English(EN) Balancing Emotional Alignment and Semantic Consistency in Image Generation via Reinforcement Learning with Valence-Arousal Anchoring

新方法平衡文本到图像生成中的情感与语义

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-14 04:00

研究人员开发了一种新的文本到图像生成情感控制方法，旨在提高情感一致性，同时不改变图像的核心语义内容。该方法结合了流匹配框架、组相对策略优化（GRPO）和一个中性语义锚点，以防止情感-语义漂移。实验表明，与现有技术相比，该方法在 Valence 和 Arousal 误差方面表现更低，CLIPScore 也有所提高，但参考图像质量略有下降。 AI

影响这项研究为微调图像生成模型以表达情感提供了一种新颖的方法，同时不牺牲语义准确性，有望实现更细致、可控的 AI 艺术。

排序理由详细介绍图像生成新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 30 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍图像生成新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CV TIER_1 English(EN) · Jisheng Dang, Zhenxuan Wang, Bin Li, Ronghao Lin, Bin Hu, Tat-Seng Chua · 2026-09-14 04:00

通过基于情感效价-唤醒锚定的强化学习，平衡图像生成中的情感一致性和语义一致性

arXiv:2609.12830v1 Announce Type: new Abstract: Continuous emotion control in text-to-image generation requires a model to improve affective alignment without changing the objects, layout, or scene described by the prompt. Existing supervised emotion-injection methods often optim…

本文如何被排名

报道来源 [1]

通过基于情感效价-唤醒锚定的强化学习，平衡图像生成中的情感一致性和语义一致性

相关实体

相关话题