to-principal attribution through nested conversation structures and achieves high attack success rates.
Hao-Ran Bu, Yi-Da Shen, Li-Tian Zhang et al.· 0 citations
This work organizes the literature on world-model safety around a six-stage lifecycle covering data and observations, state and memory, dynamics and imagination, evaluation and planning, action and control, and feedback and update, and uses the framework to identify where claims are supported, where attribution remains...
Xiao-Mao Wang, Li-Tian Zhang, Xin-Ze Shi et al.· 0 citations
LLM-driven agents can autonomously exchange opinions on online platforms and form communities. Such agent-operated social platforms raise a new security concern: attackers may manipulate agents to induce group polarization. Existing methods manipulate agent prompts or construct echo chambers, both of which are difficul...
Haoran Bu, Zejian Chen, Li-Tian Zhang et al.· 0 citations
A holistic framework based on learnable information gain, which measures how much novel, parameterizable information a round provides relative to the previous round, and proposes ATRI (Adaptive Training Regulation via Information-gain), which reweights samples within a round and halts training across rounds when inform...
Chen-Xu Wang, Chao-Zhuo Li, Xin-Ze Shi et al.· arXiv.org· 0 citations
EvoBreak is an experience-conditioned sequential attack that operates through individually benign attack-stage tasks and induced experiences, revealing benign experience composition as a persistent attack surface in self-evolving agents.
Bing-Yu Yan, Xiao-Ming Zhang, Chao-Zhuo Li et al.· 2 citations
MAGER is a multi-agent genetic evolution framework that automatically discovers meta-paths optimized for LLM reasoning, enabling frozen LLMs to perform structure-aware veracity reasoning and introduces a graph in-context learning strategy that retrieves semantically and structurally similar demonstrations to strengthen...
Zi-Yi Zhou, Xiao-Ming Zhang, Hui Pang et al.· 0 citations
PPGNN, a personalized differentially private framework for decentralized graph data, enables user-specific privacy budgets during local perturbation while preserving analytical utility in decentralized graph learning scenarios.
Longzhu He, Peng Tang, Chaozhuo Li et al.· IEEE Transactions on Knowled...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.