Pengxiang Li

3
Papers
71
Total Citations

Papers (3)

Multi-modal Agent Tuning: Building a VLM-Driven Agent for Efficient Tool Usage

ICLR 2025
37
citations

Mix-LN: Unleashing the Power of Deeper Layers by Combining Pre-LN and Post-LN

ICLR 2025
22
citations

Adaptive Classifier-Free Guidance via Dynamic Low-Confidence Masking

NeurIPS 2025arXiv
12
citations