Forecasting Benchmarks

Protein language models (pLMs) have shown strong potential for zero-shot prediction of missense variant effects, yet systematic benchmarking on viral proteins remains limited, a critical gap given the need for proactive tools that can anticipate emerging mutations ahead of experimental validation. Here we introduce ViroGym, a comprehensive benchmark evaluating pLMs across three tasks: 79 deep mutational scanning (DMS) assays covering eukaryotic viruses with 552,065 mutated sequences across 7 phenotypic readouts, 21 influenza neutralisation tasks, and a real-world pandemic prediction task for SARS-CoV-2. We benchmark well-established pLMs on fitness landscapes, antigenic diversity, and pandemic forecasting, and find that the ProGen2 family consistently achieves the strongest performance across all three tasks. Crucially, DMS and neutralisation performance reliably identifies models that generalise to real-world emergence, even though the mutation sets they surface barely overlap, revealing that complementary in vitro benchmarks capture the evolutionary constraints needed for real-world mutation forecasting.

Forecasting financial asset returns is challenging due to low signal-to-noise ratios, non-stationarity and heavy-tailed distributions. We benchmark 16 deep learning architectures for time series forecasting, including Transformers, State Space Models (SSMs), xLSTMs and recurrent hybrids, on a 15-year daily futures dataset using an end-to-end volatility-targeted Sharpe ratio optimisation task. Our results challenge prevailing assumptions in the literature. Several widely used 'state-of-the-art' architectures, including generic Transformers and SSMs, underperformed a simple AR(1) benchmark in a realistic financial forecasting task. The strongest results came from comparatively simple hybrid approaches using recurrent structures: a Variable Selection Network combined with an LSTM achieved the highest Sharpe ratio, while xLSTM showed the greatest robustness to transaction costs. These findings suggest that success on standard time-series benchmarks does not reliably transfer to finance, and that architectures tailored to the unique properties of asset returns may remain more effective than generic attention-based approaches for financial forecasting.

Accurate forecasting of transportation dynamics is essential for urban mobility and infrastructure planning. Although recent work has achieved strong performance with deep learning models, these methods typically require dataset-specific training, architecture design and hyper-parameter tuning. This paper evaluates whether general-purpose time-series foundation models can serve as forecasters for transportation tasks by benchmarking the zero-shot performance of the state-of-the-art model, Chronos-2, across ten real-world datasets covering highway traffic volume and flow, urban traffic speed, bike-sharing demand, and electric vehicle charging station data. Under a consistent evaluation protocol, we find that, even without any task-specific fine-tuning, Chronos-2 delivers state-of-the-art or competitive accuracy across most datasets, frequently outperforming classical statistical baselines and specialized deep learning architectures, particularly at longer horizons. Beyond point forecasting, we evaluate its native probabilistic outputs using prediction-interval coverage and sharpness, demonstrating that Chronos-2 also provides useful uncertainty quantification without dataset-specific training. In general, this study supports the adoption of time-series foundation models as a key baseline for transportation forecasting research.

Multi-horizon price forecasting is central to portfolio allocation, risk management, and algorithmic trading, yet deep learning architectures have proliferated faster than rigorous financial benchmarks can evaluate them. This study provides a controlled comparison of nine architectures (Autoformer, DLinear, iTransformer, LSTM, ModernTCN, N-HiTS, PatchTST, TimesNet, and TimeXer) spanning Transformer, MLP, CNN, and RNN families across cryptocurrency, forex, and equity index markets at 4-hour and 24-hour horizons. A total of 918 experiments were conducted under a strict five-stage protocol including fixed-seed Bayesian hyperparameter optimization, configuration freezing per asset class, multi-seed retraining, uncertainty aggregation, and statistical validation. ModernTCN achieves the best mean rank (1.333) with a 75 percent first-place rate, followed by PatchTST (2.000). Results reveal a clear three-tier ranking structure and show that architecture explains nearly all performance variance, while seed randomness is negligible. Rankings remain stable across horizons despite 2 to 2.5 times error amplification. Directional accuracy remains near 50 percent across all configurations, indicating that MSE-trained models lack directional skill at hourly resolution. The findings highlight the importance of architectural inductive bias over raw parameter count and provide reproducible guidance for multi-step financial forecasting.

Selecting the right deep learning model for power grid forecasting is challenging, as performance heavily depends on the data available to the operator. This paper presents a comprehensive benchmark of five modern neural architectures: two state space models (PowerMamba, S-Mamba), two Transformers (iTransformer, PatchTST), and a traditional LSTM. We evaluate these models on hourly electricity demand across six diverse US power grids for forecast windows between 24 and 168 hours. To ensure a fair comparison, we adapt each model with specialized temporal processing and a modular layer that cleanly integrates weather covariates. Our results reveal that there is no single best model for all situations. When forecasting using only historical load, PatchTST and the state space models provide the highest accuracy. However, when explicit weather data is added to the inputs, the rankings reverse: iTransformer improves its accuracy three times more efficiently than PatchTST. By controlling for model size, we confirm that this advantage stems from the architecture's inherent ability to mix information across different variables. Extending our evaluation to solar generation, wind power, and wholesale prices further demonstrates that model rankings depend on the forecast task: PatchTST excels on highly rhythmic signals like solar, while state space models are better suited for the chaotic fluctuations of wind and price. Ultimately, this benchmark provides grid operators with actionable guidelines for selecting the optimal forecasting architecture based on their specific data environments.

Can AI systems trained on the existing scientific record forecast the advances that will follow? We introduce PreScience, a dataset and benchmark for scientific forecasting built around 98K recent AI research papers, together with companion papers covering author publication histories and citation links, yielding 502K papers in total. The resulting paper records include titles, abstracts, disambiguated author identities, influential references, topic labels, citation trajectories, and metadata snapshotted to respect temporal cutoffs. We instantiate seven exemplar tasks: five paper-anchored tasks -- contribution generation, collaborator prediction, prior work selection, citation count prediction, and future combination prediction -- and two aggregate topic trend forecasting variants. We develop baselines ranging from simple heuristics and embedding methods to frontier language models and agentic systems, and introduce LACER, an LLM-based metric for evaluating similarity of generated contribution descriptions that agrees better with human judgments than existing metrics. Finally, we compose task models to generate a 12-month synthetic corpus and find that the resulting papers are systematically less diverse and less novel than human-authored research from the same period. We release the PreScience dataset (https://huggingface.co/datasets/allenai/prescience) and code (https://github.com/allenai/prescience).

The rapid expansion of renewable energy, particularly wind and solar power, has made reliable forecasting critical for power system operations. While recent deep learning models have achieved strong average accuracy, the increasing frequency and intensity of climate-driven extreme weather events pose severe threats to grid stability and operational security. Consequently, developing robust forecasting models that can withstand volatile conditions has become a paramount challenge. In this paper, we present R$^2$Energy, a large-scale benchmark for NWP-assisted renewable energy forecasting. It comprises over 10.7 million high-fidelity hourly records from 902 wind and solar stations across four provinces in China, providing the diverse meteorological conditions necessary to capture the wide-ranging variability of renewable generation. We further establish a standardized, leakage-free forecasting paradigm that grants all models identical access to future Numerical Weather Prediction (NWP) signals, enabling fair and reproducible comparison across state-of-the-art representative forecasting architectures. Beyond aggregate accuracy, we incorporate regime-wise evaluation with expert-aligned extreme weather annotations, uncovering a critical ``robustness gap'' typically obscured by average metrics. This gap reveals a stark robustness-complexity trade-off: under extreme conditions, a model's reliability is driven by its meteorological integration strategy rather than its architectural complexity. R$^2$Energy provides a principled foundation for evaluating and developing forecasting models for safety-critical power system applications.

Electrocardiograms (ECG) are electrical recordings of the heart that are critical for diagnosing cardiovascular conditions. ECG language models (ELMs) have recently emerged as a promising framework for ECG classification accompanied by report generation. However, current models cannot forecast future cardiac events despite the immense clinical value for planning earlier intervention. To address this gap, we propose CAMEL, the first ELM that is capable of inference over longer signal durations which enables its forecasting capability. Our key insight is a specialized ECG encoder which enables cross-understanding of ECG signals with text. We train CAMEL using established LLM training procedures, combining LoRA adaptation with a curriculum learning pipeline. Our curriculum includes ECG classification, metrics calculations, and multi-turn conversations to elicit reasoning. CAMEL demonstrates strong zero-shot performance across 6 tasks and 9 datasets, including ECGForecastBench, a new benchmark that we introduce for forecasting arrhythmias. CAMEL is on par with or surpasses ELMs and fully supervised baselines both in- and out-of-distribution, achieving SOTA results on ECGBench (+7.0% absolute average gain) as well as ECGForecastBench (+12.4% over fully supervised models and +21.1% over zero-shot ELMs).

Demand forecasting is critical for inventory planning, procurement, replenishment, production, and capacity decisions in heterogeneous supply chains. However, selecting the most appropriate model for each demand series remains challenging because performance varies across datasets, demand structures, horizons, and evaluation metrics. This study proposes the Adaptive Hybrid Selector (AHS), an automatic model-selection rule that combines observed predictive performance with structural demand information. AHS uses demand frequency and series variability to activate a hierarchical logic based on RMSSE, MAE, sMAPE, and BIAS. The proposed selector is compared with Overall Weighted Average (OWA), a benchmark-relative criterion based on sMAPE and MASE, and Equilibrium Ranking Aggregation (ERA), a comparator based on MAE, RMSE, and R^2 rankings. The empirical evaluation uses the Walmart, M3, M4, and M5 datasets, three training-testing partitions, 22 forecasting models, and horizons of up to 12 cycles. Selector performance is assessed ex post using Global Relative Accuracy (GRA), interpreted as an indicator of volumetric coherence between accumulated forecasted demand and observed demand. Results show that AHS provides the most robust overall behavior, especially in M5, while OWA remains competitive in more regular datasets. ERA shows lower volumetric coherence in most configurations. These findings suggest that automatic model selection should account for demand structure and be evaluated using ex post indicators of volumetric coherence.

Temporal knowledge graph (TKG) forecasting has advanced remarkably in recent years. Yet we found existing benchmarks inadvertently introduce a shortcut: For a subset of test queries, models can achieve optimal evaluation metrics without using any time information. We trace this to two pieces of evidence: inherent dataset biases and an overly simplistic task formulation that allows models to exploit biases, revealing fundamental flaws in current evaluation protocols for differentiating true forecasting ability. Moreover, existing datasets rely on fixed history that poorly reflects real-world dynamics, lack semantic context, and fail to explicitly mitigate shortcut, preventing thorough assessment of forecasting abilities. To address these, we propose TKGB, a comprehensive benchmark for evaluating the capability of TKG forecasting models. By integrating four complementary tasks with an automatic pipeline for constructing up-to-date and context-enriched TKG datasets with controllable shortcuts, TKGB reveals capability differences and failure modes hidden under conventional evaluation.

Occupancy world models use historical occupancy states to forecast future 3D scenes, but their robustness under corrupted temporal inputs remains poorly understood. Existing evaluations primarily emphasize clean forecasting accuracy and provide limited evidence about how errors enter, persist, and propagate through the occupancy perception-forecasting chain. This paper introduces OccStress, a robustness stress-testing benchmark for the occupancy forecasting chain. OccStress contains 21 corruption families with 61 severity configurations and 10,827 strict temporal anchors across 3 datasets. OccStress covers both 3D occupancy perception and 4D occupancy forecasting through two complementary tracks. This design separates model-mediated pipeline errors under standardized sensor stressors from the intrinsic sensitivity of 4D forecasting models to corrupted occupancy states. OccStress further defines temporal injection protocols to test whether errors in the current state, recent history, or earlier history affect future forecasts differently. OccStress provides aggregate metrics for evaluating robustness along the occupancy forecasting chain. Experiments with five 4D forecasters reveal that current occupancy models are substantially affected by both upstream prediction errors and direct state corruptions, and that clean performance alone is an insufficient description of source- and position-specific robustness. Code, data, and evaluation tools are available at https://insailab.org/OccStress.

Human pose forecasting predicts future human poses from past observations for applications including action recognition, sports analysis, and human-robot interaction. However, progress in the field is difficult to assess because reported results often rely on heterogeneous preprocessing, inconsistent metric implementations, and partially released artifacts. This paper re-examines human pose forecasting as an evaluation problem by auditing a wide range of recent forecasting methods, identifying several reproducibility issues, and introducing a unified training and evaluation pipeline for absolute pose forecasting. Under this unified protocol, mature cross-domain sequence models adapted from speech recognition provide strong baselines and obtain the lowest errors among the evaluated models. To study deployment-relevant noise, a paired clean/detector-noisy dataset variant is introduced using poses generated by a pose-estimation model. The experiments show that detector-generated poses cause substantial performance degradation compared with clean motion-capture inputs, but that part of this degradation can be recovered by adapting the forecaster on detector-generated pose sequences.

This paper presents a measurement-driven study of early warning for reliability breakdown events in 5G non-standalone (NSA) railway networks. Using 10 Hz metro-train measurement traces with serving- and neighbor-cell indicators, we benchmark six representative learning models, including CNN, LSTM, XGBoost, Anomaly Transformer, PatchTST, and TimesNet, under multiple observation windows and prediction horizons. Rather than proposing a new prediction architecture, this study develops a measurement-driven benchmark to quantify the feasibility and operating trade-offs of seconds-ahead reliability prediction in 5G NSA railway environments. Experimental results show that learning models can anticipate radio link failure (RLF)-related reliability breakdown events seconds in advance using lightweight radio features available on commercial devices. The presented benchmark provides insights for sensing-assisted communication control and offers an empirical foundation for integrating sensing and analytics into future mobility control.

Temporal cascade prediction is widely studied, yet its empirical foundations remain fragile. Most existing works report results under random cascade splits that mix past and future signals, rely on datasets with limited features and no downstream conversion labels, and compare increasingly complex models without systematically examining whether benchmark conclusions are protocol-dependent. This paper argues that the field should move from leakage-prone evaluation toward fidelity-aware benchmarking. We introduce a protocol suite and renewed evaluation standard for temporal cascade prediction, centered on the Full Temporal protocol, overlap-based leakage diagnostics, and analyses of performance inflation and temporal drift. To broaden the scope of benchmark tasks, we also present Taoke, a real-world e-commerce cascade dataset with rich promoter/product features and observed purchase conversions, enabling both first-stage popularity forecasting and second-stage conversion forecasting under a shared benchmark asset. Finally, we include CasTemp as a lightweight reference method and additionally probe a larger same-task internal extension to verify that this pipeline remains operational at substantially greater scale. Together, these components turn cascade prediction from a protocol-sensitive leaderboard exercise into a more reliable analysis and benchmarking problem, while still providing a practical reference pipeline for large-scale evaluation and conversion-aware modeling.

Benchmark quality is critical for meaningful evaluation and sustained progress in time series forecasting, particularly with the rise of pretrained models. Existing benchmarks often have limited domain coverage or overlook real-world settings such as tasks with covariates. Their aggregation procedures frequently lack statistical rigor, making it unclear whether observed performance differences reflect true improvements or random variation. Many benchmarks lack consistent evaluation infrastructure or are too rigid for integration into existing pipelines. To address these gaps, we propose fev-bench, a benchmark of 100 forecasting tasks across seven domains, including 46 with covariates. Supporting the benchmark, we introduce fev, a lightweight Python library for forecasting evaluation emphasizing reproducibility and integration with existing workflows. Using fev, fev-bench employs principled aggregation with bootstrapped confidence intervals to report performance along two dimensions: win rates and skill scores. We report results on fev-bench for pretrained, statistical, and baseline models and identify promising future research directions.

Artificial intelligence (AI) algorithms are a critical part of state-of-the-art digital health technology for diabetes management. Yet, access to large high-quality datasets is creating barriers that impede development of robust AI solutions. To accelerate development of transparent, reproducible, and robust AI solutions, we present Glucose-ML, a collection of 10 publicly available diabetes datasets, released within the last 7 years (i.e., 2018 - 2025). The Glucose-ML collection comprises over 300,000 days of continuous glucose monitor (CGM) data with a total of 38 million glucose samples collected from 2500+ people across 4 countries. Participants include persons living with type 1 diabetes, type 2 diabetes, prediabetes, and no diabetes. To support researchers and innovators with using this rich collection of diabetes datasets, we present a comparative analysis to guide algorithm developers with data selection. Additionally, we conduct a case study for the task of blood glucose prediction - one of the most common AI tasks within the field. Through this case study, we provide a benchmark for short-term blood glucose prediction across all 10 publicly available diabetes datasets within the Glucose-ML collection. We show that the same algorithm can have significantly different prediction results when developed/evaluated with different datasets. Findings from this study are then used to inform recommendations for developing robust AI solutions within the diabetes or broader health domain. We provide direct links to each longitudinal diabetes dataset and openly provide our code.

Canada experienced in 2023 one of the most severe wildfire seasons in recent history, causing damage across ecosystems, destroying communities, and emitting large quantities of CO2. This extreme wildfire season is symptomatic of a climate-change-induced increase in the length and severity of the fire season that affects the boreal ecosystem. Therefore, it is critical to empower wildfire management in boreal communities with better mitigation solutions. Wildfire probability maps represent an important tool for understanding the likelihood of wildfire occurrence and the potential severity of future wildfires. The massive increase in the availability of Earth observation data has enabled the development of deep learning-based wildfire forecasting models, aiming at providing precise wildfire probability maps at different spatial and temporal scales. A main limitation of such methods is their reliance on coarse-resolution environmental drivers and satellite products, leading to wildfire occurrence prediction of reduced resolution, typically around $\sim 0.1$°. This paper presents a benchmark dataset: CanadaFireSat, and baseline methods for high-resolution: 100 m wildfire forecasting across Canada, leveraging multi-modal data from high-resolution multi-spectral satellite images (Sentinel-2 L1C), mid-resolution satellite products (MODIS), and environmental factors (ERA5 reanalysis data). Our experiments consider two major deep learning architectures. We observe that using multi-modal temporal inputs outperforms single-modal temporal inputs across all metrics, achieving a peak performance of 60.3% in F1 score for the 2023 wildfire season, a season never seen during model training. This demonstrates the potential of multi-modal deep learning models for wildfire forecasting at high-resolution and continental scale.