16 papers
Confidence-Adaptive SwiGLU for Mixture-of-Experts
Shaohua Li, Xiuchao Sui, Xiaobing Sun +4
SwiGLU has become a standard gated activation in modern Transformer MLPs, yet its gate sharpness -- the smoothness and selectivity of the gating function -- is typically fixed thro…
Structured Semantic Cloaking for Jailbreak Attacks on Large Language Models
Xiaobing Sun, Perry Lam, Shaohua Li +4
Modern LLMs employ safety mechanisms that extend beyond surface-level input filtering to latent semantic representations and generation-time reasoning, enabling them to recover obf…
Aligning Medical Conversational AI through Online Reinforcement Learning with Information-Theoretic Rewards
Tanvi Verma, Yang Zhou, Rick Siow Mong Goh +1
We present Information Gain Fine-Tuning (IGFT), a novel approach for training medical conversational AI to conduct effective patient interviews and generate comprehensive History o…
Secure and Explainable Fraud Detection in Finance via Hierarchical Multi-source Dataset Distillation
Yiming Qian, Thorsten Neumann, Xueyining Huang +4
We propose an explainable, privacy-preserving dataset distillation framework for collaborative financial fraud detection. A trained random forest is converted into transparent, axi…
From Pretraining to Privacy: Federated Ultrasound Foundation Model with Self-Supervised Learning
Yuncheng Jiang, Chun-Mei Feng, Jinke Ren +15
Ultrasound imaging is widely used in clinical diagnosis due to its non-invasive nature and real-time capabilities. However, traditional ultrasound diagnostics relies heavily on phy…
EVLF-FM: Explainable Vision Language Foundation Model for Medicine
Yang Bai, Haoran Cheng, Yang Zhou +40
Despite the promise of foundation models in medical AI, current systems remain limited - they are modality-specific and lack transparent reasoning processes, hindering clinical ado…