dLLM PRM Gap Β· Causal mean-pool PRM

πŸ’» Code   β€’   πŸ€— Collection   β€’   πŸ“„ Paper

News

  • 2026-09: Accepted at NeurIPS 2026.

Public adapter release for dLLM PRM Gap. causal attention with mask-aware mean pooling for the submitted readout diagnostic

This repository contains only trainable adapter parameters and the reward head. It does not include the base model. The release configuration is recorded in config.json; the arXiv paper defines the release scope and citation.

Files

  • adapter.safetensors β€” compact adapter weights
  • config.json β€” public base-model id, architecture, and provenance

Load

from huggingface_hub import snapshot_download
from prm.checkpointing import load_diffusion_prm

path = snapshot_download("YanZhanPKU/dLLM-PRM-Gap-causal-meanpool-dream7b")
model, report = load_diffusion_prm(
    checkpoint=path,
    model_path="Dream-org/Dream-v0-Instruct-7B",
    local_files_only=False,
)

Role: causal mean-pool scorer used for the mean-pool row in the readout diagnostic.

Paper: https://arxiv.org/abs/2609.35472.

Downloads last month
-
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for YanZhanPKU/dLLM-PRM-Gap-causal-meanpool-dream7b

Adapter
(30)
this model

Collection including YanZhanPKU/dLLM-PRM-Gap-causal-meanpool-dream7b

Paper for YanZhanPKU/dLLM-PRM-Gap-causal-meanpool-dream7b