Search results

dataloader_num_workers

Status: Open.

#119 In tulerfeng/Video-R1;

· 2402227288 opened on Jan 9, 2026

可以提供Video-R1-7B-wo-image的模型吗

Status: Open.

#118 In tulerfeng/Video-R1;

· xiaotubaobei opened on Jan 7, 2026

How to change the frame when training？

Status: Open.

#113 In tulerfeng/Video-R1;

· wwxs123W opened on Dec 22, 2025

GRPO Trainer训练过程中NCCL通信问题

Status: Open.

#112 In tulerfeng/Video-R1;

· lyrig opened on Dec 20, 2025

grpo脚本中的max_pixels是什么？

Status: Open.

#111 In tulerfeng/Video-R1;

· Fu-Fu-Fu-Fu opened on Dec 18, 2025

如何调整train batch size

Status: Open.

#109 In tulerfeng/Video-R1;

· Fu-Fu-Fu-Fu opened on Dec 4, 2025

关于测试Prompt的疑惑

Status: Open.

#107 In tulerfeng/Video-R1;

· Fu-Fu-Fu-Fu opened on Dec 1, 2025

是否测试过微调模型随着SFT的step的性能变化？

Status: Open.

#105 In tulerfeng/Video-R1;

· Fu-Fu-Fu-Fu opened on Nov 29, 2025

模型在think里面回答对了，但是answer回答错误

Status: Open.

#104 In tulerfeng/Video-R1;

· Fu-Fu-Fu-Fu opened on Nov 27, 2025

Training Error

Status: Open.

#103 In tulerfeng/Video-R1;

· Lexarymade opened on Nov 3, 2025

T-GRPO training from official SFT checkpoint: loss always 0, reward always 0, all_wrong=1.0

Status: Open.

#101 In tulerfeng/Video-R1;

· Quibbler6 opened on Oct 27, 2025

Error when running RL with provided SFT model: Unrecognized image processor

Status: Open.

#100 In tulerfeng/Video-R1;

· Quibbler6 opened on Oct 25, 2025