Open Source
WTR-Drive: World-Task Decomposition with Structured Risk Supervision for End-to-End Driving
arXiv cs.RO0 views
arXiv:2611.20510v1 Announce Type: new Abstract: Proposal-conditioned scene predictions can lack a common object-level reference across candidate trajectories. Aggregate planning scores also leave the objects and times associated with violations implicit. We present WTR-Drive (World-Task-Risk Driving), a Drive-JEPA-based planner that separates shared scene estimation from proposal-conditioned reasoning. A proposal-independent World Memory estimates object states, map structure, and future motion by re-reading the original visual features at each refinement round. Proposal feedback updates a residual Task branch without writing back to World Memory. This branch guides refinement through candidate-specific retrieval and gated visual feedback. A final Task readout supplies candidate-conditioned features for scoring. We derive structured risk targets from existing candidate-scoring rollouts without additional simulator calls. For at-fault collision and time-to-collision violations, the targets encode event existence, association with shared object slots, and the first violation time bin. On the NAVSIM navtest split, WTR-Drive achieves 94.66 PDMS on v1 and 88.24 EPDMS on v2 (expert-frame sequential evaluation). These scores are 0.96 and 0.44 points above the published Drive-JEPA values, respectively. Candidate-level analysis shows that higher best-candidate scores can coexist with lower selected-trajectory scores, revealing a gap between candidate quality and selection effectiveness.
Source:https://arxiv.org/abs/2611.20510PreviousLift, See, Act: Hierarchical Robot Policy Pretraining with 3D Foundation ModelsNextAsymEEF: Bimanual Manipulation with Asymmetric End Effectors
Related posts
TaF-Foundation: Wrench-Grounded Cross-Sensor Tactile Representations
0arXiv cs.RO
SG-ODOM: Event-Driven Proprioceptive State Estimation with Stability-Gated Contact Updates for Legged Robots
0arXiv cs.RO
Herramientas ESG open source para pymes que funcionan en el navegador (y en valenciano)
0dev.to WebDev
GTA 6 Review: The Megayacht of Gaming Has Something for Nearly Everyone
0CNET News

Grand Theft Auto VI does something impossible — it lives up to the hype
0The Verge
Birdsong's "language-like" Zipf law is matched by simple syllable rules, and song with no words passes the whale controls
0dev.to