tianzhechu
/

VIRL-VL-Init

Model card Files Files and versions Community

VIRL-VL-Init

This model serves as a initial checkpoint to reproduce results in paper SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training.

Related links

Website: https://tianzhechu.com/SFTvsRL/

Github: https://github.com/LeslieTrue/SFTvsRL

Arxiv: https://arxiv.org/abs/2501.17161v1

HF: https://huggingface.co./papers/2501.17161

Downloads last month: 33

Safetensors

Model size

10.7B params

Tensor type

BF16

·

Inference Providers NEW

This model is not currently available via any of the supported Inference Providers.

The model cannot be deployed to the HF Inference API: The model has no library tag.

Collection including tianzhechu/VIRL-VL-Init

SFTvsRL Models & Data

This collection contains 4 initial checkpoints for https://github.com/LeslieTrue/SFTvsRL and necessary data for V-IRL training. • 5 items • Updated 9 days ago • 7