AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

EE-RL: Vision Language Guided Reinforcement Learning with Explorer and Expert model for End-to-End Autonomous Driving

Xiaolong Li, Lan Yang, Ruyang Li, Shan Fang, Yang Liu, Xiangmo Zhao

CVPR 2026 · Reinforcement learning › Visual and representation RL · approach: Probabilistic and autonomous RL

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add multi-level attention to the explorer (ablation)EE-RL · CARLA (train Town01, eval Town02)DS %: 51.80 -> 81.94Improved
add LoRA fine-tuning and quantization of the VLM expert (ablation)EE-RL · CARLA (train Town01, eval Town02)DS %: 64.93 -> 81.94Improved
add INT4 quantization on top of attention and LoRA (ablation)EE-RL · CARLA (train Town01, eval Town02)DS %: 75.82 -> 81.94Improved
add INT4 quantization on top of attention and LoRA (ablation)EE-RL · CARLA (train Town01, eval Town02)CoR %: 4.26 -> 3.72No clear effect
add StateHash to skip redundant VLM reasoning (two VLMs) (ablation)EE-RL · expert experiences after 1x10^6 stepsnumber of expert experiences: 17733 -> 90566Improved
add StateHash (single VLM) (ablation)EE-RL · expert experiences after 1x10^6 stepsnumber of expert experiences: 8126 -> 53783Improved
use two VLMs in parallel instead of one (ablation)EE-RL · expert experiences after 1x10^6 stepsnumber of expert experiences: 8126 -> 17733Improved
sample expert experience below 10% of batch from VLM buffer (ablation)dual replay buffer · Town02 traffic light and obstacle avoidancetask successHurt
raise expert sample ratio to 19% (traffic light) or 24% (obstacle) (ablation)dual replay buffer · Town02 sparse-critical scenariostraining duration: shortest at 19% and 24%Improved
raise expert sample ratio beyond the optimum (ablation)dual replay buffer · Town02 sparse-critical scenariostraining durationHurt
use TD3 or SAC backbone instead of DDPGEE-RL · CARLA Town05-06 testcomposite score (CS): 76.59 -> 78.66 (TD3), 80.09 (SAC)Improved
EE-RL versus VLM-RL on collision rateEE-RL · CARLA training Towns 01-03CoR %: 9.21 -> 10.76 (Town03, TD3)Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy