RobuQ delivers the first stable DiT image generation at W1.58A2 average bits via Hadamard-based robust activation quantization and layer-wise mixed-precision activations.
Imagenet large scale visual recognition challenge
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2025 2verdicts
CONDITIONAL 2representative citing papers
Franca introduces nested Matryoshka clustering and positional disentanglement in a transparent SSL pipeline to deliver open-source vision models competitive with closed proprietary systems.
citing papers explorer
-
RobuQ: Pushing DiTs to W1.58A2 via Robust Activation Quantization
RobuQ delivers the first stable DiT image generation at W1.58A2 average bits via Hadamard-based robust activation quantization and layer-wise mixed-precision activations.
-
Franca: Nested Matryoshka Clustering for Scalable Visual Representation Learning
Franca introduces nested Matryoshka clustering and positional disentanglement in a transparent SSL pipeline to deliver open-source vision models competitive with closed proprietary systems.