Search results

ui-r1-3b-e training dataset

Status: Open.

#18 In lll6gg/UI-R1;

· lezhang7 opened on Jul 15, 2025

未能复现论文精度

Status: Open.

#17 In lll6gg/UI-R1;

· ioir123ju opened on Jul 10, 2025

推理速度问题

Status: Open.

#16 In lll6gg/UI-R1;

· CTZhou-byte opened on Jun 4, 2025

About the reproduction of experimental results.

Status: Open.

#15 In lll6gg/UI-R1;

· caseclose opened on May 26, 2025

Great job on this project!

Status: Open.

#14 In lll6gg/UI-R1;

· GaoMengGladys opened on May 24, 2025

compute_loss 中的self.num_iterations 不更新吗？如果不更新，计算的per_token_kl横为0，kl惩罚无效

Status: Open.

#12 In lll6gg/UI-R1;

· moguizhizi opened on May 21, 2025

Training requires upgrading transformers from 4.49.0 due to dtype mismatch error, causing potential performance concerns

Status: Open.

#9 In lll6gg/UI-R1;

· YcZhangSing opened on May 8, 2025

How to support multi round chat training?

Status: Open.

#8 In lll6gg/UI-R1;

· xuexidi opened on May 6, 2025

How to eval grounding metric in androidcontrol?

Status: Open.

#7 In lll6gg/UI-R1;

· Quester-one opened on May 3, 2025

启动lora训练

Status: Open.

#6 In lll6gg/UI-R1;

· LucusFigoGao opened on Apr 22, 2025

How do you process AndroidControl Dataset?

Status: Open.

#5 In lll6gg/UI-R1;

· mengzchen opened on Apr 22, 2025

请教一个代码的问题

Status: Open.

#4 In lll6gg/UI-R1;

· cbigeyes opened on Apr 15, 2025