7 papers
WildTrace: Evaluating Training Data Detection in Realistic Large-Scale LLM Post-Training
Hengxiang Zhang, Junxian Liu, Hongxin Wei
Although post-training significantly enhances model performance, the use of massive undisclosed data raises concerns about benchmark contamination and copyright infringement. This…
Understanding Evaluation Illusion in Diffusion Large Language Models
Hengxiang Zhang, Jiaxi Ren, Renchunzi Xie +1
Despite the capability of parallel decoding, diffusion large language models (dLLMs) require many denoising steps to maintain generation quality, motivating recent research on effi…
Detecting Distillation Data from Reasoning Models
Hengxiang Zhang, Hyeong Kyu Choi, Sharon Li +1
Reasoning distillation has emerged as a prevailing paradigm for transferring reasoning capabilities from large reasoning models to small language models. Yet, reasoning distillatio…
ChineseSafe: A Chinese Benchmark for Evaluating Safety in Large Language Models
Hengxiang Zhang, Hongfu Gao, Qiang Hu +7
With the rapid development of Large language models (LLMs), understanding the capabilities of LLMs in identifying unsafe content has become increasingly important. While previous w…
Fine-tuning can Help Detect Pretraining Data from Large Language Models
Hengxiang Zhang, Songxin Zhang, Bingyi Jing +1
In the era of large language models (LLMs), detecting pretraining data has been increasingly important due to concerns about fair evaluation and ethical risks. Current methods diff…
Defending Membership Inference Attacks via Privacy-aware Sparsity Tuning
Hengxiang Zhang, Qiang Hu, Hongxin Wei
Over-parameterized models are typically vulnerable to membership inference attacks, which aim to determine whether a specific sample is included in the training of a given model. P…
Automatic Dataset Construction (ADC): Sample Collection, Data Curation, and Beyond
Minghao Liu, Zonglin Di, Jiaheng Wei +15
Large-scale data collection is essential for developing personalized training data, mitigating the shortage of training data, and fine-tuning specialized models. However, creating…