ModelVerificadoExterno

ReflectRL Qwen2.5 Math 7B DAPO

This model checkpoint is part of the work presented in ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning.

Fontehuggingface.co/Jinhe/ReflectRL-Qwen2.5-Math-7B-DAPO/tree/bb6557fc35419474bcf887ab5e2c81abc8ecf38a

TipoText modelModelo baseQwen/Qwen2.5-Math-7BVersãomodelPublicador@JinheCnota

Fonte do modelo

Tipo: Text model
Tamanho do arquivo: 28,4 GB
Modelo base: Qwen/Qwen2.5-Math-7B
Versão: vmodel
Atualizado: 4 de out. de 2026
Publicador: @Jinhe
Downloads: 17
Parâmetros: 7B
Fonte: Hugging Face

Descrição da fonte

ReflectRL-Qwen2.5-Math-7B-GRPO-ReflectRL

This model checkpoint is part of the work presented in ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning.

ReflectRL is a lightweight framework for learning from Golden Negative Trajectories (GNTs) during on-policy post-training. Instead of imitating failed expert trajectories directly, ReflectRL uses them as reflective context during training and gradually transitions the policy back to direct reasoning for inference.

Paper: ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning
Repository: GitHub Repository

Citation

@article{reflectrl2027,
  title={ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning},
  author={Jinhe Bi and Chennan Zhou and Zengjie Jin and Aniri and Shuo Lu and Wenke Huang and Hu Cao and Xun Xiao and Zhihong Zhu and Volker Tresp and Fei Shen and Yunpu Ma and Tat-Seng Chua},
  year={2026}
}

Fontes

1 fonte

Hugging FacePrimárioVersão bb6557fc35419474bcf887ab5e2c81abc8ecf38a

huggingface.co/Jinhe/ReflectRL-Qwen2.5-Math-7B-DAPO/tree/bb6557fc35419474bcf887ab5e2c81abc8ecf38a

Verificado 4 de out.

Artefatos de modelo

1 artefato

model.safetensors

safetensors · 28,4 GB · SHA-256 926f8e2066dd…1ea5 · Hugging Face

Baixar

Trechos de fonte

3 trechos

Readme4 de out.huggingface.co/Jinhe/ReflectRL-Qwen2.5-Math-7B-DAPO

--- license: apache-2.0 library_name: transformers pipeline_tag: text-generation base_model: Qwen/Qwen2.5-Math-7B --- # ReflectRL-Qwen2.5-Math-7B-GRPO-ReflectRL This model checkpoint is part of the work presented in [ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning](https://huggingface.co/papers/2608.03972). ReflectRL is a lightweight framework for learning from Golden Negative Trajectories (GNTs) during on-policy post-training. Instead of imitating failed expert trajectories directly, ReflectRL uses them as reflective context during training and gradually transitions the policy back to direct reasoning for inference. - **Paper:** [ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning](https://huggingface.co/papers/2608.03972) - **Repository:** [GitHub Repository](https://github.com/bibisbar/ReflectRL) ## Citation ```bibtex @article{reflectrl2027, title={ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning}, author={Jinhe Bi and Chennan Zhou and Zengjie Jin and Aniri and Shuo Lu and Wenke Huang and Hu Cao and Xun Xiao and Zhihong Zhu and Volker Tresp and Fei Shen and Yunpu Ma and Tat-Seng Chua}, year={2026} } ```

Note4 de out.huggingface.co/Jinhe/ReflectRL-Qwen2.5-Math-7B-DAPO

Source context: 17 downloads · 0 likes · Pipeline text-generation · Library transformers · Repo Jinhe/ReflectRL-Qwen2.5-Math-7B-DAPO

Note18 de jul.huggingface.co/Jinhe/ReflectRL-Qwen2.5-Math-7B-DAPO

Source context: 188 downloads · 0 likes · Repo Jinhe/ReflectRL-Qwen2.5-Math-7B-DAPO