transformers-community/group-beam-search
The transformers-community/group-beam-search model is a 0.8 billion parameter decoder-only transformer based on Qwen/Qwen3-0.6B, designed to enhance text generation diversity. It implements a variant of beam search that produces more varied output candidates by penalizing similar sequences across groups of beams. This approach is particularly useful for generating a wider range of high-quality text outputs, offering more creative and less repetitive results compared to standard beam search.
Loading preview...
Overview
This model implements a diverse beam search strategy, a variant of the standard beam search algorithm, to generate more varied and less repetitive text outputs. It is built upon the Qwen/Qwen3-0.6B base model, a 0.8 billion parameter decoder-only transformer. The core idea is to divide the total number of beams into groups and apply a diversity penalty to discourage similar sequences within these groups, thereby increasing the overall diversity of the generated candidates.
Key Capabilities
- Enhanced Output Diversity: Produces a broader range of distinct text sequences compared to traditional beam search.
- Configurable Parameters: Allows fine-tuning of generation behavior through
num_beams,num_beam_groups, anddiversity_penalty. - Compatibility: Works with decoder-only transformer models, matching the
group_beam_searchfunctionality found intransformers<4.56.0. - Post-processing: Integrates DoLa contrastive scoring for post-processing logits before token selection, further refining output quality.
Good For
- Applications requiring creative text generation where diverse outputs are preferred.
- Scenarios where avoiding repetitive or generic responses is crucial.
- Exploring a wider array of potential continuations in sequence generation tasks.