DynThinker: Unifying Static 3D and Dynamic 4D Spatial Reasoning
To address this interference, we propose DynThinker, a unified framework for jointly strengthening static and dynamic spatial reasoning.
ProofPaper ↗
Key points
- Achieving general spatial intelligence requires a unified understanding of both static 3D and dynamic 4D.
- However, we find that existing models struggle to excel at both: 3D post-training can weaken dynamic understanding, while 4D-specialized models often lack strong static 3D reasoning.
- Our analysis reveals that this spatial--dynamic interference is highly non-uniform across parameter groups, with the balance between static and dynamic capabilities particularly sensitive to changes in late-layer MLPs.
- Spatial-Dynamic Fusion (SDF) complements D-EMA by selectively integrating spatial and temporal priors from a dynamic geometry encoder into intermediate VLM representations under semantic guidance.
Sources (1)
- [1]DynThinker: Unifying Static 3D and Dynamic 4D Spatial ReasoningarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 14, 05:59 PM
To address this interference, we propose DynThinker, a unified framework for jointly strengthening static and dynamic spatial reasoning.
Achieving general spatial intelligence requires a unified understanding of both static 3D and dynamic 4D.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 14, 2026MAEM: Mitigating Recency Bias via a Replay-Free, Capacity-Controlled Dynamic Mixture of Experts for Class-Incremental Graph Learning
- Oct 14, 2026When Protection Overrides Consequences: Understanding Protection Mode in Large Language Model Decision-Making
- Oct 14, 2026A Diagnostic Study of World-Space Human Recovery Under High Camera Dynamics
- Oct 13, 2026How Far Can Coding Agents Go in Visual Generation?