Vision-Language-Models on skulldchttps://skulldc.net/tags/vision-language-models/Recent content in Vision-Language-Models on skulldcHugoen-usSat, 07 Nov 2026 10:25:32 +0000VFlash predicts whole visual plan primitives in one pass instead of token by tokenhttps://skulldc.net/2026/11/vflash-predicts-whole-visual-plan-primitives-in-one-pass-instead-of-to/Sat, 07 Nov 2026 10:25:32 +0000https://skulldc.net/2026/11/vflash-predicts-whole-visual-plan-primitives-in-one-pass-instead-of-to/An NVIDIA paper replaces coordinate-by-coordinate decoding of points, boxes and trajectories with a drafter that emits each primitive at once, reporting fewer errors and faster generation.