LexReward: A Taxonomy-Driven Reward Framework for Legal Language Models Paper • 2609.39071 • Published 13 days ago • 66
Towards Looped Models Done Right, Part II: Rethinking at Fixed Points Paper • 2610.06833 • Published 8 days ago • 34
SlimWise: Decoupling Expert Pruning Across Prefill and Decode for Efficient MoE Serving Paper • 2609.34117 • Published 15 days ago • 25
InfoBayAI/CT-Scan-Radiology-Reports-With-Findings-Dataset Viewer • Updated 4 days ago • 6.3k • 39 • 10
Triadic Linear Attention: Three-Dimensional Recurrent States for Long-Context Sequence Modeling Paper • 2609.36529 • Published 14 days ago • 37
InfoBayAI/CT-Scan-Radiology-Reports-Without-Findings-Dataset Viewer • Updated 4 days ago • 2.6k • 41 • 8
StanfordAIMI/stanford-deidentifier-with-radiology-reports-and-i2b2 Token Classification • Updated Nov 23, 2022 • 3.88k • • 14
More Choices, Fewer Decisions: Ordinal-Scale Bias in JEV-like Direct-Decision Models Paper • 2609.38827 • Published 13 days ago • 64
YuE2: Unifying Symbolic and Audio Music Generation at Frontier Quality Paper • 2609.33757 • Published 16 days ago • 234
When EOS Tokens Disagree: Understanding Length Inflation in On-Policy Distillation Paper • 2609.20511 • Published 26 days ago • 115