Hang Zhang

SEP 3, 2026 · PREPRINT

Gradients Know What Outcomes Don't: Unlocking Reinforcement Learning for LLM Reasoning with Gradient-Aligned Rewards

arXiv

This is a preprint describing a novel method for reward design in reinforcement learning applied to LLM reasoning, demonstrated on benchmark tasks without peer review or independent external validation.

AUG 17, 2026 · PREPRINT

Learning to Unlearn: Machine Unlearning via Learning the Unlearning Behaviors

arXiv

A novel machine learning method for data unlearning is presented with experimental validation, but lacks peer review, clinical context, and independent replication in a preprint format.

JUL 14, 2026 · JOURNAL ARTICLE

Disruption of rcnB modulates colistin susceptibility in Acinetobacter baumannii AB5075.

Virulence

Mechanistic in vitro study using CRISPR-Cas9 gene editing to explore rcnB function in a single bacterial strain; raises questions about colistin resistance mechanisms rather than establishing clinical evidence.