lilakk/BLEUBERI
1
0
Fork 0
mirror of https://github.com/lilakk/BLEUBERI.git synced 2026-04-19 12:58:12 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

minor update

Browse source
This commit is contained in:
yapei 2025-06-05 01:05:43 +00:00
parent 84309f0f78
commit e0d1563925
2 changed files with 1 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

4
README.md
View file

@ -39,9 +39,7 @@ Check [training](training) for details on creating training data and running GRP
### Dataset and model releases
For our 50K data pool obtained from the [Tulu3 SFT mixture](https://huggingface.co/datasets/allenai/tulu-3-sft-mixture), we collected synthetic responses from the following LLMs: `claude-3-7-sonnet@20250219`, `deepseek-chat-v3`, `gemini-2.5-pro-exp-03-25`, `o4-mini-2025-04-16`, `Llama-3.1-8B-Instruct`. The resulting dataset, [BLEUBERI-Tulu3-50k](https://huggingface.co/datasets/yapeichang/BLEUBERI-Tulu3-50k), is publicly available on Hugging Face.
Trained models reported in our main experiment table (Table 1) can be found [here](https://huggingface.co/collections/yapeichang/bleuberi-6840b3b9d02ff86c5878dafa).
For our 50K data pool obtained from the [Tulu3 SFT mixture](https://huggingface.co/datasets/allenai/tulu-3-sft-mixture), we collected synthetic responses from the following LLMs: `claude-3-7-sonnet@20250219`, `deepseek-chat-v3`, `gemini-2.5-pro-exp-03-25`, `o4-mini-2025-04-16`, `Llama-3.1-8B-Instruct`. **The resulting dataset, [BLEUBERI-Tulu3-50k](https://huggingface.co/datasets/yapeichang/BLEUBERI-Tulu3-50k), is publicly available. Trained models reported in our main experiment table can be found [here](https://huggingface.co/collections/yapeichang/bleuberi-6840b3b9d02ff86c5878dafa).**
## BLEUBERI-trained models generate more factually grounded outputs

BIN
assets/data_sources.png Normal file
View file

Binary file not shown.
Side by side

After

Width:  |  Height:  |  Size: 195 KiB