# Deep Tabular Research via Continual Experience-Driven Execution

arXiv: https://arxiv.org/abs/2603.09151 · HTML: https://mlsanity.com/paper/2603.09151/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 31 references that are in the map.
- First posted on arXiv 10 Mar 2026.

## Builds on (most cited first, 21 of 21)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [ExpeL: LLM Agents Are Experiential Learners](https://mlsanity.com/paper/2308.10144/) (2023)
- [LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Models](https://mlsanity.com/paper/2212.04088/) (2022)
- [Understanding the planning of LLM agents: A survey](https://mlsanity.com/paper/2402.02716/) (2024)
- [Chain-of-Table: Evolving Tables in the Reasoning Chain for Table Understanding](https://mlsanity.com/paper/2401.04398/) (2024)
- [TableBench: A Comprehensive and Complex Benchmark for Table Question Answering](https://mlsanity.com/paper/2408.09174/) (2024)
- [Dynamic Planning with a LLM](https://mlsanity.com/paper/2308.06391/) (2023)
- [Tabular Representation, Noisy Operators, and Impacts on Table Structure Understanding Tasks in LLMs](https://mlsanity.com/paper/2310.10358/) (2023)
- [LLM Agents Making Agent Tools](https://mlsanity.com/paper/2502.11705/) (2025)
- [AIT-QA: Question Answering Dataset over Complex Tables in the Airline Industry](https://mlsanity.com/paper/2106.12944/) (2021)
- [Multimodal Table Understanding](https://mlsanity.com/paper/2406.08100/) (2024)
- [Exploring the Numerical Reasoning Capabilities of Language Models: A Comprehensive Analysis on Tabular Data](https://mlsanity.com/paper/2311.02216/) (2023)
- [A Survey on Large Language Model-based Agents for Statistics and Data Science](https://mlsanity.com/paper/2412.14222/) (2024)
- [A Survey on Deep Tabular Learning](https://mlsanity.com/paper/2410.12034/) (2024)
- [SAMULE: Self-Learning Agents Enhanced by Multi-level Reflection](https://mlsanity.com/paper/2509.20562/) (2025)
- [RealHiTBench: A Comprehensive Realistic Hierarchical Table Benchmark for Evaluating LLM-Based Table Analysis](https://mlsanity.com/paper/2506.13405/) (2025)
- [ST-Raptor: LLM-Powered Semi-Structured Table Question Answering](https://mlsanity.com/paper/2508.18190/) (2025)
- [Testing the Limits of Unified Sequence to Sequence LLM Pretraining on Diverse Table Data Tasks](https://mlsanity.com/paper/2310.00789/) (2023)
- [Youtu-LLM: Unlocking the Native Agentic Potential for Lightweight Large Language Models](https://mlsanity.com/paper/2512.24618/) (2025)
- [Deep Learning within Tabular Data: Foundations, Challenges, Advances and Future Directions](https://mlsanity.com/paper/2501.03540/) (2025)
- [LLM-Symbolic Integration for Robust Temporal Tabular Reasoning](https://mlsanity.com/paper/2506.05746/) (2025)

## Built on by (most cited first, 5 of 5)

- [From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI](https://mlsanity.com/paper/2606.14502/) (2026)
- [Disentangling Long-Term Memory via Latent Neuro-Symbolic Reasoning](https://mlsanity.com/paper/2609.18461/) (2026)
- [WFM: Wiki Foundation Model for Complex Agentic Reasoning](https://mlsanity.com/paper/2609.18182/) (2026)
- [LADDER: Graph-Guided Diffusion Language Models for Efficient Multi-Hop Reasoning](https://mlsanity.com/paper/2609.24346/) (2026)
- [ATTUNER: Recomputation-Free KV Cache Reuse via Query-Side Adaptation](https://mlsanity.com/paper/2609.36722/) (2026)

## Abstract (arXiv)

Large language models often struggle with complex long-horizon analytical tasks over unstructured tables, which typically feature hierarchical and bidirectional headers and non-canonical layouts. We formalize this challenge as Deep Tabular Research (DTR), requiring multi-step reasoning over interdependent table regions. To address DTR, we propose a novel agentic framework that treats tabular reasoning as a closed-loop decision-making process. We carefully design a coupled query and table comprehension for path decision making and operational execution. Specifically, (i) DTR first constructs a hierarchical meta graph to capture bidirectional semantics, mapping natural language queries into an operation-level search space; (ii) To navigate this space, we introduce an expectation-aware selection policy that prioritizes high-utility execution paths; (iii) Crucially, historical execution outcomes are synthesized into a siamese structured memory, i.e., parameterized updates and abstracted texts, enabling continual refinement. Extensive experiments on challenging unstructured tabular benchmarks verify the effectiveness and highlight the necessity of separating strategic planning from low-level execution for long-horizon tabular reasoning.