mlsanityData to 7 Oct 2026

LOB-Bench: Benchmarking Generative AI for Finance -- an Application to Limit Order Book Data

arXiv 2502.09172 ↗First posted 13 Feb 2025cs.LG cs.CE q-fin.CPcited by 4

Not on the map. Why?

Abstract (arXiv)

While financial data presents one of the most challenging and interesting sequence modelling tasks due to high noise, heavy tails, and strategic interactions, progress in this area has been hindered by the lack of consensus on quantitative evaluation paradigms. To address this, we present LOB-Bench, a benchmark, implemented in python, designed to evaluate the quality and realism of generative message-by-order data for limit order books (LOB) in the LOBSTER format. Our framework measures distributional differences in conditional and unconditional statistics between generated and real LOB data, supporting flexible multivariate statistical evaluation. The benchmark also includes features commonly used LOB statistics such as spread, order book volumes, order imbalance, and message inter-arrival times, along with scores from a trained discriminator network. Lastly, LOB-Bench contains "market impact metrics", i.e. the cross-correlations and price response functions for specific events in the data. We benchmark generative autoregressive state-space models, a (C)GAN, as well as a parametric LOB model and find that the autoregressive GenAI approach beats traditional model classes.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 19 of its 48 references that are in the map. First posted on arXiv 13 Feb 2025.

Builds on

19 of 48 references.

References · 19

  1. 2017Attention Is All You Needcited by 59,932
  2. 2024The Llama 3 Herd of Modelscited by 14,192
  3. 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,898
  4. 2021Efficiently Modeling Long Sequences with Structured State Spacescited by 2,424
  5. 2023RWKV: Reinventing RNNs for the Transformer Eracited by 814
  6. 2024Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrencecited by 155
  7. 20191D Convolutional Neural Networks and Applications: A Surveycited by 152
  8. 2023Learning-Rate-Free Learning by D-Adaptationcited by 113
  9. 2024A Survey of Large Language Models for Financial Applications: Progress, Prospects and Challengescited by 86
  10. 2017Benchmark Dataset for Mid-Price Forecasting of Limit Order Book Data with Machine Learning Methodscited by 49
  11. 2018Market Making via Reinforcement Learningcited by 45
  12. 2019Get Real: Realism Metrics for Robust Limit Order Book Market Simulationscited by 40
7 more
  1. 2024PARDEN, Can You Repeat That? Defending against Jailbreaks via Repetitioncited by 29
  2. 2022Learning to simulate realistic limit order book markets from data as a World Agentcited by 27
  3. 2023Generative AI for End-to-End Limit Order Book Modelling: A Token-Level Autoregressive Generative Model of Message Flow Using a Deep State Space Networkcited by 19
  4. 2023Conditional Generators for Limit Order Book Environments: Explainability, Challenges, and Robustnesscited by 14
  5. 2023JAX-LOB: A GPU-Accelerated limit order book simulator to unlock large scale reinforcement learning for tradingcited by 12
  6. 2023LOB-Based Deep Learning Models for Stock Price Trend Prediction: A Benchmark Studycited by 9
  7. 2022DSLOB: A Synthetic Limit Order Book Dataset for Benchmarking Forecasting Algorithms under Distributional Shiftcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 2; 2026: 2.

4 papers.

  1. 2026TradeFM: A Generative Foundation Model for Trade-flow and Market Microstructurecited by 5
  2. 2025Predicting Liquidity-Aware Bond Yields using Causal GANs and Deep Reinforcement Learning with LLM Evaluationcited by 2
  3. 2025Reinforcement Learning in Queue-Reactive Models: Application to Optimal Executioncited by 2
  4. 2026ReLOBGen: Replayable Limit Order Book Message Generationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025ByteGen: A Tokenizer-Free Generative Model for Orderbook Events in Byte Space
  2. 2025DiffVolume: Diffusion Models for Volume Generation in Limit Order Books
  3. 2025TRADES: Generating Realistic Market Simulations with Diffusion Models
  4. 2026High-Quality Synthetic Financial Time-Series using a GAN-Diffusion Framework
  5. 2026FlowLOB: Efficient and Controllable Limit Order Book Generation with Flow Matching
  6. 2025Towards Competent AI for Fundamental Analysis in Finance: A Benchmark Dataset and Evaluation
  7. 2024FinBen: A Holistic Financial Benchmark for Large Language Models
  8. 2026DiffLOB: Diffusion Models for Counterfactual Generation in Limit Order Books