ByteBrief
We're a portrait publication through and through. Turn your phone back and your briefing picks up right where you left it.
(We tried widescreen once. It wasn't us.)

Sun Yat-sen University researchers introduced GeoThinker-F, a method that drops the 3D encoder to cut spatial reasoning latency from 1.45 seconds to 0.40 seconds. This change makes the model 3.6 times faster while maintaining scores near the original GeoThinker on benchmarks like VSI-Bench, with results posted to arXiv on October 14.
Tap to vote and see what everyone thinks.
Summary by ByteBrief