CompGPT: Probing the Compositional Understanding of Visual Language Models
Evan Wang, Felix Heide
Princeton University

CRAB: Computational Reproducibility Agent Benchmark
Zachary Siegel, Sayash Kapoor, Benedikt Stroebl, Nitya Nadgir, Arvind Narayanan
Princeton University

How do Large Language Models Navigate Conflicts between Honesty and Helpfulness?
Ryan Liu1*, Theodore R. Sumers1*, Ishita Dasgupta2, Thomas L. Griffiths1
1Princeton University, 2Google Deepmind (* equal contribution)

Privacy-Preserving In-Context Learning with Differentially Private Few-Shot Generation
Xinyu Tang1 Richard Shin2 Huseyin A. Inan3 Andre Manoel3 Niloofar Mireshghallah4 Zinan Lin5 Sivakanth Gopi5 Janardhan Kulkarni5 Robert Sim3
1Princeton University 2Microsoft Semantic Machines 3M365 Research 4University of Washington 5Microsoft Research

QuRating: Selecting High-Quality Data for Training Language Models
Alexander Wettig, Aatmik Gupta, Saumya Malik, Danqi Chen
Princeton University

Replacing AI Agent Leaderboards with Pareto Curve Evaluations
Sayash Kapoor, Benedikt Stroebl, Zachary Siegel, Arvind Narayanan
Princeton University

Trainable Transformer in Transformer
Abhishek Panigrahi* Sadhika Malladi* Mengzhou Xia Sanjeev Arora
Princeton University * Equal Contribution