✨CAR-bench: Evaluating the Consistency and Limit-Awareness of LLM Agents under Real-World Uncertainty

📝 Summary:
CAR-bench evaluates LLM agent reliability under real-world uncertainty, focusing on consistency and capability awareness in in-car assistants. It introduces Hallucination and Disambiguation tasks. Baseline LLMs struggle with disambiguation and often hallucinate, highlighting a need for more relia...

🔹 Publication Date: Published on Jan 29

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.22027
• PDF: https://arxiv.org/pdf/2601.22027
• Github: https://github.com/CAR-bench/car-bench

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#LLMAgents #AI #AutonomousVehicles #AIReliability #AIUncertainty

🔥 SimWAM: A Simple World Action Model for End-to-End Autonomous Driving

💡 The paper presents SimWAM, a simple yet effective world action model for end to end autonomous driving. The problem addressed is that existing methods for world action models require costly future generation at inference, which can lead to high latency. To solve this, SimWAM uses video generation purely as a training signal and co-trains a pre-trained video expert and a lightweight action expert with joint flow matching.

An isolated attention mask keeps action prediction independent of future frames, allowing the video branch to be discarded after training and leaving a self-contained planner that directly predicts trajectories. The video backbone could be replaced and the action expert scaled independently without modifying the learning objective or inference pipeline.

The authors further apply reinforcement learning to optimize a compositional driving reward beyond trajectory imitation. The results show that SimWAM achieves 91.5 PDM on NAVSIM, surpasses state of the art WAM-based planners with substantially lower latency, and transfers zero-shot to nuScenes.

Overall, SimWAM is a simple yet solid baseline that could readily benefit from advances in video generation for efficient autonomous driving. The code and model weights are available for further development and improvement.

📅 Published on Aug 7

🔗 Links:
• GitHub: https://github.com/huggingface
• arXiv: https://arxiv.org/abs/2608.07468
• PDF: https://arxiv.org/pdf/2608.07468

🤖 Models citing this paper:
• https://huggingface.co/H-EmbodVis/SimWAM

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#AutonomousDriving #WorldActionModel #EndToEndLearning #AutonomousVehicles #ActionPredictionModels