6 papers
DEER: Disentangled Mixture of Experts with Instance-Adaptive Routing for Generalizable Machine-Generated Text Detection
Guoxin Ma, Xiaoming Liu, Hongyang Chen +6
Detecting machine-generated text has become a critical challenge amid the rapid advancement of LLMs, yet existing detectors degrade severely under domain shift. Through systematic…
Can Reasoning Path still be Effective as Input? Bridging Post-Reasoning to Chain-of-Thought Compression
Chengzhengxu Li, Xiaoming Liu, Zhaohan Zhang +5
Recent developments have enabled advanced reasoning in Large Language Models (LLMs) via long Chain-of-Thought (CoT), trading efficiency during inference for performance. Existing w…
Auditing Agent Harness Safety
Chengzhi Liu, Yichen Guo, Yepeng Liu +8
LLM agents increasingly run inside execution harnesses that dispatch tools, allocate resources, and route messages between specialized components. However, a harness can return a c…
On Distinguishing Capability Elicitation from Capability Creation in Post-Training: A Free-Energy Perspective
Yuhao Li, Shengchao Liu
Debates about large language model post-training often treat supervised fine-tuning (SFT) as imitation and reinforcement learning (RL) as discovery. But this distinction is too coa…
A Minimal Model of Representation Collapse: Frustration, Stop-Gradient, and Dynamics
Louie Hong Yao, Yuhao Li, Shengchao Liu
Self-supervised representation learning is central to modern machine learning because it extracts structured latent features from unlabeled data and enables robust transfer across…
MGT-Prism: Enhancing Domain Generalization for Machine-Generated Text Detection via Spectral Alignment
Shengchao Liu, Xiaoming Liu, Chengzhengxu Li +4
Large Language Models have shown growing ability to generate fluent and coherent texts that are highly similar to the writing style of humans. Current detectors for Machine-Generat…