# On the Divergence of Accuracy and Mechanism Consistency in Time Series World Models

arXiv: https://arxiv.org/abs/2610.01842 · HTML: https://mlsanity.com/paper/2610.01842/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in World models for language agents, one of 11 directions in World models.
- Builds on 22 of its 56 references that are in the map.
- First posted on arXiv 1 Oct 2026.

## Placed in

- World models / [World models for language agents](https://mlsanity.com/wm/world-models-for-language-agents/)

## Builds on (most cited first, 22 of 22)

- [A Time Series is Worth 64 Words: Long-term Forecasting with Transformers](https://mlsanity.com/paper/2211.14730/) (2022)
- [Cosmos World Foundation Model Platform for Physical AI](https://mlsanity.com/paper/2501.03575/) (2025)
- [V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning](https://mlsanity.com/paper/2506.09985/) (2025)
- [Learning Interactive Real-World Simulators](https://mlsanity.com/paper/2310.06114/) (2023)
- [DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning](https://mlsanity.com/paper/2411.04983/) (2024)
- [Long-term Forecasting with TiDE: Time-series Dense Encoder](https://mlsanity.com/paper/2304.08424/) (2023)
- [Monash Time Series Forecasting Archive](https://mlsanity.com/paper/2105.06643/) (2021)
- [GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluation](https://mlsanity.com/paper/2410.10393/) (2024)
- [Chronos-2: From Univariate to Universal Forecasting](https://mlsanity.com/paper/2510.15821/) (2025)
- [Estimating Counterfactual Treatment Outcomes over Time Through Adversarially Balanced Representations](https://mlsanity.com/paper/2002.04083/) (2020)
- [From Tables to Time: Extending TabPFN-v2 to Time Series Forecasting](https://mlsanity.com/paper/2501.02945/) (2025)
- [DAG: A Dual Correlation Network for Time Series Forecasting with Exogenous Variables](https://mlsanity.com/paper/2509.14933/) (2025)
- [ChronosX: Adapting Pretrained Time Series Models with Exogenous Variables](https://mlsanity.com/paper/2503.12107/) (2025)
- [Hybrid2 Neural ODE Causal Modeling and an Application to Glycemic Response](https://mlsanity.com/paper/2402.17233/) (2024)
- [CoRA: Covariate-Aware Adaptation of Time Series Foundation Models](https://mlsanity.com/paper/2510.12681/) (2025)
- [CLARITY: Medical World Model for Guiding Treatment Decisions by Simulating Context-Aware Disease Trajectories](https://mlsanity.com/paper/2512.08029/) (2025)
- [From Observations to States: Latent Time Series Forecasting](https://mlsanity.com/paper/2602.00297/) (2026)
- [Wastewater Treatment Plant Data for Nutrient Removal System](https://mlsanity.com/paper/2407.05346/) (2024)
- [Medical world models: representing medical states, modelling clinical dynamics and guiding intervention policies](https://mlsanity.com/paper/2606.16721/) (2026)
- [Training and Evaluating Causal Forecasting Models for Time-Series](https://mlsanity.com/paper/2411.00126/) (2024)
- [DoTime: A Synthetic Benchmark Generator for Interventional and Counterfactual Time Series](https://mlsanity.com/paper/2607.27263/) (2026)
- [Beyond Static Forecasting: Unleashing the Power of World Models for Mobile Traffic Extrapolation](https://mlsanity.com/paper/2604.08199/) (2026)

## Abstract (arXiv)

A time series world model (TSWM) predicts a controlled system's state from its observed history and planned actions and exogenous inputs. Current approaches build forecasters with actions as covariates, trained and evaluated on prediction error under the executed plan. Yet world models compare unexecuted plans, but their responses to changed plans remain untested. We ask which design choices matter and whether accurate forecasters respond to changed plans as real systems do. We address both with a formalization and benchmark. The formalization separates state, actions and exogenous inputs, distinguishes continuous, mode and event actions, and introduces mechanism consistency, a metric built on declared action-state relations with known directions, such as a vasopressor raising blood pressure: it checks whether shifting an action moves the forecast in the declared direction. The benchmark consolidates eight public datasets with real actions from engineered infrastructure and clinical care, varying prediction space, plan fusion and plan encoding across seven backbones and five seeds. First, a frozen latent prediction space lowers MAE by 9.9% over observation space and gated output fusion lowers it by 12.7% over input concatenation on average, with both improving all eight datasets; temporal plan encoding changes average MAE by at most 2.2%. Second, prediction error and mechanism consistency diverge: the lowest-error configuration is at or below chance in consistency on four of five datasets with declared mechanisms, and no design choice avoids this. Finally, directional supervision, a loss penalizing the wrong-signed part of the response to a shifted action, significantly raises consistency on penalized mechanisms with no change in MAE. Together they give TSWMs a recipe: a frozen latent space and output-side fusion for accuracy, and a training objective for mechanism consistency.