Audits Under Resource, Data, and Access Constraints: Scaling Laws For Less Discriminatory Alternatives

This is experimental HTML to improve accessibility. We invite you to report rendering errors. Use Alt+Y to toggle on accessible reporting links and Alt+Shift+Y to toggle off. Learn more about this project and help improve conversions.

Audits Under Resource, Data, and Access Constraints: Scaling Laws For Less Discriminatory Alternatives Sarah H. Cen Stanford University Palo Alto, CA 94304 shcen@stanford.edu &Salil Goyal Stanford University Palo Alto, CA 94304 salilg@stanford.edu &Zaynah Javed Stanford University Palo Alto, CA 94304 zjaved@stanford.edu Ananya Karthik Stanford University Palo Alto, CA 94304 ananya23@stanford.edu &Percy Liang Stanford University Palo Alto, CA 94304 pliang@cs.stanford.edu &Daniel E. Ho Stanford University Palo Alto, CA 94304 deho@stanford.edu Abstract AI audits play a critical role in AI account

related reading

• Statistical Fairness - Turing Commonsalan-turing-institute.github.io
• Papers · Nikhil Garggargnikhil.com
• IsoCompute Playbook: Optimally Scaling Sampling Compute for RL Training of LLMscompute-optimal-rl-llm-scaling.github.io
• MAI-Thinking-1: Building a Hill-Climbing Machinemicrosoft.ai
• DataRater: Meta-Learned Dataset Curationarxiv.org
• Classificationfairmlbook.org
• VaultGemma: The world's most capable differentially private LLMresearch.google
• arxiv.org/pdf/2511.08544arxiv.org
• A Bitter Lesson for Data Filteringarxiv.org
• [2411.12925] Loss-to-Loss Prediction: Scaling Laws for All Datasetsarxiv.org
• Frontiers | On Consequentialism and Fairnessfrontiersin.org
• MAI-Thinking-1: Building a Hill-Climbing Machinemicrosoft.ai