# TFRBench: A Reasoning Benchmark for Evaluating Forecasting Systems

arXiv: https://arxiv.org/abs/2604.05364 · HTML: https://mlsanity.com/paper/2604.05364/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Time series reasoning, one of 17 directions in LLM reasoning.
- Builds on 4 of its 34 references that are in the map.
- First posted on arXiv 7 Apr 2026.

## Placed in

- LLM reasoning / [Time series reasoning](https://mlsanity.com/lr/time-series-reasoning/)

## Builds on (most cited first, 4 of 4)

- [Chronos-2: From Univariate to Universal Forecasting](https://mlsanity.com/paper/2510.15821/) (2025)
- [Explainable Artificial Intelligence (XAI) on TimeSeries Data: A Survey](https://mlsanity.com/paper/2104.00950/) (2021)
- [MTBench: A Multimodal Time Series Benchmark for Temporal Reasoning and Question Answering](https://mlsanity.com/paper/2503.16858/) (2025)
- [Forecasting Economics and Financial Time Series: ARIMA vs. LSTM](https://mlsanity.com/paper/1803.06386/) (2018)

## Built on by (most cited first, 6 of 6)

- [Agentic Forecasting using Sequential Bayesian Updating of Linguistic Beliefs](https://mlsanity.com/paper/2604.18576/) (2026)
- [Nexus : An Agentic Framework for Time Series Forecasting](https://mlsanity.com/paper/2605.14389/) (2026)
- [Reasoning-Aware Training for Time Series Forecasting](https://mlsanity.com/paper/2605.08625/) (2026)
- [LLM-based Agents for Forecasting and Prediction: Methods, Training, Evaluation, and Applications](https://mlsanity.com/paper/2608.23058/) (2026)
- [TimeLitmus: A Diagnostic Benchmark for Cross-Modal Understanding and Explanation Faithfulness in Event-Conditioned Time-Series Prediction](https://mlsanity.com/paper/2609.24677/) (2026)
- [SEER: Self-Evolving Event Reasoning and Retrieval for Time Series Forecasting](https://mlsanity.com/paper/2610.04109/) (2026)

## Abstract (arXiv)

We introduce TFRBench, the first benchmark designed to evaluate the reasoning capabilities of forecasting systems. Traditionally, time-series forecasting has been evaluated solely on numerical accuracy, treating foundation models as ``black boxes.'' Unlike existing benchmarks, TFRBench provides a protocol for evaluating the reasoning generated by forecasting systems--specifically their analysis of cross-channel dependencies, trends, and external events. To enable this, we propose a systematic multi-agent framework that utilizes an iterative verification loop to synthesize numerically grounded reasoning traces. Spanning ten datasets across five domains, our evaluation confirms that this reasoning is causally effective; useful for evaluation; and prompting LLMs with our generated traces significantly improves forecasting accuracy compared to direct numerical prediction (e.g., avg. ~40.2%→56.6%), validating the quality of our reasoning. Conversely, benchmarking experiments reveal that off-the-shelf LLMs consistently struggle with both reasoning (lower LLM-as-a-Judge scores) and numerical forecasting, frequently failing to capture domain-specific dynamics. TFRBench thus establishes a new standard for interpretable, reasoning-based evaluation in time-series forecasting. Our benchmark is available at: https://tfrbench.github.io