tianzhechu/SFTvsRL_Data

Dataset

4

stars

5

commits

2

linked in READMEs

Feb 8, 2025

updated

README

SFTvsRL_Data

This dataset contains necessary data to reproduce results in paper SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training.

Dataset structures

SFTvsRL_DATA/
  - VIRL_Routes/
    - VLN_mini.zip # containing 18 routes in VLN mini benchmark, for eval
    - nyc_1k_routes.zip # containing 1k routes in NYC, for training
  - SFT_data/ # we are still preparing
    - gp-l/
      - data.json
    - virl-l/
      - data.json

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

Contributors

tianzhechu

5 commits

tianzhechu/SFTvsRL_Data

Dataset

4

stars

5

commits

2

linked in READMEs

Feb 8, 2025

updated

README

SFTvsRL_Data

This dataset contains necessary data to reproduce results in paper SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training.

Dataset structures

SFTvsRL_DATA/
  - VIRL_Routes/
    - VLN_mini.zip # containing 18 routes in VLN mini benchmark, for eval
    - nyc_1k_routes.zip # containing 1k routes in NYC, for training
  - SFT_data/ # we are still preparing
    - gp-l/
      - data.json
    - virl-l/
      - data.json

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

Contributors

tianzhechu

5 commits