Advances in Agentic Computer Graphics
cs.CV
Submitted: 2026-11-01
Updated: 2026-11-01
Comments: 27 pages, 19 Figures
Project page: https://dwzhu-pku.github.io/PaperBanana
License: http://creativecommons.org/licenses/by/4.0/
The gist: Computer graphics is built on code.
Terminology
Abstract
Computer graphics is built on code. Large language model (LLM) agents can now generate graphics code and iteratively refine it through various feedback. This capability opens the door to a potentially new paradigm of computer graphics, with graphics algorithms increasingly built around agentic systems to automate progressively complex tasks. Over the past years, research on agentic graphics has expanded rapidly across a diverse range of tasks. This rapid growth makes it increasingly difficult for manual literature reviews to keep pace with new developments. At the same time, similar ideas are often described using different terminology across these tasks, further complicating efforts to develop a systematic understanding of the field. In this report, we address these challenges through two contributions. First, we introduce a unified taxonomy for characterizing agentic systems across diverse computer graphics tasks. Second, we develop an agentic literature system that can assist human researchers in analyzing existing and newly published papers at scale. Building on these contributions, we analyze emerging research trends and discuss open challenges in coordinating agents and tools across graphics tasks, designing interfaces and representations for human-agent collaboration, sustaining long-horizon planning and refinement, and reliably evaluating interactive graphics systems.
Sources
- PaperBanana: Automating Academic Illustration for AI Scientists
- Zero-to-CAD: Agentic Synthesis of Interpretable CAD Programs at Million-Scale Without Real Data
- Generating CAD Code with Vision-Language Models for 3D Designs
- MultiMat: Multimodal Program Synthesis for Procedural Materials using Large Multimodal Models
- CADSmith: Multi-Agent CAD Generation with Programmatic Geometric Validation
- HOLODECK 2.0: Vision-Language-Guided 3D World Generation with Editing
- ScanRefer: 3D Object Localization in RGB-D Scans using Natural Language
- CADReview: Automatically Reviewing CAD Programs with Error Detection and Correction
- UniCAD: A Unified Benchmark and Universal Model for Multi-Modal Multi-Task CAD
- DiagGen: Agentic Generation of Deformable Assets with Sim-based Diagnostics for Robotic Simulation
- Evaluating Large Language Models Trained on Code
- A Design-Intent-Preserving Agent Harness for Parametric CAD Modeling
- Towards Physically Grounded Articulated 3D Assets: A Multi-Agent Pipeline and Benchmark
- BlenderLLM: Training Large Language Models for Computer-Aided Design with Self-improvement
- BIMgent: Towards Autonomous Building Modeling via Computer-use Agents
- A Survey on Code Generation with LLM-based Agents
- CADWorld: Computer-Use Benchmark for Long-Horizon Computer-Aided Design
- CIT-CAD: Constraint Intent Tree-based CAD Code Generation and Verification
- ViCo: Visual-oriented Coding with Self-Reflection for Chart Replication
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs