arXiv Machine Learning
RT-Safe: Benchmarking Agent Safety in Real-Time Embodied Environment
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Jul 17
SafeRelBench: A Spatial-Relation-Aware Benchmark for Process-Level Safety in VLM-Driven Embodied Agents
arXiv:2607. 14543v1 Announce Type: cross Abstract: Vision-language models (VLMs) are increasingly used as the reasoning backbone of embodied agents, enabling robots to interpret visual scenes, follow language instructions, and plan multi-step actions.
arXiv AI
Jul 21
DSBench: A Comprehensive Benchmark for Evaluating External and In-Cabin Risks
arXiv:2511. 14592v3 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) show great promise for autonomous driving, but their suitability for safety-critical scenarios is largely unexplored, raising safety concerns.
arXiv AI
Jun 30
When Stopping Fails: Rethinking Minimal Risk Conditions through Human-Interactive Autonomous Driving for Safe Transportation Systems
arXiv:2606. 29115v1 Announce Type: cross Abstract: Autonomous vehicles (AVs) are increasingly deployed in urban environments, yet their safety frameworks remain primarily designed around collision avoidance and minimal risk condition (MRC) behaviors such as slowing or stopping when uncertainty arises.
arXiv Computation and Language
Sep 30
VehicleArena: A Realistic Urban Environment for Multi-Agent Driving
arXiv:2609.35916v1 Announce Type: cross Abstract: Real-world embodied agents often pursue independent objectives within a shared physical environment, where their actions can alter the conditions fac...
arXiv Machine Learning
Sep 3
PRISM: An Agentic Multi-Model Architecture for Proactive Safety in Autonomous Transportation Systems
PRISM (Proactive Risk Intelligence and Safety Management) is an agentic multi-model architecture designed to shift autonomous transportation safety from reactive crash avoidance to proactive, continuous risk management. It uses inverse crash‑probability modeling to transform binary crash classifiers into dynamic safety scores, and runs three specialized models—trajectory kinematics, environmental risk, and VRU interaction—coordinated by a reinforcement‑learning reasoning layer. Across 1,296 naturalistic driving scenarios, PRISM achieved a mean safety score of 68/100, classified 77.6% of situations as advisory, and flagged 3.8% as near‑misses, with 11% requiring intervention or emergency response, highlighting trajectory risk and VRU proximity as key safety factors.
arXiv AI
Sep 12
ReactHuman: A Physics-Grounded Benchmark for Human-Like Reactive Decision-Making in Embodied Multimodal LLMs
ReactHuman is a physics‑grounded benchmark that tests whether multimodal large language models (MLLMs) can make immediate, safety‑critical decisions in simulated humanoid scenarios involving sudden household hazards. The benchmark includes 17 event families, over 1,000 reproducible scenes generated from 240 Hz rigid‑body simulation, and a five‑metric suite evaluating reactions on reasonableness, safety, and physical grounding. Evaluation of seven MLLMs reveals that reactive safety remains unsolved, with models frequently mishandling hazards, relying on appearance over motion, and missing key interception points.