Playable Video Generation

arXiv 2021 30.6 method

TLDR

Unsupervised learning of playable video generation where user controls video by selecting discrete actions, using self-supervised encoder-decoder with action bottleneck.

Reasoning

The paper introduces a novel unsupervised problem and a self-supervised framework that learns discrete action labels from unlabelled videos. Strengths include a clear problem formulation and demonstration on diverse datasets, but the abstract lacks explicit comparison to world models or reinforcement learning, and the evaluation details are not fully described.

Read-first score

Read-first score 30.6, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 11.

Reproducibility 25%

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

Recency 8%

Uses a gentle age decay so recent papers surface without erasing older foundations. 2021

Methodology quality 25%

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

Topical relevance 42%

15.7

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

Field roles

Candidate

Rank sensitivity

Stability: volatile; rank range: 37.

Keyword Scores

video world model

world dynamics prediction

interactive world model

world model

world simulator

generative world model

model-based reinforcement learning world model

Resources

Paper