mlsanityData to 14 Oct 2026

MultiHiertt: Numerical Reasoning over Multi Hierarchical Tabular and Textual Data

arXiv 2206.01347 ↗First posted 3 Jun 2022cs.AI cs.LGcited by 121

Not on the map. Why?

Cited by 25 in LLM reasoning, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Numerical reasoning over hybrid data containing both textual and tabular content (e.g., financial reports) has recently attracted much attention in the NLP community. However, existing question answering (QA) benchmarks over hybrid data only include a single flat table in each document and thus lack examples of multi-step numerical reasoning across multiple hierarchical tables. To facilitate data analytical progress, we construct a new large-scale benchmark, MultiHiertt, with QA pairs over Multi Hierarchical Tabular and Textual data. MultiHiertt is built from a wealth of financial reports and has the following unique characteristics: 1) each document contain multiple tables and longer unstructured texts; 2) most of tables contained are hierarchical; 3) the reasoning process required for each question is more complex and challenging than existing benchmarks; and 4) fine-grained annotations of reasoning processes and supporting facts are provided to reveal complex numerical reasoning. We further introduce a novel QA model termed MT2Net, which first applies facts retrieving to extract relevant supporting facts from both tables and text and then uses a reasoning module to perform symbolic reasoning over retrieved facts. We conduct comprehensive experiments on various baselines. The experimental results show that MultiHiertt presents a strong challenge for existing baselines whose results lag far behind the performance of human experts. The dataset and code are publicly available at https://github.com/psunlpgroup/MultiHiertt.

Key facts. Cited by 121 papers in the map (data through 14 Oct 2026). Builds on 47 of its 66 references that are in the map. First posted on arXiv 3 Jun 2022.

Builds on

47 of 66 references.

References · 47

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
  4. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  5. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  6. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,148
  7. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  8. 2020Longformer: The Long-Document Transformercited by 3,413
  9. 2018Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Taskcited by 1,139
  10. 2015Teaching Machines to Read and Comprehendcited by 1,049
  11. 2019DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphscited by 903
  12. 2017Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learningcited by 687
35 more
  1. 2019MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalismscited by 662
  2. 2021FinQA: A Dataset of Numerical Reasoning over Financial Datacited by 562
  3. 2015Compositional Semantic Parsing on Semi-Structured Tablescited by 515
  4. 2020TAPAS: Weakly Supervised Table Parsing via Pre-trainingcited by 434
  5. 2018The Web as a Knowledge-base for Answering Complex Questionscited by 410
  6. 2021TAT-QA: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Financecited by 408
  7. 2019FinBERT: Financial Sentiment Analysis with Pre-trained Language Modelscited by 385
  8. 2019Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biasescited by 280
  9. 2020HybridQA: A Dataset of Multi-Hop Question Answering over Tabular and Textual Datacited by 264
  10. 2022UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Modelscited by 231
  11. 2021MultiModalQA: Complex Question Answering over Text, Tables and Imagescited by 179
  12. 2021FeTaQA: Free-form Table Question Answeringcited by 175
  13. 2020FinBERT: A Pretrained Language Model for Financial Communicationscited by 161
  14. 2020Global Table Extractor (GTE): A Framework for Joint Table Identification and Cell Structure Recognition Using Visual Contextcited by 147
  15. 2020Injecting Numerical Reasoning Skills into Language Modelscited by 133
  16. 2020Unsupervised Question Decomposition for Question Answeringcited by 118
  17. 2021HiTab: A Hierarchical Table Dataset for Question Answering and Natural Language Generationcited by 118
  18. 2020Birds have four legs?! NumerSense: Probing Numerical Commonsense Knowledge of Pre-trained Language Modelscited by 110
  19. 2020TUTA: Tree-based Transformers for Generally Structured Table Pre-trainingcited by 100
  20. 2022TableFormer: Robust Transformer Modeling for Table-Text Encodingcited by 91
  21. 2018How Much Reading Does Reading Comprehension Require? A Critical Investigation of Popular Benchmarkscited by 88
  22. 2020Understanding tables with intermediate pre-trainingcited by 68
  23. 2019Avoiding Reasoning Shortcuts: Adversarial Evaluation, Training, and Model Development for Multi-Hop QAcited by 63
  24. 2020Do Language Embeddings Capture Scales?cited by 63
  25. 2019NumNet: Machine Reading Comprehension with Numerical Reasoningcited by 59
  26. 2021AIT-QA: Question Answering Dataset over Complex Tables in the Airline Industrycited by 44
  27. 2019A Multi-Type Multi-Span Network for Reading Comprehension that Requires Discrete Reasoningcited by 41
  28. 2020Learning by Fixing: Solving Math Word Problems with Weak Supervisioncited by 25
  29. 2020An Empirical Investigation of Contextualized Number Predictioncited by 21
  30. 2021TSQA: Tabular Scenario Based Question Answeringcited by 21
  31. 2021NOAHQA: Numerical Reasoning with Interpretable Graph Question Answering Datasetcited by 14
  32. 2019Financial Event Extraction Using Wikipedia-Based Weak Supervisioncited by 11
  33. 2019Are You for Real? Detecting Identity Fraud via Dialogue Interactionscited by 8
  34. 2019A framework for anomaly detection using language modeling, and its applications to financecited by 6
  35. 2019A Time Series Analysis of Emotional Loading in Central Bank Statementscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 9; 2023: 16; 2024: 30; 2025: 35; 2026: 31.

121 papers.

Sort:

By area:

  1. 2025Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Modelscited by 665
  2. 2022Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoningcited by 350
  3. 2024What matters when building vision-language models?cited by 326
  4. 2024Mulberry: Empowering MLLM with o1-like Reasoning and Reflection via Collective Monte Carlo Tree Searchcited by 175
  5. 2024Building and better understanding vision-language models: insights and future directionscited by 173
  6. 2022A Survey of Deep Learning for Mathematical Reasoningcited by 124
  7. 2025Revisiting Reinforcement Learning for LLM Reasoning from A Cross-Domain Perspectivecited by 76
  8. 2025Eagle 2: Building Post-Training Data Strategies from Scratch for Frontier Vision-Language Modelscited by 70
  9. 2024Are LLMs Capable of Data-based Statistical and Causal Reasoning? Benchmarking Advanced Quantitative Reasoning with Datacited by 63
  10. 2023EHRXQA: A Multi-Modal Question Answering Dataset for Electronic Health Records with Chest X-ray Imagescited by 53
  11. 2025FineVision: Open Data Is All You Needcited by 53
  12. 2025DeepAnalyze: Agentic Large Language Models for Autonomous Data Sciencecited by 50
38 more
  1. 2023DocMath-Eval: Evaluating Math Reasoning Capabilities of LLMs in Understanding Long and Specialized Documentscited by 45
  2. 2022ReasTAP: Injecting Table Reasoning Skills During Pre-training via Synthetic Reasoning Examplescited by 29
  3. 2024StructLM: Towards Building Generalist Models for Structured Knowledge Groundingcited by 29
  4. 2026LocateAnything: Fast and High-Quality Vision-Language Grounding with Parallel Box Decodingcited by 28
  5. 2024Is It Really Long Context if All You Need Is Retrieval? Towards Genuinely Difficult Long Context NLPcited by 27
  6. 2024Evaluating LLMs' Mathematical Reasoning in Financial Document Question Answeringcited by 25
  7. 2023Mathematical Language Models: A Surveycited by 24
  8. 2024TAT-LLM: A Specialized Language Model for Discrete Reasoning over Tabular and Textual Datacited by 19
  9. 2024Language Modeling on Tabular Data: A Survey of Foundations, Techniques and Evolutioncited by 19
  10. 2024FinDVer: Explainable Claim Verification over Long and Hybrid-Content Financial Documentscited by 19
  11. 2023MMHQA-ICL: Multimodal In-context Learning for Hybrid Question Answering over Text, Tables and Imagescited by 18
  12. 2024Leopard: A Vision Language Model For Text-Rich Multi-Image Taskscited by 18
  13. 2025RAG over Tables: Hierarchical Memory Index, Multi-Stage Retrieval, and Benchmarkingcited by 18
  14. 2022Answering Numerical Reasoning Questions in Table-Text Hybrid Contents with Graph-based Encoder and Tree-based Decodercited by 17
  15. 2023S3HQA: A Three-Stage Approach for Multi-hop Text-Table Hybrid Question Answeringcited by 17
  16. 2024SEC-QA: A Systematic Evaluation Corpus for Financial QAcited by 17
  17. 2024TQA-Bench: Evaluating LLMs for Multi-Table Question Answeringcited by 17
  18. 2025FinDER: Financial Dataset for Question Answering and Evaluating Retrieval-Augmented Generationcited by 17
  19. 2025FinAuditing: A Financial Taxonomy-Structured Multi-Document Benchmark for Evaluating LLMscited by 14
  20. 2024Efficient Prompting for LLM-based Generative Internet of Thingscited by 12
  21. 2025A Survey on Mathematical Reasoning and Optimization with Large Language Modelscited by 12
  22. 2025RealHiTBench: A Comprehensive Realistic Hierarchical Table Benchmark for Evaluating LLM-Based Table Analysiscited by 12
  23. 2024cPAPERS: A Dataset of Situated and Multimodal Interactive Conversations in Scientific Paperscited by 11
  24. 2024OmChat: A Recipe to Train Multimodal Language Models with Strong Long Context and Video Understandingcited by 11
  25. 2025DocR1: Evidence Page-Guided GRPO for Multi-Page Document Understandingcited by 11
  26. 2026SpreadsheetBench 2: Evaluating Agents on End-to-End Business Spreadsheet Workflowscited by 11
  27. 2022DyRRen: A Dynamic Retriever-Reranker-Generator Model for Numerical Reasoning over Tabular and Textual Datacited by 10
  28. 2023Multi-View Graph Representation Learning for Answering Hybrid Numerical Reasoning Questioncited by 10
  29. 2025Tabular Data Understanding with LLMs: A Survey of Recent Advances and Challengescited by 10
  30. 2023SEER : A Knapsack approach to Exemplar Selection for In-Context HybridQAcited by 9
  31. 2024Knowledge-Aware Reasoning over Multimodal Semi-structured Tablescited by 9
  32. 2025Multimodal Large Language Models for Text-rich Image Understanding: A Comprehensive Reviewcited by 9
  33. 2025Optimizing Retrieval Strategies for Financial Question Answering Documents in Retrieval-Augmented Generation Systemscited by 9
  34. 2025MLLM-Selector: Necessity and Diversity-driven High-Value Data Selection for Enhanced Visual Instruction Tuningcited by 9
  35. 2023LoFT: Enhancing Faithfulness and Diversity for Table-to-Text Generation via Logic Form Controlcited by 8
  36. 2025Towards Temporal-Aware Multi-Modal Retrieval Augmented Generation in Financecited by 8
  37. 2025FAITH: A Framework for Assessing Intrinsic Tabular Hallucinations in Financecited by 8
  38. 2025Granite Embedding R2 Modelscited by 8

50 most cited of 121 All 121 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2024H-STAR: LLM-driven Hybrid SQL-Text Adaptive Reasoning on Tables
  2. 2026Document-Level Numerical Reasoning across Single and Multiple Tables in Financial Reports
  3. 2025TabularMath: Understanding Math Reasoning over Tables with Large Language Models
  4. 2025Plugging Schema Graph into Multi-Table QA: A Human-Guided Framework for Reducing LLM Reliance
  5. 2024Accurate and Regret-aware Numerical Problem Solver for Tabular Question Answering
  6. 2026From Documents to Reasoning: A Validated Synthetic Data Pipeline and Semantic-Aware Fine-Tuning for Financial Numerical Reasoning
  7. 2022Reasoning over Hybrid Chain for Table-and-Text Open Domain QA
  8. 2025SciTrek: Evaluating and Improving Long-Context Numerical Reasoning over Scientific Articles