7 304 593

xziayro

xziayro

AI & ML interests

None yet

Recent Activity

liked a Space 3 days ago

gen6scp/sana-zero

liked a model 4 days ago

ostris/Flex.1-alpha

reacted to mkurman's post with ❤️ 4 days ago

Blurred-Thoughts Supervised Fine-Tuning (BT-SFT) 🤖 Can we teach a model to think completely on its own without reinforcement learning? Actually, yes. We can do straightforward supervised fine-tuning using a relatively simple trick: blurring a part of CoT thoughts. But why is this effective? We observed that various models differ in their thinking processes, and fine-tuning one model on another model’s thoughts (CoT) can sometimes be inefficient—often resulting in the model simply memorizing reasoning rather than learning how to actually think. I discovered that this process can still be efficient if we clearly indicate when the model should start and stop thinking and uncover only a part of CoT and the expected answer, blurring the other part of CoT. This approach allows the model to learn only a portion of the thought process while still arriving at an expected answer. To demonstrate this, you can watch my experimental BT-SFT on meditsolutions/Llama-3.2-SUN-2.5B-chat model, which was fine-tuned on 151 million tokens from the Magpie-Align/Magpie-Reasoning-V2-250K-CoT-Deepseek-R1-Llama-70B dataset. Enjoy! 🚀 PS. If you were curious enough to read this, leave me a comment. It's always nice to chat with open-minded and intelligent ppl.

View all activity

Organizations

None yet

xziayro's activity

liked a Space 3 days ago

Running on Zero

🖼

Sana-1.6B Zero

Nvidia Sana

liked a model 4 days ago

ostris/Flex.1-alpha

Text-to-Image • Updated 16 days ago • 18.8k • 330

reacted to mkurman's post with ❤️ 4 days ago

Post

1927

Blurred-Thoughts Supervised Fine-Tuning (BT-SFT) 🤖

Can we teach a model to think completely on its own without reinforcement learning? Actually, yes.

We can do straightforward supervised fine-tuning using a relatively simple trick: blurring a part of CoT thoughts. But why is this effective?

We observed that various models differ in their thinking processes, and fine-tuning one model on another model’s thoughts (CoT) can sometimes be inefficient—often resulting in the model simply memorizing reasoning rather than learning how to actually think.

I discovered that this process can still be efficient if we clearly indicate when the model should start and stop thinking and uncover only a part of CoT and the expected answer, blurring the other part of CoT. This approach allows the model to learn only a portion of the thought process while still arriving at an expected answer.

To demonstrate this, you can watch my experimental BT-SFT on meditsolutions/Llama-3.2-SUN-2.5B-chat model, which was fine-tuned on 151 million tokens from the Magpie-Align/Magpie-Reasoning-V2-250K-CoT-Deepseek-R1-Llama-70B dataset.

Enjoy! 🚀

PS. If you were curious enough to read this, leave me a comment. It's always nice to chat with open-minded and intelligent ppl.