PyMETA: Evaluating Student Code Diagnosis on and Beyond the First Execution Error Paper • 2606.30610 • Published 29 days ago
WhoSaidIt: Human-LLM Collaborative Annotation for Text-Based Multilingual Speaker-Attribute Classification Paper • 2605.26070 • Published May 25
"What makes a question inquisitive?" A Study on Type-Controlled Inquisitive Question Generation Paper • 2205.08056 • Published May 19, 2022
Set-Based Prompting: Provably Solving the Language Model Order Dependency Problem Paper • 2406.06581 • Published Jun 4, 2024 • 1
Aligning Superhuman AI with Human Behavior: Chess as a Model System Paper • 2006.01855 • Published Jun 2, 2020
The Benefits of Label-Description Training for Zero-Shot Text Classification Paper • 2305.02239 • Published May 3, 2023
ToMChallenges: A Principle-Guided Dataset and Diverse Evaluation Tasks for Exploring Theory of Mind Paper • 2305.15068 • Published May 24, 2023
Ambiguity-Aware In-Context Learning with Large Language Models Paper • 2309.07900 • Published Sep 14, 2023 • 5
Training Models to Generate, Recognize, and Reframe Unhelpful Thoughts Paper • 2307.02768 • Published Jul 6, 2023 • 15
Improving Open Language Models by Learning from Organic Interactions Paper • 2306.04707 • Published Jun 7, 2023 • 3