arXiv AI
RepoAtlas: Guiding Coding Agents via Evolving Multimodal Repository Views
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Aug 11
Can Coding Agents Solve Repository-Level Issues with Rendered Code? An Exploratory Study of Visual Representations
arXiv:2608. 09268v1 Announce Type: cross Abstract: Visual modality has recently been explored as a way to compress textual tokens, including rendering code as images for static code understanding.
arXiv Computer Vision
Aug 25
VinciCoder: Unifying Multimodal Code Generation via Coarse-to-fine Visual Reinforcement Learning
VinciCoder is a unified framework for multimodal code generation that addresses the limitations of single-task models by training on a large-scale curated corpus of 1.3 M direct generation pairs and 300 k visual‑refinement tasks. It introduces a coarse‑to‑fine Visual Reinforcement Learning (ViRL) approach that uses visual similarity across multi‑scale patches to provide an implementation‑agnostic reward, improving alignment between rendered outputs and input visuals. Experiments on diverse benchmarks show VinciCoder outperforms existing methods, and ablation studies confirm the effectiveness of ViRL.
arXiv AI
Sep 2
Adaptive Critical Token-Aware Retrieval for Repository-Level Code Generation
arXiv:2609.01601v1 Announce Type: cross Abstract: The repository-level code generation task requires synthesizing code that satisfies task requirements while remaining consistent with the target repo...
arXiv AI
Jun 11
Exploration Structure in LLM Agents for Multi-File Change Localization
arXiv:2606. 11976v1 Announce Type: cross Abstract: Software engineering tools increasingly rely on LLM based agents to localize files to change to resolve a software issue.
arXiv AI
Sep 15
DeepDiscovery: A Location-Inference Framework for Task-Level Repository Understanding
arXiv:2606.22906v2 Announce Type: replace-cross Abstract: Large language models have shown strong performance on software engineering (SE) tasks, yet understanding large industrial repositories remai...
arXiv Computation and Language
Sep 1
Beyond NL2Code: A Structured Survey of Multimodal Code Intelligence
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.