World Models Meet Language Models: On the Complementarity of Concrete and Abstract Reasoning

Hugging Face Daily Papers Papers

Summary

This paper proposes Privileged-Future On-Policy Self-Distillation (PF-OPSD) for controlled concrete reasoning, combining world models' visual simulation with language models' abstract reasoning to improve prediction accuracy and robustness on two new benchmarks.

World models and multimodal large language models (MLLMs) provide complementary capabilities for predicting future outcomes from static visual observations. World models can generate concrete visual rollouts of possible futures, while MLLMs can reason abstractly over questions, goals, and rules. However, generated rollouts are stochastic and may be visually plausible but task-incorrect, making it necessary to determine when visual simulation is useful, whether a rollout is credible, and how it should influence the final answer. We formulate this problem as controlled concrete reasoning, where a model learns to invoke, verify, and integrate visual future simulation alongside abstract reasoning. To study this setting, we construct two human-verified benchmarks, VRQABench for controllable spatial lookahead and OpenWorldQA for open-domain physical prediction, and propose Privileged-Future On-Policy Self-Distillation (PF-OPSD). During training, PF-OPSD uses ground-truth future videos and answers only as teacher-side privileged context to evaluate on-policy concrete-reasoning trajectories, while the deployable student never observes true futures at test time. Experimental results show that PF-OPSD outperforms baseline by 10.6% and 10.9% on VRQABench and OpenWorldQA, respectively, while increasing robustness to noisy or conflicting rollouts. Our code and dataset are available at https://github.com/yczhou001/PF-OPSD.
Original Article
View Cached Full Text

Cached at: 06/03/26, 07:36 AM

Paper page - World Models Meet Language Models: On the Complementarity of Concrete and Abstract Reasoning

Source: https://huggingface.co/papers/2606.03603

Abstract

Controlled concrete reasoning combines visual simulation with abstract reasoning through a training method that uses privileged future information to improve prediction accuracy and robustness.

World modelsandmultimodal large language models(MLLMs) provide complementary capabilities for predicting future outcomes from static visual observations.World modelscan generate concretevisual rolloutsof possible futures, while MLLMs can reason abstractly over questions, goals, and rules. However, generated rollouts are stochastic and may be visually plausible but task-incorrect, making it necessary to determine whenvisual simulationis useful, whether a rollout is credible, and how it should influence the final answer. We formulate this problem ascontrolled concrete reasoning, where a model learns to invoke, verify, and integrate visual future simulation alongsideabstract reasoning. To study this setting, we construct two human-verified benchmarks, VRQABench for controllable spatial lookahead and OpenWorldQA for open-domain physical prediction, and propose Privileged-FutureOn-Policy Self-Distillation(PF-OPSD). During training,PF-OPSDuses ground-truth future videos and answers only as teacher-sideprivileged contextto evaluate on-policy concrete-reasoning trajectories, while the deployable student never observes true futures at test time. Experimental results show thatPF-OPSDoutperforms baseline by 10.6% and 10.9% on VRQABench and OpenWorldQA, respectively, while increasing robustness to noisy or conflicting rollouts. Our code and dataset are available at https://github.com/yczhou001/PF-OPSD.

View arXiv pageView PDFGitHub1Add to collection

Get this paper in your agent:

hf papers read 2606\.03603

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/2606.03603 in a model README.md to link it from this page.

Datasets citing this paper2

#### YCZhou/vrqa_bench Updatedabout 1 hour ago #### YCZhou/openworld_qa Updatedabout 1 hour ago

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.03603 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

More Thought, Less Accuracy? On the Dual Nature of Reasoning in Vision-Language Models

Papers with Code Trending

This paper uncovers that prolonged reasoning in vision-language models can impair perceptual grounding, causing recognition failures on basic visual questions. It proposes Vision-Anchored Policy Optimization (VAPO) to steer reasoning toward visually grounded trajectories, achieving state-of-the-art performance with the VAPO-Thinker-7B model.