5 papers
Can Hallucinations Be Useful? Solving Multi-Hop Questions With SLMs By Chaining System-I/II Reasoning
Saptarshi Sengupta, Suhang Wang
Recently, there has been increased interest in Small Language Models (SLMs), which are fast, show good performance, and have lower hardware demands than large language models (LLMs…
ICCU: In-Context Continual Unlearning via Pattern-Induced Refusal Rules
Ruihao Pan, Suhang Wang
Machine unlearning aims to remove the influence of specific data from trained language models. In real-world deployments, unlearning requests often arrive sequentially, which chall…
LLM Benchmark Datasets Should Be Contamination-Resistant
Ali Al-Lawati, Jason Lucas, Dongwon Lee +1
Benchmark datasets are critical for reproducible, reliable, and discriminative evaluation of LLMs. However, recent studies reveal that many benchmark datasets are included in pretr…
SpecDetect: Simple, Fast, and Training-Free Detection of LLM-Generated Text via Spectral Analysis
Haitong Luo, Weiyao Zhang, Suhang Wang +4
The proliferation of high-quality text from Large Language Models (LLMs) demands reliable and efficient detection methods. While existing training-free approaches show promise, the…
Aligning the Spectrum: Hybrid Graph Pre-training and Prompt Tuning across Homophily and Heterophily
Haitong Luo, Suhang Wang, Weiyao Zhang +3
Graph ``pre-training and prompt-tuning'' aligns downstream tasks with pre-trained objectives to enable efficient knowledge transfer under limited supervision. However, current meth…