Search results

Shift Equivariant Losses

Status: Open.

#181 In NVlabs/RADIO;

· lthilnklover opened on May 7, 2026

Inquiry: Future support for ConvNeXt-based backbones or Multi-Scale feature outputs?

Status: Open.

#179 In NVlabs/RADIO;

· wangzhenlin123 opened on Mar 18, 2026

How to further distill the RadioV4 model？

Status: Open.

#178 In NVlabs/RADIO;

· Pixel16Bit opened on Mar 6, 2026

load model error

Status: Open.

#177 In NVlabs/RADIO;

· cxmscb opened on Mar 3, 2026

Question about model license and distillation

Status: Open.

#176 In NVlabs/RADIO;

· tcourat opened on Feb 16, 2026

Questions regarding SAM3 backbone replacement and training requirements for C-RADIOv4

Status: Open.

#175 In NVlabs/RADIO;

· JuiceCoffe opened on Feb 13, 2026

[C-RADIOv4] Strong outlier artifacts in spatial features (PCA visualization)

Status: Open.

#174 In NVlabs/RADIO;

· Lwt-diamond opened on Feb 11, 2026

Convergence issues when using C-Radio (v2/v3/v4) as Vision Encoder in LLaVA-style pretraining

Status: Open.

#167 In NVlabs/RADIO;

· haoyi199815 opened on Feb 4, 2026

[Bug] KeyError: "Couldn't find the key blocks.0.ls1.gamma nor blocks.0.ls1.grandma" when loading C-RADIOv3-L

Status: Open.

#164 In NVlabs/RADIO;

· wangjl-nb opened on Jan 22, 2026

Inquiry regarding the specific SigLIP 2 checkpoint used for c-radio_v3 training

Status: Open.

#163 In NVlabs/RADIO;

· wangjl-nb opened on Jan 13, 2026

Confusion on checkpoint on torch hub v.s. huggingface

Status: Open.

#157 In NVlabs/RADIO;

· jrryzh opened on Nov 19, 2025

Upscale embeddings for the RADIO Model

Status: Open.

#156 In NVlabs/RADIO;

· ivanlyubimtsev opened on Aug 30, 2025