# Low-Rank Dynamics-Effective Latent Carriers for Counterfactual Rollout in Learned World Models

arXiv: https://arxiv.org/abs/2608.15156 · HTML: https://mlsanity.com/paper/2608.15156/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Latent predictive world models, one of 11 directions in World models.
- Builds on 26 of its 29 references that are in the map.
- First posted on arXiv 15 Aug 2026.

## Placed in

- World models / [Latent predictive world models](https://mlsanity.com/wm/latent-predictive-world-models/)

## Builds on (most cited first, 26 of 26)

- [World Models](https://mlsanity.com/paper/1803.10122/) (2018)
- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [Learning Latent Dynamics for Planning from Pixels](https://mlsanity.com/paper/1811.04551/) (2018)
- [Concept Bottleneck Models](https://mlsanity.com/paper/2007.04612/) (2020)
- [V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning](https://mlsanity.com/paper/2506.09985/) (2025)
- [Genie: Generative Interactive Environments](https://mlsanity.com/paper/2402.15391/) (2024)
- [Designing and Interpreting Probes with Control Tasks](https://mlsanity.com/paper/1909.03368/) (2019)
- [Interaction Networks for Learning about Objects, Relations and Physics](https://mlsanity.com/paper/1612.00222/) (2016)
- [CLEVRER: CoLlision Events for Video REpresentation and Reasoning](https://mlsanity.com/paper/1910.01442/) (2019)
- [Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection](https://mlsanity.com/paper/2004.07667/) (2020)
- [LEACE: Perfect linear concept erasure in closed form](https://mlsanity.com/paper/2306.03819/) (2023)
- [Finding Alignments Between Interpretable Causal Variables and Distributed Neural Representations](https://mlsanity.com/paper/2303.02536/) (2023)
- [DeepMDP: Learning Continuous Latent Space Models for Representation Learning](https://mlsanity.com/paper/1906.02736/) (2019)
- [Physion: Evaluating Physical Prediction from Vision in Humans and Machines](https://mlsanity.com/paper/2106.08261/) (2021)
- [SlotFormer: Unsupervised Visual Dynamics Simulation with Object-Centric Models](https://mlsanity.com/paper/2210.05861/) (2022)
- [Inducing Causal Structure for Interpretable Neural Networks](https://mlsanity.com/paper/2112.00826/) (2021)
- [Intuitive physics understanding emerges from self-supervised pretraining on natural videos](https://mlsanity.com/paper/2502.11831/) (2025)
- [Action-Sufficient State Representation Learning for Control with Structural Constraints](https://mlsanity.com/paper/2110.05721/) (2021)
- [How Reliable are Causal Probing Interventions?](https://mlsanity.com/paper/2408.15510/) (2024)
- [What Makes Video World Model Latents Action-Relevant: Prediction over Reconstruction](https://mlsanity.com/paper/2606.07687/) (2026)
- [Is Forward Prediction Enough? Physical State Grounding for JEPA World Models](https://mlsanity.com/paper/2608.06799/) (2026)
- [What Do World Models Learn in RL? Probing Latent Representations in Learned Environment Simulators](https://mlsanity.com/paper/2603.21546/) (2026)
- [GAUGE: A Measurement-Grounded Benchmark for Physical Fidelity in Simulation Engines and Video World Models](https://mlsanity.com/paper/2608.05948/) (2026)
- [CRONOS: Benchmarking Counterfactual Physical Consistency in Video Models](https://mlsanity.com/paper/2605.23699/) (2026)
- [Causal Physics Steering in Video World Models via Concept Activation Vectors](https://mlsanity.com/paper/2605.24322/) (2026)
- [Latent State Design for World Models under Sufficiency Constraints](https://mlsanity.com/paper/2605.01694/) (2026)

## Built on by (most cited first, 2 of 2)

- [RolloutFaith: Auditing Persistent Internal Interventions in Visual World Model](https://mlsanity.com/paper/2609.36843/) (2026)
- [When Generative Models Act as Simulators](https://mlsanity.com/paper/2610.12847/) (2026)

## Abstract (arXiv)

We ask whether a small, directly addressable hidden-state intervention can place a learned world model on an intended counterfactual future and then let the model's own dynamics carry that future forward. In a controlled two-object collision environment, we study a 192-dimensional recurrent model trained on factual and locally edited counterfactual trajectories. Candidate carriers are learned from training-only counterfactual-minus-factual hidden differences, and an affine map predicts carrier coordinates from the factual state and requested edit without access to the native counterfactual hidden state at test time. For bounded single-component velocity edits, rank 4 is the smallest tested rank on the preregistered grid that satisfies the development criteria. A one-shot rank-4 patch launches a 12-transition autonomous rollout without future observations, teacher forcing, repeated hidden-state correction, or physical-state clamping. The frozen procedure satisfies the preregistered 2-of-3 fresh-checkpoint replication rule and remains reusable at nearby anchors. The same Single-derived carrier and Single-only affine map also support bounded same-object two-component requests. Across the matched training regimes, broader counterfactual support was associated mainly with better Joint rollout accuracy and more additive Joint hidden responses. Composition-related structure is enriched in the rank-4 subspace but is not confined to it, and local recurrent diagnostics show strong one-step coupling from the carrier to the rest of the hidden state. A position-edit stress test fails the required specificity controls. Together, these results support a compact dynamics-effective intervention-entry interface, not a closed four-dimensional state or an intrinsic state dimension.