Jun Zhao

SEP 8, 2026 · PREPRINT

SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research?

arXiv

This is a benchmarking study of AI agent capabilities in mechanistic interpretability research, establishing measurement methodology but not generating evidence about model safety, alignment, or real-world discovery outcomes.

AUG 12, 2026 · JOURNAL ARTICLE

Peroxisome proliferator-activated receptor gamma (PPARγ) as a mechano-metabolic transducer: coordinating lipid homeostasis through mechanical cues

Molecular Biomedicine

This is a narrative review proposing a mechanistic framework linking mechanical signals to PPARγ-mediated lipid metabolism; it raises conceptual questions rather than reporting empirical evidence from primary trials.