UCM: Unifying Camera Control and Memory with Time-aware Positional Encoding Warping for World Models

arXiv 26.2 2026 43.4 method

TLDR

UCM unifies camera control and long-term memory in video-based world models using time-aware positional encoding warping, outperforming state-of-the-art on real and synthetic benchmarks.

Reasoning

The paper introduces a novel mechanism (time-aware positional encoding warping) to address key limitations in world models: long-term consistency and camera control. Strengths include a dual-stream diffusion transformer for efficiency and a scalable data curation strategy. Weaknesses are not explicitly discussed in the abstract, but the claims are well-supported by experiments on both real-world and synthetic benchmarks.

Read-first score

Read-first score 43.4, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 48.

Recency 6%

100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Topical relevance 29%

68.6

Uses existing LLM keyword relevance scores normalized to 0-100. world model,world simulator,generative world model,interactive world model,video world model,world dynamics prediction,model-based reinforcement learning world model

Methodology quality 18%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=benchmark,experiment

Reproducibility 18%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=github

Citation impact 18%

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. cited_by_count=0

Citation velocity 12%

Citation velocity estimates citations per publication-year to reduce old-paper bias. velocity=0.00

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 355.

Keyword Scores

world model

generative world model

video world model

interactive world model

world simulator

world dynamics prediction

model-based reinforcement learning world model

Deep Analysis

Innovations

Unifying camera control and long-term memory for world models via a time-aware positional encoding warping mechanism
Efficient dual-stream diffusion transformer for high-fidelity generation with reduced computational overhead
Scalable data curation strategy using point-cloud-based rendering to simulate scene revisiting, enabling training on over 500K monocular videos

Methodology

UCM introduces a time-aware positional encoding warping mechanism that establishes explicit spatial correspondence between frames, unifying long-term memory and precise camera control. It employs an efficient dual-stream diffusion transformer to generate high-fidelity videos while reducing computational overhead. Training is performed on over 500K monocular videos curated via a point-cloud-based rendering pipeline that simulates scene revisiting.

Key Results

Extensive experiments on real-world and synthetic benchmarks show that UCM significantly outperforms state-of-the-art methods in long-term scene consistency and achieves precise camera controllability in high-fidelity video generation.

Resources

Paper Project