# Synthetic Data-Driven Prompt Tuning for Financial QA over Tables and Documents

arXiv: https://arxiv.org/abs/2511.06292 · HTML: https://mlsanity.com/paper/2511.06292/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in LLM synthetic data generation, one of 13 directions in Synthetic data and self-improvement for LLMs.
- Builds on 20 of its 48 references that are in the map.
- First posted on arXiv 9 Nov 2025.

## Placed in

- Synthetic data and self-improvement for LLMs / [LLM synthetic data generation](https://mlsanity.com/synth/llm-synthetic-data-generation/)

## Builds on (most cited first, 20 of 20)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [BloombergGPT: A Large Language Model for Finance](https://mlsanity.com/paper/2303.17564/) (2023)
- [Automatic Prompt Optimization with "Gradient Descent" and Beam Search](https://mlsanity.com/paper/2305.03495/) (2023)
- [TinyStories: How Small Can Language Models Be and Still Speak Coherent English?](https://mlsanity.com/paper/2305.07759/) (2023)
- [TAT-QA: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Finance](https://mlsanity.com/paper/2105.07624/) (2021)
- [Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing](https://mlsanity.com/paper/2406.08464/) (2024)
- [Making Large Language Models Better Reasoners with Step-Aware Verifier](https://mlsanity.com/paper/2206.02336/) (2022)
- [PromptAgent: Strategic Planning with Language Models Enables Expert-level Prompt Optimization](https://mlsanity.com/paper/2310.16427/) (2023)
- [MultiHiertt: Numerical Reasoning over Multi Hierarchical Tabular and Textual Data](https://mlsanity.com/paper/2206.01347/) (2022)
- [DocMath-Eval: Evaluating Math Reasoning Capabilities of LLMs in Understanding Long and Specialized Documents](https://mlsanity.com/paper/2311.09805/) (2023)
- [PRompt Optimization in Multi-Step Tasks (PROMST): Integrating Human Feedback and Heuristic-based Sampling](https://mlsanity.com/paper/2402.08702/) (2024)
- [Towards Trustworthy and Aligned Machine Learning: A Data-centric Survey with Causality Perspectives](https://mlsanity.com/paper/2307.16851/) (2023)
- [AutoPDL: Automatic Prompt Optimization for LLM Agents](https://mlsanity.com/paper/2504.04365/) (2025)
- [REVOLVE: Optimizing AI Systems by Tracking Response Evolution in Textual Optimization](https://mlsanity.com/paper/2412.03092/) (2024)
- [Enhancing Financial Question Answering with a Multi-Agent Reflection Framework](https://mlsanity.com/paper/2410.21741/) (2024)
- [Synthetic Text Generation for Training Large Language Models via Gradient Matching](https://mlsanity.com/paper/2502.17607/) (2025)
- [Prompt Stability Matters: Evaluating and Optimizing Auto-Generated Prompt in General-Purpose Systems](https://mlsanity.com/paper/2505.13546/) (2025)
- [FinQAPT: Empowering Financial Decisions with End-to-End LLM-driven Question Answering Pipeline](https://mlsanity.com/paper/2410.13959/) (2024)
- [PREMISE: Scalable and Strategic Prompt Optimization for Efficient Mathematical Reasoning in Large Models](https://mlsanity.com/paper/2506.10716/) (2025)
- [GPT-3 Models are Few-Shot Financial Reasoners](https://mlsanity.com/paper/2307.13617/) (2023)

## Abstract (arXiv)

Financial documents like earning reports or balance sheets often involve long tables and multi-page reports. Large language models have become a new tool to help numerical reasoning and understanding these documents. However, prompt quality can have a major effect on how well LLMs perform these financial reasoning tasks. Most current methods tune prompts on fixed datasets of financial text or tabular data, which limits their ability to adapt to new question types or document structures, or they involve costly and manually labeled/curated dataset to help build the prompts. We introduce a self-improving prompt framework driven by data-augmented optimization. In this closed-loop process, we generate synthetic financial tables and document excerpts, verify their correctness and robustness, and then update the prompt based on the results. Specifically, our framework combines a synthetic data generator with verifiers and a prompt optimizer, where the generator produces new examples that exposes weaknesses in the current prompt, the verifiers check the validity and robustness of the produced examples, and the optimizer incrementally refines the prompt in response. By iterating these steps in a feedback cycle, our method steadily improves prompt accuracy on financial reasoning tasks without needing external labels. Evaluation on DocMath-Eval benchmark demonstrates that our system achieves higher performance in both accuracy and robustness than standard prompt methods, underscoring the value of incorporating synthetic data generation into prompt learning for financial applications.