RealCompanion: Benchmarking Human Understanding from Reasoning over Longitudinal Real-World Conversations Paper • 2610.01780 • Published 15 days ago • 288
Scaling Properties of Same-Family On-Policy Distillation Paper • 2609.32722 • Published 20 days ago • 327
WISE-ATTA: When to Ask for Labels in Budgeted Active Test-Time Adaptation Paper • 2609.37687 • Published 17 days ago • 7
One Proposal for Every Margin: Zero-Shot Amortized Sequential Importance Sampling for Binary Matrices Paper • 2609.35514 • Published 18 days ago • 5
Understanding On-Policy Distillation: A Mechanistic Interpretability Perspective via Sparse Crosscoders Paper • 2609.35210 • Published 18 days ago • 7