$RSNG reasoning

Latest papers

1. SafeDriveVLA: Navigation-Conditioned World Model Dreaming for Conflict-Aware End-to-End Autonomous Driving cs.RO · 14 Oct · 1 hit

   …first driving VLA that (1) explicitly predicts a three-level navigation-compliance mode to decouple conflict reasoning from action generation, and (2) uses navigation-conditioned world model dreaming to inform this…

2. DynThinker: Unifying Static 3D and Dynamic 4D Spatial Reasoning cs.CV · 14 Oct · 3 hits

   …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…

3. GeoThinker-$F$: Endogenous Predictive World Representations for Fast Spatial Intelligence cs.CV · 14 Oct · 1 hit

   …speedup, reducing latency from $1.45\,\mathrm{s}$ to $0.40\,\mathrm{s}$ while retaining strong spatial reasoning performance.

4. Player-1: Learning to Play from Synthetic Experience cs.CV · 14 Oct · 1 hit

   …models on 6,044 curated gameplay trajectories that pair recorded actions with task instructions and visual reasoning. We construct Player-1-Gym, sixteen synthetic 2D and 3D environments designed around control…

5. Recovering Reliable Supervision via Critique-Guided On-Policy Distillation cs.LG · 14 Oct · 1 hit

   …the student uses the revision without the critique for training. On two Qwen3 students and six mathematical reasoning benchmarks, CriticOPD improves OPD, with the largest gains on competition problems.

6. PhySGA: A Physics-Space Grounded Agentic Framework for Generalizable and Interactive 3D Composition cs.AI · 14 Oct · 1 hit

   …a collaborative feedforward-feedback architecture: an Assembly Agent focuses on geometric and kinematic reasoning for global step planning and coarse adjustment, while a Critic Agent reasons over contact dynamics…

7. VASPO: Self-Distillation with Privileged Value Estimation cs.CL · 14 Oct · 2 hits

   On-policy self-distillation provides dense supervision for reasoning models via a privileged self-teacher. However, this supervision reflects local token preferences…

8. Hindsight-Privileged RLVR: Single-Rollout Token Credit Without Group Sampling cs.LG · 14 Oct · 1 hit

   …(RLVR) trains language models from checked task outcomes, but assigning a terminal reward to individual reasoning tokens or actions remains difficult. Group-relative methods such as Group Relative Policy…