Personalize Your LLM: Fake it then Align It
Published in NAACL Findings, 2025
Authors: Yijing Zhang, Dyah Adila, Changho Shin, Frederic Sala
Publications
2025
Theoretical Physics Benchmark (TPBench) – a Dataset and Study of AI Reasoning Capabilities in Theoretical Physics
Published in Machine Learning: Science and Technology, 2025
Authors: Daniel J.H. Chung, Zhiqi Gao, Yurii Kvasiuk, Tianyi Li, Moritz Münchmeyer, Maja Rudolph, Frederic Sala, Sai Chaitanya Tadepalli
Alignment, Simplified: Steering LLMs with Self-Generated Preferences
Published in Preprint, 2025
Authors: Dyah Adila, Changho Shin, Yijing Zhang, Frederic Sala
Pretrained Hybrids with MAD Skills
Published in Conference on Language Modeling (COLM), 2025
Authors: Nicholas Roberts, Samuel Guo, Zhiqi Gao, Satya Sai Srinath Namburi, Sonia Cromp, Chengjun Wu, Chengyu Duan, Frederic Sala
Weak-to-strong generalization through the data-centric lens
Published in ICLR 2025, 2025
Authors: Changho Shin, John Cooper, Frederic Sala
Test-time Scaling Techniques in Theoretical Physics–A Comparison of Methods on the TPBench Dataset
Published in NeurIPS 2025 Machine Learning and the Physical Sciences (ML4PS) Workshop, 2025
Authors: Zhiqi Gao, Tianyi Li, Yurii Kvasiuk, Sai Chaitanya Tadepalli, Maja Rudolph, Daniel J.H. Chung, Frederic Sala, Moritz Münchmeyer
2024
Stronger Than You Think: Benchmarking Weak Supervision on Realistic Tasks
Published in NeurIPS 2024, 2024
Authors: Tianyi Zhang, Linrong Cai, Jeffrey Li, Nicholas Roberts, Neel Guha, Frederic Sala
The AlCHEmist: Automated Labeling 500x CHEaper than LLM Data Annotators
Published in NeurIPS 2024 (Spotlight), 2024
Authors: Tzu-Heng Huang, Catherine Cao, Vaishnavi Bhargava, Frederic Sala
Pearls from Pebbles: Improved Confidence Functions for Auto-labeling
Published in NeurIPS 2024, 2024
Authors: Harit Vishwakarma, Sui Jiet Tay, Satya Sai Srinath Namburi, Frederic Sala, Ramya Korlakai Vinayak
Zero-Shot Robustification of Zero-Shot Models
Published in ICLR 2024, 2024
Authors: Dyah Adila, Changho Shin, Linrong Cai, Frederic Sala
Look Who’s Talking Now: Covert Channels From Biased LLMs
Published in EMNLP Findings 2024, 2024
Authors: Daniel Silva, Frederic Sala, Ryan Gabrys
OTTER: Effortless Label Distribution Adaptation of Zero-shot Models
Published in NeurIPS 2024, 2024
Authors: Changho Shin, Jitian Zhao, Sonia Cromp, Harit Vishwakarma, Frederic Sala
2023
Ask Me Anything: A simple strategy for prompting language models
Published in ICLR 2023, 2023
Authors: Simran Arora, Avanika Narayan, Mayee F. Chen, Laurel Orr, Neel Guha, Kush Bhatia, Ines Chami, Frederic Sala, Christopher Ré
Multimodal Data Curation Via Object Detection And Filter Ensembles
Published in ICCV Workshop: Towards the Next Generation of Computer Vision Datasets (TNGCV) 2023, 2023
Authors: Tzu-Heng Huang, Changho Shin, Sui Jiet Tay, Dyah Adila, Frederic Sala
Resonant anomaly detection with multiple reference datasets
Published in Journal of High Energy Physics 2023, 2023
Authors: Mayee F. Chen, Benjamin Nachman, Frederic Sala
ScriptoriumWS: A Code Generation Assistant for Weak Supervision
Published in DL4C Workshop at ICLR 2023, 2023
Authors: Tzu-Heng Huang, Catherine Cao, Spencer Schoenberg, Harit Vishwakarma, Nicholas Roberts, Frederic Sala
Skill-it! A Data-Driven Skills Framework for Understanding and Training Language Models
Published in NeurIPS 2023, 2023
Authors: Mayee F. Chen, Nicholas Roberts, Kush Bhatia, Jue Wang, Ce Zhang, Frederic Sala, Christopher Ré
Mitigating Source Bias for Fairer Weak Supervision
Published in NeurIPS 2023, 2023
Authors: Changho Shin, Sonia Cromp, Dyah Adila, Frederic Sala
The Cost of Compression: Investigating the Impact of Compression on Parametric Knowledge in Language Models
Published in EMNLP Findings 2023, 2023
Authors: Srinath Namburi, Makesh Narsimhan Sreedhar, Srinath Srinivasan, Frederic Sala
Domain Generalization via Nuclear Norm Regularization
Published in CPAL 2023, 2023
Authors: Zhenmei Shi, Yifei Ming, Ying Fan, Frederic Sala, Yingyu Liang
Embroid: Unsupervised Prediction Smoothing Can Improve Few-Shot Classification
Published in NeurIPS 2023, 2023
Authors: Neel Guha, Mayee F Chen, Kush Bhatia, Azalia Mirhoseini, Frederic Sala, Christopher Ré
Geometry-Aware Adaptation for Pretrained Models
Published in NeurIPS 2023, 2023
Authors: Nicholas Roberts, Xintong Li, Dyah Adila, Sonia Cromp, Tzu-Heng Huang, Jitian Zhao, Frederic Sala
Product Manifold Representations for Learning on Biological Pathways
Published in ICML Workshop for Computational Biology 2023, 2023
Authors: Daniel McNeela, Frederic Sala, Anthony Gitter
Promises and Pitfalls of Threshold-based Auto-labeling
Published in NeurIPS 2023, 2023
Authors: Harit Vishwakarma, Heguang Lin, Frederic Sala, Ramya Korlakai Vinayak
Train `n Trade: Foundations of Parameter Markets
Published in NeurIPS 2023, 2023
Authors: Tzu-Heng Huang, Harit Vishwakarma, Frederic Sala
Generative Modeling Helps Weak Supervision (and Vice Versa)
Published in ICLR 2023, 2023
Authors: Benedikt Boecking, Nicholas Roberts, Willie Neiswanger, Stefano Ermon, Frederic Sala, Artur Dubrawski
The Credential is Not Enough: Combining Honeypots and Fake Credentials for Cyber-Defense
Published in GameSec-23 2023, 2023
Authors: Sonia Cromp, Mark Bilinski, Ryan Gabrys, Frederic Sala
2022
Universalizing weak supervision
Published in ICLR 2022, 2022
Authors: Changho Shin, Winfred Li, Harit Vishwakarma, Nicholas Roberts, Frederic Sala
AutoWS-Bench-101: Benchmarking Automated Weak Supervision with 100 Labels
Published in NeurIPS 2022, 2022
Authors: Nicholas Roberts*, Xintong Li*, Tzu-Heng Huang, Dyah Adila, Spencer Schoenberg, Cheng-Yu Liu, Lauren Pick, Haotian Ma, Aws Albarghouthi, Frederic Sala
Lifting Weak Supervision To Structured Prediction
Published in NeurIPS 2022, 2022
Authors: Harit Vishwakarma, Nicholas Roberts, Frederic Sala
NAS-Bench-360: Benchmarking Neural Architecture Search on Diverse Tasks
Published in NeurIPS 2022, 2022
Authors: Renbo Tu*, Nicholas Roberts*, Mikhail Khodak, Junhong Shen, Frederic Sala, Ameet Talwalkar