6 papers
Learning from Less: Measuring the Effectiveness of RLVR in Low Data and Compute Regimes
Justin Bauer, Thomas Walshe, Derek Pham +4
Fine-tuning Large Language Models (LLMs) typically relies on large quantities of high-quality annotated data, or questions with well-defined ground truth answers in the case of Rei…
Automating Benchmark Design
Amanda Dsouza, Harit Vishwakarma, Zhengyang Qi +6
The rapid progress and widespread deployment of LLMs and LLM-powered agents has outpaced our ability to evaluate them. Hand-crafted, static benchmarks are the primary tool for asse…
ASAT: Adaptive Scoring and Thresholding with Human Feedback for Robust Out-of-Distribution Detection
Daisuke Yamada, Harit Vishwakarma, Ramya Korlakai Vinayak
Machine Learning (ML) models are trained on in-distribution (ID) data but often encounter out-of-distribution (OOD) inputs during deployment---posing serious risks in safety-critic…
Prune 'n Predict: Optimizing LLM Decision-making with Conformal Prediction
Harit Vishwakarma, Alan Mishler, Thomas Cook +3
Large language models (LLMs) are empowering decision-making in several applications, including tool or API usage and answering multiple-choice questions (MCQs). However, incorrect…
Taming False Positives in Out-of-Distribution Detection with Human Feedback
Harit Vishwakarma, Heguang Lin, Ramya Korlakai Vinayak
Robustness to out-of-distribution (OOD) samples is crucial for safely deploying machine learning models in the open world. Recent works have focused on designing scoring functions…
Pearls from Pebbles: Improved Confidence Functions for Auto-labeling
Harit Vishwakarma, Reid, Chen +4
Auto-labeling is an important family of techniques that produce labeled training sets with minimum manual labeling. A prominent variant, threshold-based auto-labeling (TBAL), works…