Entity · paper

EvolveNav: Proactive Preflection and Self-Evolving Memory for Zero-Shot Object Goal Navigation

paperactiveevolvenav-proactive-preflection-and-self-evolving-memory-for-zero-shot-object-goal-navigation-1cf21696·1 events·first seen Jun 17, 2026

Aliases: EvolveNav: Proactive Preflection and Self-Evolving Memory for Zero-Shot Object Goal Navigation

Co-occurring entities

EvolveNav

More like this (12)

Self-Refining Agentic Reinforcement Learning for Vision-Conditioned UAV Navigation EvolveNav Learning Adaptive Safety Margins for Visual Navigation Visually Grounded Self-Reflection for Vision-Language Models via Reinforcement Learning Visual Verification Enables Inference-time Steering and Autonomous Policy Improvement Verifiable Self-Evolution for Open-Ended Dialogue Skills via Future-Feedback Prediction InSight: Self-Guided Skill Acquisition via Steerable VLAs InSight: Self-Guided Skill Acquisition via Steerable VLAs Humanoid-GPT: Scaling Data and Structure for Zero-Shot Motion Tracking TempoVLA: Learning Speed-Controllable Vision-Language-Action Policies Online Neural Space Time Memory for Dynamic Novel View Synthesis Holonomy Memory Reinforcement Learning

Recent events (1)

4arXiv · cs.AI·Jun 17, 2026·source ↗

EvolveNav: Self-evolving memory and preflection for zero-shot object-goal navigation

EvolveNav is a new framework for Zero-Shot Object-Goal Navigation (ZS-OGN) that enables test-time improvement through a self-evolving agentic rule memory built from past trajectories. A retrieval strategy based on upper confidence bound balances semantic relevance and historical success when selecting rules, while a memory-guided preflection module forecasts action outcomes before execution to reduce inefficient exploration. The method achieves a 10.1% improvement in success rate over existing zero-shot baselines with fewer unnecessary steps.

Evaluation and Benchmarking Agent and Tool Ecosystem EvolveNav: Proactive Preflection and Self-Evolving Memory for Zero-Shot Object Goal Navigation EvolveNav