Bo Ji

The paper investigates how various fine-tuning methods can be used both to intentionally misalign and subsequently realign large language models (LLMs), revealing distinct strengths for attack and defense mechanisms.

The paper introduces SecGoal, a benchmark dataset and framework, demonstrating that fine-tuning smaller LLMs on this dataset significantly improves the precision of extracting formalizable security goals from natural language protocol documents.

The paper introduces EvaluatAR, a cross-device evaluation framework that standardizes the testing of bystander Privacy-Enhancing Technologies (PETs) in Augmented Reality (AR) to enable rapid, reproducible prototyping across different hardware.

The paper proposes a novel temporal and structural credit assignment framework to efficiently optimize multi-agent LLM systems by decomposing the error signal and using targeted, discrete gradient updates.

The paper introduces AgentSchool, an advanced LLM-powered multi-agent simulator that models learning as state transitions to provide a robust, ethically viable testbed for educational research and pedagogical reform.

The paper introduces Singularity-aware Adam (S-Adam), a novel optimizer that stabilizes deep learning training in non-smooth loss landscapes by dynamically damping updates based on local geometric instability.