Papers for

data visualization teams

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Graph captioning improves by describing key structural motifs

Structurally Speaking: Motif-Oriented Graph Captioning through Bidirectional Graph-Text Translation

Abstract: Graph captions should help readers understand graph structure, rather than simply translate adjacency matrices into long textual edge lists. A useful graph caption abstracts connectivity into recognizable motifs, such as hubs, paths, cycles, cliques, and bridges, because these motifs provide compact structural units that are easier to read, compare, and recover. In this paper, we study motif-oriented graph captioning as a bidirectional graph-text translation task, where captions must both preserve enough topology for graph recovery and express the graph through concise motif-level descriptions. We show that direct prompting of GPT-5.1 often produces graph-recoverable captions by enumerating node-to-node connections, but these captions are verbose and can contain inconsistent motif interpretations. To address this gap, we introduce Structurally Speaking, a lightweight structured prompting protocol that guides translation between explicit connectivity and motif-level abstraction. Experiments on a synthetic motif-based dataset show that structured prompting produces shorter and more motif-consistent captions while maintaining comparable graph recovery. These results suggest that explicit topology-to-motif reasoning guidance can make LLM-generated graph captions more interpretable without model fine-tuning.

Thu 10 SeptComputation and LanguageMachine Learning
The gist
Graphs, which show how things are connected, can be hard to describe clearly using simple lists of connections. The authors study how to write shorter, easier-to-understand descriptions by focusing on common patterns called motifs, like hubs or cycles. They introduce a method that helps computers generate captions that highlight these patterns, making the descriptions more concise while still allowing the original graph to be reconstructed. Testing on simulated graphs showed their approach keeps accuracy but makes the captions easier to read and more consistent.
Open → 2609.10923v1

AI reconstructs editable Python code from scientific figures

SciFigure2Code: An AI-Reconstructed Benchmark for Scientific Figure-to-Code

Abstract: Scientific figures are the interface through which research claims are inspected and reused, but final published panels rarely expose the data or plotting code that produced them. Recovering this hidden provenance from pixels is therefore underdetermined. We introduce SciFigure2Code, an AI-reconstructed benchmark that instead evaluates presentation recovery: generating editable Python programs that preserve how a scientific panel is arranged and read. Role-specialized Codex agents generate, execute, visually refine, and audit silver-standard presentation programs that capture geometry, visual hierarchy, encodings, annotations, and typography without claiming to recover original measurements or author source code. This reconstruction-and-audit protocol turns final published panels into auditable reference packages; the resulting resource contains 6,740 reviewed panels and SciFigureBench, a balanced 337-panel test set across 31 chart subtypes, five domains, and three complexity levels. Across 14 zero-shot models in image-only and caption-assisted settings, execution, multi-component layouts, axes, legends, and scientific labels remain weak. Claude Opus 4.7 achieves the highest image-only Overall score, Claude Opus 4.6 leads caption-assisted reconstruction, and two-stage plan-then-code prompting improves Overall for all four tested models. SciFigure2Code provides an auditable testbed for agents that construct editable, visually faithful scientific figure presentations.

Tue 8 SeptComputer Vision and Pattern Recognition
The gist
Scientific papers often show figures without revealing the data or code behind them, making it hard to reuse or check them. The authors created SciFigure2Code, a system that turns images of scientific figures into Python programs that recreate their appearance and layout, but not the original data. They used AI agents to generate and refine these programs, building a large dataset and test set with many types of charts. The system helps test AI models on how well they can recreate the visual presentation of scientific charts from images.
Open → 2609.08155v1