Overview

Forecasting benchmarks test how well models predict future values or events in settings such as energy, weather, health, finance, and prediction markets. They compare classical statistical methods, machine-learning models, time-series foundation models, and language-model agents using fixed test sets, rolling or live evaluations, and zero-shot or fine-tuned setups. Stronger benchmarks use leakage controls, simple baselines, probabilistic scores, and tests across changing regimes, missing data, extreme events, or downstream decisions. A central challenge is that model rankings can change with the dataset, evaluation protocol, metric, or operating conditions, so low average error alone does not establish reliable or useful forecasts.

Latest papers

CardsList

247 papers21 papers in the last four weeks, down 54% on the four weeks before. 0.2% of all new papers.

  1. ForecastGym: A Live, Controlled, and Replayable Benchmark for Forecasting Agents

    Nov 17, 2026Yicheng Tao, Yuante Li, Zipei FanForecasting BenchmarksProbabilistic Forecasting

  2. RehearsalBench: Can AI Predict What Happens When You Change a Company?

    Nov 17, 2026Mika Siddiqui, Tharaka PereraAI-Assisted Decision MakingForecasting Benchmarks

  3. SeasonRainBench: Global Seasonal Precipitation

    Nov 17, 2026Qiuyan Sun, Mengxuan Chen, Haoyuan Liang +5Precipitation ForecastingForecasting Benchmarks

  4. Benchmarking the State of AI for Crop Yield Forecasting: A Global Assessment Across Modeling Paradigms

    Nov 14, 2026Michiel Kallenberg, Philip Janz, Christoph Jörges +4Crop Yield PredictionForecasting Benchmarks

  5. A systematic comparison of physics-informed neural networks and mechanistic models for epidemic modeling and forecasting

    Nov 13, 2026Martina Rama, Yuhan Li, Bruno Lepri +2Epidemic ForecastingPhysics-Informed ML

  6. CastWorld: Benchmarking Context-Aware Agentic Time Series Forecasting Beyond Predictive Accuracy

    Nov 11, 2026Bokai Pan, Ruijie Yang, Mingyue Cheng +2Context-Aware Time Series ForecastingAgentic Time Series Forecasting

  7. HydroTuring: A physics probe benchmark testing mass, energy, and momentum conservation in hydrologic models

    Nov 8, 2026Zhi Li, Changming Li, Siavash Shams +17HydrologyForecasting Benchmarks

  8. Update Slope Is Bounded by Prior Calibration: A Structural Bound and Its Contingent Empirical Signature

    Nov 6, 2026Ben Jenkins, Mihaela CardeiForecastingForecasting Benchmarks

  9. Grid-Oriented Probabilistic Load Forecasting for AI Data Centers

    Nov 6, 2026Qingyi Yao, Ding Lin, Tianqiao Zhao +2Electricity Load ForecastingProbabilistic Time Series Forecasting

  10. Benchmarking Machine Learning for Chaotic Three-Body Dynamics in Astrophysics

    Nov 5, 2026Sheng Wan, Pavan Vynatheya, Duo Xu +2Chaotic Dynamical SystemsMultibody Dynamics

  11. EnergyBench: a live, gate-aligned benchmark of day-ahead forecasts for European power markets

    Nov 5, 2026Evgenii Genov, Ander Martinez Alonso, Christoph BergmeirForecasting BenchmarksElectricity Price Forecasting

  12. SynForecast: Controlled Synthetic Time Series for Forecasting Research

    Nov 2, 2026Olivier Sprangers, Cristian Challu, Jan Rathfelder +2Time Series GenerationSynthetic Data Pretraining

  13. FinForesight: A Diagnostic Backtesting Benchmark for Deep Research Systems in Financial Forecasting

    Nov 1, 2026Fengyuan Liu, Mukai Li, Jieshun You +5Financial ForecastingForecasting Benchmarks

  14. Benchmarking Forecasting Agents on Yesterday's Tomorrows

    Oct 29, 2026Yuxuan Wang, Yishu Wang, Jiaqi DengForecasting BenchmarksAI Agent Evaluation

  15. Investor's Last Exam: Can AI Agents Do Equity Investment Without Seeing the Future?

    Oct 28, 2026Huangfei Jiang, Anning Hu, Kai Sheng +7Financial Reasoning BenchmarksAI Agent Evaluation

  16. RainCheck: Harnessing Historical Weather Data with Machine Learning for Rainfall Forecasting in Bangladesh

    Oct 28, 2026Shahran Rahman Alve, Muhammad Zawad Mahmud, Samiha Islam +1Precipitation ForecastingForecasting Benchmarks

  17. Beyond Local Accuracy: Benchmarking the Rollout Reliability of PDE Foundation Models

    Oct 22, 2026Jaewan Park, Seid Koric, Giri Krishnan +2PDE Foundation ModelsPDE Surrogate Modeling

  18. WhatIf-Bench: Benchmarking Intervention-Aware Time Series Forecasting

    Oct 19, 2026Arjun Ashok, Mihir Parmar, Palash Goyal +6Time Series ForecastingTime Series Forecasting with Covariates

  19. Beyond Static Leaderboards: Capability-Focused Evaluation with Structural Diversity for Time Series Foundation Models

    Oct 19, 2026Hongyin Zhang, Mingyi Xiang, Rongzhao Chen +7Forecasting BenchmarksTime-Series Foundation Models

  20. PathFM: Fast Generative Forecasting of Multivariate Trajectories

    Oct 15, 2026Jiachen Yao, Xiyuan Zhang, Annan Yu +5Time Series ForecastingMultivariate Time Series Forecasting

  21. Can AI Forecast the Extremes? Evaluating Probabilistic Weather Models on Tropical Cyclones

    Oct 15, 2026Nathan Chalumeau, Xavier Renard, Remi Meynadier +3Tropical Cyclone Intensity ForecastingWeather Forecasting

  22. When Does "Post-Call" Begin? An Availability Contract for Earnings-Call NLP Benchmarks

    Oct 15, 2026Xiaoyu Ma, Kelvin J. L. Koa, Yuncong Liu +1Look-Ahead Bias in Language ModelsForecasting Benchmarks

  23. JockeyBench: Evaluating LLM Agents on Horse-Racing Decision-Making

    Oct 14, 2026Zhongwei Xie, Xingsu Li, Jingqiao Yu +8LLM Agent EvaluationForecasting Benchmarks

  24. Beyond RMSE and Granger Gain: Multi-Horizon Forecast-Error Entropy for Nonlinear Predictive Coupling

    Oct 13, 2026Andrei VelichkoTime-Series Causal DiscoveryMultivariate Time Series Forecasting