Marcus Rohrbach
2 papers · Latest:
Computer Vision
SIEVES: Selective Prediction Generalizes through Visual Evidence Scoring
SIEVES improves selective prediction in MLLMs by scoring visual evidence localization, tripling coverage on OOD tasks without model-specific training.
2604.25855
Computer VisionReCap: Lightweight Referential Grounding for Coherent Story Visualization
ReCap is a lightweight framework improving character consistency in story visualization by selectively conditioning on previous frames, outperforming SOTA.
2604.18575
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.