{"slug": "slideforge-an-llm-agent-for-controllable-editing-of-slides-as-structured", "title": "SLIDEFORGE: An LLM Agent for Controllable Editing of Slides as Structured Artifacts", "summary": "Researchers at the University of Illinois Urbana-Champaign introduced SLIDEFORGE, an LLM agent framework for controllable slide editing that builds a Deck State Graph to preserve layout, style, and native editability. In experiments, SLIDEFORGE outperformed direct prompting, screenshot-based agents, and generic code-agent baselines across component recovery, preservation, restyling consistency, visual quality, and native editability. The code is available on GitHub.", "body_md": "arXiv:2609.03109v1 Announce Type: new\nAbstract: Current AI agents compellingly describe slides. However, AI-assisted slide editing requires more than understanding: the output must retain layout, style, component structure, and native editability. Towards, AI-assisted slide editing, existing agents operate on screenshots or weak document representations and often fragment coherent visual units, rasterize editable content, or break layout. In contrast, for controllable slide editing, we introduce an agentic framework, SLIDEFORGE, which builds a Deck State Graph, an executable slide state that links visual decomposition, native pptx object structure, and perceptual organization. By recovering human-referable components while retaining fine-grained editable structure, SLIDEFORGE supports theme-preserving reconstruction through slide-native operations and rendered-state verification. We further introduce an evaluation paradigm for controllable slide transformation that jointly measures component recovery, preservation, restyling consistency, visual quality, and native editability. Experiments show that SLIDEFORGE outperforms direct prompting, screenshot-based agents, and generic code-agent baselines across these dimensions. Code is available at https://github.com/UIUC-MONET/SLIDEFORGE.", "url": "https://wpnews.pro/news/slideforge-an-llm-agent-for-controllable-editing-of-slides-as-structured", "canonical_source": "https://arxiv.org/abs/2609.03109", "published_at": "2026-09-04 04:00:00+00:00", "updated_at": "2026-09-04 04:24:25.123698+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-agents", "ai-research"], "entities": ["University of Illinois Urbana-Champaign", "SLIDEFORGE", "GitHub"], "alternates": {"html": "https://wpnews.pro/news/slideforge-an-llm-agent-for-controllable-editing-of-slides-as-structured", "markdown": "https://wpnews.pro/news/slideforge-an-llm-agent-for-controllable-editing-of-slides-as-structured.md", "text": "https://wpnews.pro/news/slideforge-an-llm-agent-for-controllable-editing-of-slides-as-structured.txt", "jsonld": "https://wpnews.pro/news/slideforge-an-llm-agent-for-controllable-editing-of-slides-as-structured.jsonld"}}