Beyond VLM-Based Rewards: Diffusion-Native Latent Reward Modeling

• URL: http://arxiv.org/abs/2602.11146v1
• Date: Wed, 11 Feb 2026 18:57:29 GMT
• Title: Beyond VLM-Based Rewards: Diffusion-Native Latent Reward Modeling
• Authors: Gongye Liu, Bo Yang, Yida Zhi, Zhizhou Zhong, Lei Ke, Didan Deng, Han Gao, Yongxiang Huang, Kaihao Zhang, Hongbo Fu, Wenhan Luo,
• Abstract summary: DiNa-LRM is a diffusion-native latent reward model that formulates preference learning directly on noisy diffusion states.<n>Our method introduces a noise-calibrated Thurstone likelihood with diffusion-noise-dependent uncertainty.<n>Across image alignment benchmarks, DiNa-LRM substantially outperforms existing diffusion-based reward baselines.
• Score: 58.59644539594293
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Preference optimization for diffusion and flow-matching models relies on reward functions that are both discriminatively robust and computationally efficient. Vision-Language Models (VLMs) have emerged as the primary reward provider, leveraging their rich multimodal priors to guide alignment. However, their computation and memory cost can be substantial, and optimizing a latent diffusion generator through a pixel-space reward introduces a domain mismatch that complicates alignment. In this paper, we propose DiNa-LRM, a diffusion-native latent reward model that formulates preference learning directly on noisy diffusion states. Our method introduces a noise-calibrated Thurstone likelihood with diffusion-noise-dependent uncertainty. DiNa-LRM leverages a pretrained latent diffusion backbone with a timestep-conditioned reward head, and supports inference-time noise ensembling, providing a diffusion-native mechanism for test-time scaling and robust rewarding. Across image alignment benchmarks, DiNa-LRM substantially outperforms existing diffusion-based reward baselines and achieves performance competitive with state-of-the-art VLMs at a fraction of the computational cost. In preference optimization, we demonstrate that DiNa-LRM improves preference optimization dynamics, enabling faster and more resource-efficient model alignment.

Related papers

• Causal Autoregressive Diffusion Language Model [70.7353007255797]
  CARD reformulates the diffusion process within a strictly causal attention mask, enabling dense, per-token supervision in a single forward pass.<n>Our results demonstrate that CARD achieves ARM-level data efficiency while unlocking the latency benefits of parallel generation.
  arXiv  Detail & Related papers  (2026-01-29T17:38:29Z)
• APLOT: Robust Reward Modeling via Adaptive Preference Learning with Optimal Transport [37.21695864040979]
  The reward model (RM) plays a crucial role in aligning Large Language Models (LLMs) with human preferences through Reinforcement Learning.<n>This paper introduces an effective enhancement to BT-based RMs through an adaptive margin mechanism.
  arXiv  Detail & Related papers  (2025-10-13T03:13:28Z)
• G$^2$RPO: Granular GRPO for Precise Reward in Flow Models [74.21206048155669]
  We propose a novel Granular-GRPO (G$2$RPO) framework that achieves precise and comprehensive reward assessments of sampling directions.<n>We introduce a Multi-Granularity Advantage Integration module that aggregates advantages computed at multiple diffusion scales.<n>Our G$2$RPO significantly outperforms existing flow-based GRPO baselines.
  arXiv  Detail & Related papers  (2025-10-02T12:57:12Z)
• Divergence Minimization Preference Optimization for Diffusion Model Alignment [66.31417479052774]
  Divergence Minimization Preference Optimization (DMPO) is a principled method for aligning diffusion models by minimizing reverse KL divergence.<n>DMPO can consistently outperform or match existing techniques across different base models and test sets.
  arXiv  Detail & Related papers  (2025-07-10T07:57:30Z)
• VARD: Efficient and Dense Fine-Tuning for Diffusion Models with Value-based RL [28.95582264086289]
  VAlue-based Reinforced Diffusion (VARD) is a novel approach that first learns a value function predicting expection of rewards from intermediate states.<n>Our method maintains proximity to the pretrained model while enabling effective and stable training via backpropagation.
  arXiv  Detail & Related papers  (2025-05-21T17:44:37Z)
• Taming Diffusion Prior for Image Super-Resolution with Domain Shift SDEs [36.65594293655289]
  DoSSR is a Domain Shift diffusion-based SR model that capitalizes on the generative powers of pretrained diffusion models.<n>At the core of our approach is a domain shift equation that integrates seamlessly with existing diffusion models.<n>Our proposed method achieves state-of-the-art performance on synthetic and real-world datasets, while notably requiring only 5 sampling steps.
  arXiv  Detail & Related papers  (2024-09-26T12:16:11Z)
• FIND: Fine-tuning Initial Noise Distribution with Policy Optimization for Diffusion Models [10.969811500333755]
  We introduce a Fine-tuning Initial Noise Distribution (FIND) framework with policy optimization. Our method achieves 10 times faster than the SOTA approach.
  arXiv  Detail & Related papers  (2024-07-28T10:07:55Z)
• Inference-Time Alignment of Diffusion Models with Direct Noise Optimization [45.77751895345154]
  We propose a novel alignment approach, named Direct Noise Optimization (DNO), that optimize the injected noise during the sampling process of diffusion models. By design, DNO operates at inference-time, and thus is tuning-free and prompt-agnostic, with the alignment occurring in an online fashion during generation. We conduct extensive experiments on several important reward functions and demonstrate that the proposed DNO approach can achieve state-of-the-art reward scores within a reasonable time budget for generation.
  arXiv  Detail & Related papers  (2024-05-29T08:39:39Z)
• Confronting Reward Overoptimization for Diffusion Models: A Perspective of Inductive and Primacy Biases [76.9127853906115]
  Bridging the gap between diffusion models and human preferences is crucial for their integration into practical generative. We propose Temporal Diffusion Policy Optimization with critic active neuron Reset (TDPO-R), a policy gradient algorithm that exploits the temporal inductive bias of diffusion models. Empirical results demonstrate the superior efficacy of our methods in mitigating reward overoptimization.
  arXiv  Detail & Related papers  (2024-02-13T15:55:41Z)
• Low-Light Image Enhancement with Wavelet-based Diffusion Models [50.632343822790006]
  Diffusion models have achieved promising results in image restoration tasks, yet suffer from time-consuming, excessive computational resource consumption, and unstable restoration. We propose a robust and efficient Diffusion-based Low-Light image enhancement approach, dubbed DiffLL.
  arXiv  Detail & Related papers  (2023-06-01T03:08:28Z)
• How Much is Enough? A Study on Diffusion Times in Score-based Generative Models [76.76860707897413]
  Current best practice advocates for a large T to ensure that the forward dynamics brings the diffusion sufficiently close to a known and simple noise distribution. We show how an auxiliary model can be used to bridge the gap between the ideal and the simulated forward dynamics, followed by a standard reverse diffusion process.
  arXiv  Detail & Related papers  (2022-06-10T15:09:46Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.