SciVisAgentBench: A Benchmark for Evaluating Scientific Data Analysis and Visualization Agents

2026-03-30

AI Summary

This work addresses the lack of a systematic and reproducible evaluation benchmark for intelligent agents in multi-step scientific data analysis and visualization tasks. To this end, we propose the first structured evaluation framework tailored to scientific visualization, organized along four dimensions: application domain, data type, task complexity, and visualization operations, comprising 108 expert-designed, diverse cases. Our multimodal evaluation pipeline integrates large language model scoring, image-based metrics, code checkers, rule validators, and custom evaluators, with consistency validated through expert studies. The study establishes performance baselines for both SciVis-specific and general-purpose programming agents, revealing current limitations in their capabilities, and releases an extensible evaluation platform to support ongoing research and diagnostic analysis.

Abstract

Recent advances in large language models (LLMs) have enabled agentic systems that translate natural language intent into executable scientific visualization (SciVis) tasks. Despite rapid progress, the community lacks a principled and reproducible benchmark for evaluating these emerging SciVis agents in realistic, multi-step analysis settings. We present SciVisAgentBench, a comprehensive and extensible benchmark for evaluating scientific data analysis and visualization agents. Our benchmark is grounded in a structured taxonomy spanning four dimensions: application domain, data type, complexity level, and visualization operation. It currently comprises 108 expert-crafted cases covering diverse SciVis scenarios. To enable reliable assessment, we introduce a multimodal outcome-centric evaluation pipeline that combines LLM-based judging with deterministic evaluators, including image-based metrics, code checkers, rule-based verifiers, and case-specific evaluators. We also conduct a validity study with 12 SciVis experts to examine the agreement between human and LLM judges. Using this framework, we evaluate representative SciVis agents and general-purpose coding agents to establish initial baselines and reveal capability gaps. SciVisAgentBench is designed as a living benchmark to support systematic comparison, diagnose failure modes, and drive progress in agentic SciVis. The benchmark is available at https://scivisagentbench.github.io/.

Problem

• scientific visualization
• agent evaluation
• benchmark
• multimodal evaluation
• LLM-based agents

Innovation

• scientific visualization
• agent benchmark
• multimodal evaluation
• large language models
• structured taxonomy