AI Paper Atlas

Research map / Multi-agent systems and games / LLM agents

Improving Agent Behaviors with RL Fine-tuning for Autonomous Driving

Zhenghao Peng, Wenjie Luo, Yiren Lu, Tianyi Shen, Cole Gulino, Ari Seff, Justin Fu

ECCV 2024 · Multi-agent systems and games › LLM agents · approach: Robust and efficient agent execution · 14 citations (OpenAlex)

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 closed-loop RL fine-tuning to pre-trained modelMotionLM 1M · WOSACComposite: 0.490 -> 0.597Improved
add closed-loop RL fine-tuning to pre-trained modelMotionLM 10M · WOSACComposite: 0.549 -> 0.608Improved
add RL fine-tuningMotionLM 1M · WOSACCollision: 0.544 -> 0.863Improved
add RL fine-tuningMotionLM 1M · WOSACADE: 6.332 -> 2.436Improved
add behavioral-cloning pre-training before RL (ablation)RL fine-tuned model · WOSACComposite: 0.320 -> 0.597Improved
add collision penalty (weight 2) to reward (ablation)RL fine-tuning, 1M model · WOSACCollision: 0.834 -> 0.863Improved
add collision penalty (weight 2) to reward (ablation)RL fine-tuning, 1M model · WOSACADE: 2.405 -> 2.436Hurt
increase collision weight to 10 (ablation)RL fine-tuning, 1M model · WOSACCollision: 0.863 -> 0.831Hurt
RL fine-tune sim agents for policy evaluationMotionLM 10M · Policy evaluation vs log-playbackrank correlation: 0.845 -> 0.866Improved
RL fine-tune sim agents for policy evaluationMotionLM 10M · Policy evaluation vs log-playbackabsolute error: 10.014 -> 9.509Improved
fine-tune 10M model versus state-of-the-art MVTEFine-tuned 10M · WOSACComposite: 0.608 vs 0.645Hurt

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