# InsightChain: Optimized Chain-of-Insight Analytics for LLM-driven Data Visualization

arXiv: https://arxiv.org/abs/2609.06438 · HTML: https://mlsanity.com/paper/2609.06438/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 23 of its 45 references that are in the map.
- First posted on arXiv 6 Sep 2026.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)

## Builds on (most cited first, 23 of 23)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [Large Language Models Are Human-Level Prompt Engineers](https://mlsanity.com/paper/2211.01910/) (2022)
- [Automatic Prompt Optimization with "Gradient Descent" and Beam Search](https://mlsanity.com/paper/2305.03495/) (2023)
- [Optimizing Instructions and Demonstrations for Multi-Stage Language Model Programs](https://mlsanity.com/paper/2406.11695/) (2024)
- [Chart-to-Text: A Large-Scale Benchmark for Chart Summarization](https://mlsanity.com/paper/2203.06486/) (2022)
- [ChartGPT: Leveraging LLMs to Generate Charts from Abstract Natural Language](https://mlsanity.com/paper/2311.01920/) (2023)
- [Is GPT-4 a Good Data Analyst?](https://mlsanity.com/paper/2305.15038/) (2023)
- [A Comprehensive Analysis of the Effectiveness of Large Language Models as Automatic Dialogue Evaluators](https://mlsanity.com/paper/2312.15407/) (2023)
- [InsightBench: Evaluating Business Analytics Agents Through Multi-Step Insight Generation](https://mlsanity.com/paper/2407.06423/) (2024)
- [Calliope: Automatic Visual Data Story Generation from a Spreadsheet](https://mlsanity.com/paper/2010.09975/) (2020)
- [DAComp: Benchmarking Data Agents across the Full Data Intelligence Lifecycle](https://mlsanity.com/paper/2512.04324/) (2025)
- [Multimodal DeepResearcher: Generating Text-Chart Interleaved Reports From Scratch with Agentic Framework](https://mlsanity.com/paper/2506.02454/) (2025)
- [Reinforcement Learning Foundations for Deep Research Systems: A Survey](https://mlsanity.com/paper/2509.06733/) (2025)
- [CoSteer: Collaborative Decoding-Time Personalization via Local Delta Steering](https://mlsanity.com/paper/2507.04756/) (2025)
- [Data-to-Dashboard: Multi-Agent LLM Framework for Insightful Visualization in Enterprise Analytics](https://mlsanity.com/paper/2505.23695/) (2025)
- [Aligning Language Models Using Follow-up Likelihood as Reward Signal](https://mlsanity.com/paper/2409.13948/) (2024)
- [SpecSteer: Synergizing Local Context and Global Reasoning for Efficient Personalized Generation](https://mlsanity.com/paper/2603.16219/) (2026)
- [SRR-Judge: Step-Level Rating and Refinement for Enhancing Search-Integrated Reasoning in Search Agents](https://mlsanity.com/paper/2602.07773/) (2026)
- [InsightEval: An Expert-Curated Benchmark for Assessing Insight Discovery in LLM-Driven Data Agents](https://mlsanity.com/paper/2511.22884/) (2025)
- [IE as Cache: Information Extraction Enhanced Agentic Reasoning](https://mlsanity.com/paper/2604.14930/) (2026)
- [AsymSpec: Context-Asymmetric Speculative Decoding for Agentic LLMs](https://mlsanity.com/paper/2608.26004/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are increasingly used for automated data visualization, yet existing approaches often frame visualization generation as a single-step mapping from user query to figure or code, overlooking the iterative analytical reasoning process of expert analysts. We present InsightChain, a four-stage visualization prompting pipeline (Explore--Focus--Test--Present) that emulates expert analytical workflows, together with VG-COPRO, a vision-guided automatic prompt optimization (APO) method adapted to jointly optimize such multi-stage, executable pipelines. To address the evaluation gap for complex data visualization, we introduce the Insight Progression Metric (IPM), a rubric combining four text-based dimensions with a vision-based dimension. We assess IPM through a 100-chain human pilot and an expanded 300-chain agent-based evaluation spanning all ten domains. Experiments on public datasets show that InsightChain consistently outperforms competing prompting baselines. Existing APO methods fail to yield consistent gains on this multi-stage task, whereas VG-COPRO improves performance in both in-domain and cross-domain settings.