研究库 论文知识库

内容库 / 论文

arXiv:2610.07250 · Agent 智能体

Internalizing Agent Experience into Diffusion Model Weights via On-Policy Context Distillation

Internalizing Agent Experience into Diffusion Model Weights via On-Policy Context Distillation

类型：arxiv
标识：2610.07250
链接：https://arxiv.org/abs/2610.07250
主分类：agent
形态：method
TLDR：Wrapping an image generation model in an agentic harness can effectively boost Text-to-Image task performance: the harness can leverage memory, skills, workflow orchestration, result verification, and iterative refinement to continually construct and revise prompts, thereby eliciting better images. These gains, however, remain external to the diffusion model and are realized only while the full harness runs. We propose Diffusion On-Policy Context Distillation (D-OPCD), which treats the agent-improved prompt as privileged context and distills the knowledge encoded in the agent harness into the
副分类：multimodal
待LLM分类：否
来源文件：
/inbox/tom/_candidates/2026-10-08-agent-rag-longcontext-candidates.json

相关 · 按语义匹配

笔记批判性精读 · Entropy-Valley "Length-Adaptive Decoding for Masked Diffusion MT" `arXiv:2608.22274` · dLLM 长度自适应解码新锚

笔记 Cool Papers · cs.CL (papers.cool) · RSS 摘要

仓库 modelscope/richdreamer

笔记 LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuning

笔记 Yannic Kilcher (YouTube) · RSS 摘要

仓库 CompVis/stable-diffusion

arXiv

2610.07250

TLDR

Wrapping an image generation model in an agentic harness can effectively boost Text-to-Image task performance: the harness can leverage memory, skills, workflow orchestration, result verification, and iterative refinement to continually construct and revise prompts, thereby eliciting better images. These gains, however, remain external to the diffusion model and are realized only while the full harness runs. We propose Diffusion On-Policy Context Distillation (D-OPCD), which treats the agent-improved prompt as privileged context and distills the knowledge encoded in the agent harness into the

相关解读

arXiv ↗ 中文全文