Hugging Face's logo

Spaces:
Ruurd
/
llada-8b-instruct
Running on Zero

App Files Community
llada-8b-instruct / src /diffusion_lm /bigram.py

Commit History

2 days ago

  1. Add optional temperature-scaled Gumbel randomized Viterbi, retaining fast exact MAP decoding

    Ruurd
    Verified signature
    67a9c62

  2. Add optional expanded-corpus bigram and exact trigram guidance

    Ruurd
    Verified signature
    b31a17a

  3. Add optional expanded-corpus bigram and exact trigram guidance

    Ruurd
    Verified signature
    40f551e

  4. Fade repetition penalties over early block iterations; retain constant-strength option

    Ruurd
    Verified signature
    4742018

  5. Add optional joint bigram repetition-risk penalty

    Ruurd
    Verified signature
    82e83e1

  6. Decouple PMI weight from sequence sampling temperature

    Ruurd
    Verified signature
    35914ff

  7. Add optional bigram-guided sequence sampling and candidate-occurrence repetition counts

    Ruurd
    Verified signature
    a9be260

3 days ago

  1. Add optional PMI Viterbi transition repetition penalty; preserve existing defaults

    Ruurd
    Verified signature
    6e41363

4 days ago

  1. Deploy native LLaDA-Instruct with shared diffusion and PMI Viterbi controls

    Ruurd
    Verified signature
    153c673