The paper surveys Multimodal Code Intelligence, focusing on tasks where code is generated, edited, refined, or reasoned about under visually grounded inputs such as screenshots, charts, and videos. It categorizes the field by the role of code—rendered artifact, editable structure, intermediate reasoning trace, or executable tool interface—and organizes benchmarks into four domains: Graphical User Interface, Scientific Visualization, Structured Graphics, and Frontier Tasks and Frameworks. The authors argue that reliable evaluation must include evidence of semantics and interaction beyond visual fidelity, and propose four verification-centered research directions to advance the field toward evidence-grounded executable systems.
arXiv:2610.10066v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities in both visual understanding and code generation. However, existing...
arXiv:2610.20538v1 Announce Type: cross Abstract: Multimodal code intelligence connects the symbolic world of programs with the visual world their execution creates. Interest in this capability is bu...
arXiv:2605. 13527v3 Announce Type: replace Abstract: Reusable skills have become a core substrate for improving agent capabilities, yet most existing skill packages encode reusable behavior primarily as textual prompts, executable code, or learned routines.
Vision-language models (VLMs) have shown strong capabilities in generating visualization code from textual or visual specifications. However, real-world visualization authoring is inherently iterative: users frequently revise existing visualizations to repair flawed charts or adapt them to desired styles.
arXiv:2605. 26144v2 Announce Type: replace-cross Abstract: We present VISTA (VIsual Spec-To-App Benchmark), a benchmark for evaluating the end-to-end web-app generation capabilities of LLM-based agents.
arXiv:2607. 26769v1 Announce Type: cross Abstract: Multimodal large language models increasingly use sketches, annotations, tools, and intermediate images during reasoning, but it remains unclear whether they truly rely on these visual states.
arXiv:2609.24362v1 Announce Type: new Abstract: Sandboxed computer environments support multi-step reasoning with tools, executable programs, and persistent files, yet their extension from language m...
ATP‑Bench proposes a new benchmark for evaluating agentic tool planning in multimodal large language models (MLLMs) that generate interleaved text-and-image responses. The benchmark contains 7,702 QA pairs, including 1,592 visual‑question‑answer pairs, across eight categories and 25 visual‑critical intents, all verified by humans. A Multi‑Agent MLLM‑as‑a‑Judge (MAM) system is introduced to assess tool‑call precision, missed opportunities, and overall response quality without relying on ground‑truth references.
The paper investigates general‑purpose coding agents that generate images by writing and executing programs, contrasting them with conventional generative models. On the AgentDrawBench benchmark, these agents excel at preserving structure and meeting precise visual constraints, achieving competitive performance in tasks where compositional accuracy outweighs photorealism, though they fall short in perceptual richness. The study further shows that the agents’ visual code can serve as an external language of thought, retaining substantial information for visual question answering and suggesting an internal, compositionally structured representation that guides visual operations across different drawing interfaces.
MatToolBench is a new benchmark that evaluates multimodal GUI agents on professional materials science software. It contains 204 tasks across 10 tools in three modalities—GUI operation, OriginPro scripting, and code-based database queries—executed inside a Windows 11 VM. The benchmark offers fine-grained, expert-decomposed scoring and a high-performing multimodal judge for aesthetic assessment, revealing that strong general benchmark performance does not transfer to scientific workflows.
VETR (Verified Execution Trajectories for Refinement) is a framework that enables visual agents to recursively improve symbolic graphics reasoning by extracting actionable evidence from execution traces. It retains aligned program/render records and uses an execution refiner to compile observed writes and state changes into constrained alternative programs, which are then validated against captured runtime state. Evaluations on BlenderGym and BlenderBench with GPT‑5.6‑luna and Qwen3.8‑27B demonstrate that VETR consistently outperforms existing refinement strategies such as BlenderAlchemy and VIGA.
CoWorld Reasoner is an agentic framework that leverages a Video World Model (VWM) as a unified visual reasoning engine for Vision‑Language Models (VLMs). It exposes multiple visual operations through tools and skills, allowing the VLM to decide when to imagine, select an operation, generate visual content, and interpret the resulting evidence. The framework includes a VLM‑VWM Harness Co‑Evolution mechanism that jointly refines tools, skills, and execution code based on feedback, while keeping model weights fixed. Across seven benchmarks covering spatial, temporal, causal, and perceptual reasoning, CoWorld Reasoner outperforms direct VLM reasoning and improves accuracy on all benchmarks, achieving notable gains on MMSI and ERQA.