paper-with-me

Papers

OracleZoom: On-Policy Self-Distillation Inspired Reference-Constrained Recursive Image Super Resolution

2026-09-06 · Shubhashis Roy Dipta, Sourajit Saha, Shaswati Saha, Nobin Sarwar hf

Recursive Super-Resolution (SR) extends fixed-scale SR to extreme magnification by repeatedly feeding predictions back into the same model, analogous to zooming an image repeatedly. However, ground truth availability at every scale, especially at depth, remains challenging as the required source resolution grows geometrically, leaving deeper predictions unsupervised. We present OracleZoom, an on-policy distillation-inspired, reference-constrained framework that trains on its trajectory while carrying the last ground-truth evidence beyond the supervision boundary. Direct and cross-scale supervision constrain verifiable content, while a no-reference quality objective guides unresolved fine-scale detail. A KL-constrained pretrained latent prior limits quality-driven drift, while EMA consistency stabilizes the supervision boundary. Across seven datasets, OracleZoom achieves the state-of-the-art SR quality across zooming scales, averaging 0.713 CLIPIQA, with larger gains on deeper scales, while significantly reducing hallucinations. Code, data, and models are available at https://dipta007.github.io/OracleZoom/ .

📄 PDF Abstract BibTeX arXiv:2609.06490

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Preference-Based Self-Distillation: Beyond KL Matching via Reward Regularization

2026-05-06 · Xin Yu, Liuchen Liao, Yiwen Zhang, Yingchen Yu 외 arxiv

On-policy distillation is an efficient alternative to reinforcement learning, offering dense token-level training signals. However, its reliance on a stronger external teacher has driven recent work on on-policy self-dis…

Reinforcement LearningMathematical Reasoning

Self-Distilled Policy Gradient

2026-06-02 · Yifeng Liu, Shiyuan Zhang, Yifan Zhang, Quanquan Gu arxiv

On-policy self-distillation, where a language model conditions on privileged context to supervise its own generations, is a promising source of dense supervision for sparse-reward reinforcement learning. Actually, it can…

Reinforcement Learning

β-OPSD: Deriving with Policy Optimization, Training with Self-Distillation

2026-07-30 · Jiawei Xu, Minghui Liu, Juzheng Zhang, Tom Goldstein 외 arxiv

On-policy self-distillation (OPSD) is a promising approach to improve reasoning language models, but it remains brittle in practice: making it work reliably often requires substantial engineering effort. We identify a st…

Mathematical ReasoningReinforcement Learning

GenEvolve: Self-Evolving Image Generation Agents via Tool-Orchestrated Visual Experience Distillation

2026-05-20 · Sixiang Chen, Zhaohu Xing, Tian Ye, Xinyu Geng 외 arxiv

Open-ended image generation is no longer a simple prompt-to-image problem. High-quality generation often requires an agent to combine a model's internal generative ability with external resources. As requests become more…

Image Generation

ROSD: Reflective On-Policy Self-Distillation for Language Model Reasoning across Domains

2026-05-27 · Ziqi Zhao, Xinyu Ma, Liu Yang, Yujie Feng 외 arxiv

On-policy self-distillation (OPSD) improves the reasoning performance of large language models (LLMs) by providing dense token-level supervision for on-policy rollouts. However, existing OPSD methods often yield limited …

Domain Generalization