RIWANav: Recursive World-Action Models with Self-Improvement for Urban Navigation
In this paper, we introduce RIWANAV, a post-training framework that casts the coupled adaptation of a world model and an action model (policy) as task-specific recursive self-improvement (RSI).
ProofPaper ↗
Key points
- Long-horizon urban navigation requires sequential local decisions whose errors can compound over time.
- Imitation learning (IL) rarely learns from failures, while physical trial-and-error reinforcement learning (RL) is costly.
- Action-conditioned world models can provide imagined feedback by predicting visual consequences for candidate actions.
- Experiments show that RIWANAV outperforms training baselines and prior methods, validating the proposed recursive self-improvement loop between the policy and world model.
Sources (1)
- [1]RIWANav: Recursive World-Action Models with Self-Improvement for Urban NavigationarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 04:33 PM
In this paper, we introduce RIWANAV, a post-training framework that casts the coupled adaptation of a world model and an action model (policy) as task-specific recursive self-improvement (RSI).
Long-horizon urban navigation requires sequential local decisions whose errors can compound over time.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 6, 2026SPW-Nav Streams Language-Guided Panoramic Video in Real Time
- Oct 5, 2026From Traces to Agentic Worlds: Agentic Language World Models for Interactive Environment Simulation
- Sep 30, 2026GroundingPI: A Grounding Foundation Model towards Physical Intelligence with Visual Primitives
- Sep 30, 2026GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed
- Sep 29, 2026Introducing Quine: An AI research system designed for the complexity of biology
- Jun 9, 2026Powering the future of robotics in Europe