StylisticBias: A Few Human Visual Cues Drive Most Social Biases in MLLMs Paper • 2606.20527 • Published 27 days ago • 3
Functional Attention: From Pairwise Affinities to Functional Correspondences Paper • 2605.31559 • Published May 29 • 2
BA-T: An Iterative Transformer for Two-View Bundle Adjustment Paper • 2606.03287 • Published Jun 2 • 2
Active Tabular Augmentation via Policy-Guided Diffusion Inpainting Paper • 2605.10315 • Published May 11
RealICU: Do LLM Agents Understand Long-Context ICU Data? A Benchmark Beyond Behavior Imitation Paper • 2605.13542 • Published May 13 • 8
Trees to Flows and Back: Unifying Decision Trees and Diffusion Models Paper • 2605.00414 • Published May 1 • 7
How Much Is One Recurrence Worth? Iso-Depth Scaling Laws for Looped Language Models Paper • 2604.21106 • Published Apr 27 • 10
KV Packet: Recomputation-Free Context-Independent KV Caching for LLMs Paper • 2604.13226 • Published Apr 14 • 11
FINER: MLLMs Hallucinate under Fine-grained Negative Queries Paper • 2603.17662 • Published Mar 18 • 3
sebis at ArchEHR-QA 2026: How Much Can You Do Locally? Evaluating Grounded EHR QA on a Single Notebook Paper • 2603.13962 • Published Mar 14