LLaVA-OneVision-2.5 shows codec input beats uniform frame sampling on video tasks
liuziwei7 · x · 2026-11-19
LMMS Lab released LLaVA-OneVision-2.5, the first clear evidence that codec (compressed bitstream) input beats uniform frame sampling on general video understanding. Pixeltable adds a practical caveat: codec mode selects patches by encoder bit-cost (dense I-frames, motion-rich P-frame patches), a signal lost if you decode to JPEG and discard the source; frame sampling still wins some benchmarks (TOMATO 39.2 vs 36.0, LVBench 62.0 vs 61.6), so keep source video and treat frames as recomputable.
More from Models
- 17 LLMs draw the same pelican on a bicycle at max reasoning effort — nadermx · 2026-11-19
- Gemini 4: Neon (low) matches or beats Hydrogen (medium) at lower cost — fundal_alb · 2026-11-19
- White House, Treasury weigh requiring US firms to disclose use of Chinese open-weight AI models — kevinnbass · 2026-11-19
- Google report details autonomous AI hacking system, but no confirmed breaches — TruthIsAllYouNeed_ · 2026-11-19
- Rare sight: broad, cross-constituency public praise for Gemini models — matt_slotnick · 2026-11-19
- Seedream 5.0 Pro accused of charging credits while showing 'Completed' with no output — JohnTheWorldfucker · 2026-11-19