TRIAGE: Direction-Aware Mismatch Stabilization of Native NVFP4 Reinforcement Learning
TRIAGE stabilizes native NVFP4 RL on Qwen3 models, matching full precision with up to 2.3x rollout throughput.
3 stories, newest first · one card per story
TRIAGE stabilizes native NVFP4 RL on Qwen3 models, matching full precision with up to 2.3x rollout throughput.
K-interval quantized softmax nearly matches full-softmax pretraining loss when calibration and the gradient surrogate are chosen carefully.
A research paper on low-precision EEG decoders examines adversarial robustness across compression techniques.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views