5 papers
CauTion: Knowing When to Trust LLMs for Ensemble Causal Discovery
Bo Peng, Kaiwen Wu, Sirui Chen +3
Causal discovery from observational data remains challenging due to the fundamental limitations of purely statistical methods, such as statistical distinguishability within equival…
Agent Tools Orchestration Leaks More: Dataset, Benchmark, and Mitigation
Yuxuan Qiao, Dongqin Liu, Hongchang Yang +2
LLM-based agents increasingly use multiple external tools to complete complex tasks. We study Tools Orchestration Privacy Risk (TOP-R): an agent may combine individually non-sensit…
LLMs know their vulnerabilities: Uncover Safety Gaps through Natural Distribution Shifts
Qibing Ren, Hao Li, Dongrui Liu +7
Safety concerns in large language models (LLMs) have gained significant attention due to their exposure to potentially harmful data during pre-training. In this paper, we identify…
Data Darwinism Part II: DataEvolve -- AI can Autonomously Evolve Pretraining Data Curation
Tiantian Mi, Dongming Shan, Zhen Huang +6
Data Darwinism (Part I) established a ten-level hierarchy for data processing, showing that stronger processing can unlock greater data value. However, that work relied on manually…
OASIS: Open Agent Social Interaction Simulations with One Million Agents
Ziyi Yang, Zaibin Zhang, Zirui Zheng +20
There has been a growing interest in enhancing rule-based agent-based models (ABMs) for social media platforms (i.e., X, Reddit) with more realistic large language model (LLM) agen…