PulseAugur
LIVE 07:05:46
research · [5 sources] ·
0
research

New benchmarks and models push AI's ability to understand research papers and generate code

By PulseAugur Editorial · Summary by gemini-2.5-flash-lite from 5 sources

Researchers have developed two new frameworks for chart-to-code generation, aiming to improve the accuracy and versatility of converting visual data into executable scripts. One approach, Chart2NCode, introduces a dataset of 176,000 charts with aligned scripts in Python, R, and LaTeX, and a model called CharLuMA that efficiently adapts to different coding languages. The other framework, CharTide, utilizes a data-centric approach with a 2 million sample dataset and an Inquiry-Driven RL framework to enhance visual perception and code logic, achieving competitive results against models like GPT-4o and GPT-5. AI

Summary written by gemini-2.5-flash-lite from 5 sources. How we write summaries →

IMPACT Advances in chart-to-code generation could streamline data visualization workflows and improve the accessibility of data analysis tools.

RANK_REASON The cluster contains two new research papers detailing novel frameworks and datasets for chart-to-code generation.

Read on arXiv cs.CV →

COVERAGE [5]

  1. arXiv cs.AI TIER_1 · Yelin Chen, Fanjin Zhang, Suping Sun, Yunhe Pang, Yuanchun Wang, Jian Song, Xiaoyan Li, Lei Hou, Shu Zhao, Jie Tang, Juanzi Li ·

    RPC-Bench: A Fine-grained Benchmark for Research Paper Comprehension

    arXiv:2601.14289v2 Announce Type: replace-cross Abstract: Understanding research papers remains challenging for foundation models due to specialized scientific discourse and complex figures and tables, yet existing benchmarks offer limited fine-grained evaluation at scale. To add…

  2. arXiv cs.CL TIER_1 · Zhihan Zhang, Lizi Liao ·

    Aligned Multi-View Scripts for Universal Chart-to-Code Generation

    arXiv:2604.24559v1 Announce Type: new Abstract: Chart-to-code generation converts a chart image into an executable plotting script, enabling faithful reproduction and editable visualizations. Existing methods are largely Python-centric, limiting practical use and overlooking a cr…

  3. arXiv cs.CL TIER_1 · Lizi Liao ·

    Aligned Multi-View Scripts for Universal Chart-to-Code Generation

    Chart-to-code generation converts a chart image into an executable plotting script, enabling faithful reproduction and editable visualizations. Existing methods are largely Python-centric, limiting practical use and overlooking a critical source of supervision: the same chart can…

  4. arXiv cs.CV TIER_1 · Xiangxi Zheng, Kuang He, Jiayi Hu, Ping Yu, Rui Yan, Yuan Yao, Peng Hou, Anxiang Zeng, Alex Jinpeng Wang ·

    CharTide: Data-Centric Chart-to-Code Generation via Tri-Perspective Tuning and Inquiry-Driven Evolution

    arXiv:2604.22192v1 Announce Type: new Abstract: Chart-to-code generation demands strict visual precision and syntactic correctness from Vision-Language Models (VLMs). However, existing approaches are fundamentally constrained by data-centric limitations: despite the availability …

  5. arXiv cs.CV TIER_1 · Alex Jinpeng Wang ·

    CharTide: Data-Centric Chart-to-Code Generation via Tri-Perspective Tuning and Inquiry-Driven Evolution

    Chart-to-code generation demands strict visual precision and syntactic correctness from Vision-Language Models (VLMs). However, existing approaches are fundamentally constrained by data-centric limitations: despite the availability of growing chart-to-code datasets, simply scalin…

RELATED ENTITIES

RELATED TOPICS