⌁ Awesome World Model Hub Papers · Datasets · Projects
中文 GitHub
← Back to papers

Semantic Communications with World Models

arXiv 25.10 2025 52.9 method, application

TLDR

Proposes a world foundation model-aided semantic video transmission framework that uses prediction to reduce bandwidth while maintaining performance.

Reasoning

The paper innovatively integrates world models into semantic communication to predict future frames, reducing transmission overhead. Strengths include addressing low bandwidth and varying channel conditions with multiple strategies. Weaknesses are reliance on simulation results without real-world validation and potential prediction error accumulation.

Read-first score

Read-first score 52.9, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 39.

Recency 8%
86.7

Uses a gentle age decay so recent papers surface without erasing older foundations. 2025

Methodology quality 25%
60

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

Topical relevance 42%
55.7

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

Reproducibility 25%
30

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=none

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 397.

Keyword Scores

world model
9
generative world model
8
video world model
8
world dynamics prediction
8
world simulator
2
interactive world model
2
model-based reinforcement learning world model
2

Deep Analysis

Innovations

  • World foundation model (WFM)-aided semantic video transmission framework that predicts future frames from current frame and textual guidance, enabling omission of transmissions when predictions are reliable.
  • Depth-based feedback module to determine whether transmission of the current frame is needed, mitigating prediction error amplification.
  • Segmentation-assisted partial transmission method to repair degraded frames, balancing performance and bandwidth cost.
  • Active transmission strategy for mobile scenarios that exploits camera trajectory information to proactively schedule transmissions before channel quality deteriorates.

Methodology

The proposed framework uses a world foundation model (WFM) to generate future frames based on the current frame and textual guidance, allowing transmissions to be omitted when predictions remain reliable. A lightweight depth-based feedback module decides whether transmission is needed, and a segmentation-assisted partial transmission method repairs degraded frames. For mobile scenarios, an active transmission strategy leverages camera trajectory to schedule transmissions proactively.

Key Results

Simulation results show that the proposed framework significantly reduces transmission overhead while maintaining task performances across varying scenarios and channel conditions.

Limitations

  • Minor prediction errors tend to amplify over time, though a depth-based feedback module is introduced to mitigate this issue.

Resources

Paper

Tags