Hallucination in World Models is Predictable and Preventable
Summary
This paper demonstrates that hallucinations in world models are a data coverage issue, and proposes data-centric signals to detect them and coverage-aware sampling techniques to mitigate them, using the new MMBench2 dataset.
View Cached Full Text
Cached at: 06/26/26, 06:05 AM
Paper page - Hallucination in World Models is Predictable and Preventable
Source: https://huggingface.co/papers/2606.27326
Abstract
World models exhibit hallucinations in low-data regions of state-action space, which can be detected and mitigated using data-centric signals and coverage-aware sampling techniques.
Modern generativeworld modelsrender increasingly realistic action-controllable futures, yet they frequently hallucinate: rollouts remain visually fluent while drifting from the ground-truth dynamics. We hypothesize thathallucinationconcentrates in low-coverage regions of thestate-action space, where lightweightdata-centric signalscan both detect it and guide mitigation. To test this, we introduce MMBench2, a 427-hour, 210-task dataset forvisual world modelingwithground-truth actions,rewards, and live simulators, and train a 350M-parameter world model on it. We identify three distincthallucinationmodes: perceptual, action-marginalized, and scene-diverging -- each anchored to a different stage of the pipeline, and develop three signals that accurately predict where the model will fail. To close coverage gaps at training time, we develop acoverage-aware samplingtechnique; to close them online, ourhallucinationpredictors serve ascuriosity rewardsfor targeted data collection, yielding a data-efficient finetuning recipe that adapts the pretrained world model to entirely unseen environments with as few as 50 real environment trajectories. Overall, our findings reveal thathallucinationinworld modelsis inherently a data coverage issue, and that the same signals used to detect it can also be used for mitigation. An interactive web version of our paper is available at https://www.nicklashansen.com/mmbench2
View arXiv pageView PDFProject pageGitHub5Add to collection
Get this paper in your agent:
hf papers read 2606\.27326
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2606.27326 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2606.27326 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2606.27326 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
HalluWorld: A Controlled Benchmark for Hallucination via Reference World Models
HalluWorld is a controlled benchmark framework for evaluating hallucination in large language models using explicit reference world models across synthetic environments like gridworlds, chess, and realistic terminal tasks. It enables fine-grained analysis of failure modes such as perceptual hallucination, multi-step state tracking, and causal simulation, revealing that frontier models still struggle with complex reasoning not solved by extended thinking.
Hallucinations in LLMs: A Lifecycle-Based Survey of Causes, Detection, Mitigation, and Prevention
This survey paper presents a lifecycle-based framework for understanding hallucinations in LLMs, covering causes, detection, mitigation, and prevention across data, training, and inference stages.
The Detectability Gap: Hidden Heterogeneity in Hallucination Detection Across Language Models
This paper identifies a 'detectability gap' in hallucination detection: hallucinations split into high-agreement (Ghost) and low-agreement (Flickering) regimes with a 0.35–0.46 AUC gap, persisting across four models and three factual QA datasets even after freezing regime assignments and using stricter trajectory-based tests. The authors argue aggregate detection metrics hide model-dependent heterogeneity and call for regime-conditioned evaluation.
From Architecture to Output: Structural Origins of Hallucination in Large Language Models and the Amplifying Role of Data
This paper analyzes hallucination in large language models as a structural consequence of three architectural decisions: self-attention's co-occurrence learning, maximum likelihood estimation training objective, and autoregressive decoding's left-to-right commitment. It maps each mechanism to specific hallucination types and argues that dataset pathologies amplify but do not cause these vulnerabilities.
Detecting hallucinations in local models without eating VRAM: What we learned testing 1.5B to 120B models
The post benchmarks a zero-GPU, CPU-only string-normalization entropy metric (Spanda) against semantic entropy across 1.5B–120B local and frontier models, finding that 7B–27B models are well-served by cheap exact-match entropy while 120B models collapse into identical confident hallucinations that even NLI models miss.