Xiao Zhang

5 papers University of Chicago

• FACT: Mitigating Inconsistent Hallucinations in LLMs via Fact-Driven Alternating Code-Text Training
• GASP: Efficient Black-Box Generation of Adversarial Suffixes for Jailbreaking LLMs
• Hierarchical Implicit Neural Emulators
• Investigating and Mitigating Catastrophic Forgetting in Medical Knowledge Injection through Internal Knowledge Augmentation Learning
• PROFIT: A Specialized Optimizer for Deep Fine Tuning