7 papers
Relational Response Fields: A General Theory of Black-Box LLM Response Consistency and Recovery
Song Zichen
Black-box language-model reliability is commonly pursued by sampling, prompting, voting, verifying, or iteratively revising individual answers. We ask a prior question: \emph{what…
PPCR-IM: A System for Multi-layer DAG-based Public Policy Consequence Reasoning and Social Indicator Mapping
Zichen Song, Weijia Li
Public policy decisions are typically justified using a narrow set of headline indicators, leaving many downstream social impacts unstructured and difficult to compare across polic…
FTS: A Framework to Find a Faithful TimeSieve
Songning Lai, Ninghui Feng, Haochen Sui +5
The field of time series forecasting has garnered significant attention in recent years, prompting the development of advanced models like TimeSieve, which demonstrates impressive…
EM-MIAs: Enhancing Membership Inference Attacks in Large Language Models through Ensemble Modeling
Zichen Song, Sitan Huang, Zhongfeng Kang
With the widespread application of large language models (LLM), concerns about the privacy leakage of model training data have increasingly become a focus. Membership Inference Att…
Layer Importance and Hallucination Analysis in Large Language Models via Enhanced Activation Variance-Sparsity
Zichen Song, Sitan Huang, Yuxin Wu +1
Evaluating the importance of different layers in large language models (LLMs) is crucial for optimizing model performance and interpretability. This paper first explores layer impo…
AVSS: Layer Importance Evaluation in Large Language Models via Activation Variance-Sparsity Analysis
Zichen Song, Yuxin Wu, Sitan Huang +1
The evaluation of layer importance in deep learning has been an active area of research, with significant implications for model optimization and interpretability. Recently, large…