arXiv AI
What Makes a Latent Space Shared? Direct and Recoverable Compatibility in Motion VAEs
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Computer Vision
Sep 22
Algebraic Consistency Alone Does Not Certify Temporal Structure in Latent Action Models
The paper examines latent action models that encode transitions between video frames using algebraic constraints such as additive composition and antisymmetric reversal. It demonstrates that these algebraic consistency conditions do not reliably certify temporal structure, as unconstrained models can achieve similar error reductions and constrained models still outperform unconstrained ones even after temporal pairings are destroyed. The authors find that preserving temporal pairing offers no consistent advantage on downstream tasks and that a direct repair objective yields only marginal improvement, recommending a more rigorous validation protocol.
arXiv Computer Vision
Sep 30
Why Cross-Skeleton Retargeting Is Non-Identifiable: Structural Limits of Generative Motion Models
arXiv:2609.37297v1 Announce Type: new Abstract: Cross-skeleton motion generation trains generative models to carry action structure and motion intention from one body to another. Yet a target motion...
arXiv Computer Vision
Aug 28
R2M-Bench: Evaluating Revisit Memory via Relative Consistency in Interactive Video World Models
R2M-Bench is a benchmark that evaluates revisit memory in interactive video world models by comparing a revisit pair to two control pairs from the same rollout: a gap‑matched non‑revisit pair and a short‑range pair. It introduces MemoryGain (MG) and Normalized Memory Ratio (NMR) to quantify the revisit advantage over generic temporal stability and normalize it by short‑to‑baseline dynamics. Across 300 instances and seven models, NMR correlates with human judgments and reduces the influence of slow‑motion artifacts, with DreamX‑World‑Memo achieving the highest NMR.
arXiv Computer Vision
Oct 9
Slot3R: Set-Associative Spatial Memory for Streaming 3D Reconstruction
Slot3R introduces a set‑associative spatial memory that allows multiple states to coexist at the same 3D location, addressing the limitation of Point3R where spatial proximity forces merging of distinct observations. By keeping the pretrained Point3R backbone frozen and adding a bounded sparse readout, Slot3R improves point‑cloud accuracy by up to 72% on benchmark datasets and reduces trajectory error across pose benchmarks, while maintaining real‑time performance up to 1000 frames without exhausting memory.
arXiv Machine Learning
Oct 19
Native Geometry vs. Linear Recoverability: How Alignment Changes Platonic Scaling
arXiv:2610.22749v1 Announce Type: new Abstract: Do larger models learn increasingly similar representations of the same world? The Platonic Representation Hypothesis (PRH) predicts that they should,...
arXiv AI
Jun 17
OmniDrive: An LLM-Choreographed Multi-Agent World Model with Unified Latent Co-Compression for Multi-View Driving Video Generation
arXiv:2606. 17536v1 Announce Type: cross Abstract: Generative world models for autonomous driving face two unresolved tensions: heterogeneous control injection, where free-form language, HD-maps, trajectories, and camera poses reside in incompatible representational spaces, and post-hoc cross-view fusion, where per-camera latents fail to encode global 3-D geometry.