GeoThinker-$F$: Endogenous Predictive World Representations for Fast Spatial Intelligence

Key points

• We propose GeoThinker-$F$, an efficient framework that internalizes spatiotemporal perception by transforming causal MLLM hidden states into predictive latent world representations.1
• Our key insight is that intermediate semantic representations contain sufficient information to predict useful representations of current geometry and future semantics, without requiring an exact reproduction of features from a dedicated 3D encoder.1
• To realize this idea, GeoThinker-$F$ adopts a decoupled two-stage training strategy.1
• Extensive evaluations show that GeoThinker-$F$ eliminates runtime dependence on external 3D geometry encoders while maintaining competitive performance across diverse spatial intelligence benchmarks.1

Sources (1)

• [1]GeoThinker-$F$: Endogenous Predictive World Representations for Fast Spatial Intelligence
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 14, 05:59 PM

  Existing geometry-aware Multimodal Large Language Models (MLLMs) increasingly rely on external 3D encoders to provide structural priors, but executing these additional encoder paths at inference introduces substantial latency and computational overhead.

  We propose GeoThinker-$F$, an efficient framework that internalizes spatiotemporal perception by transforming causal MLLM hidden states into predictive latent world representations.

Extractive summary: sentences quoted from the sources.