PaSBench-Video: A Streaming Video Benchmark for Proactive Safety Warning

• URL: http://arxiv.org/abs/2606.02443v1
• Date: Mon, 01 Jun 2026 16:14:01 GMT
• Title: PaSBench-Video: A Streaming Video Benchmark for Proactive Safety Warning
• Abstract summary: Video-capable large language models (MLLMs) could serve as always-on safety monitors that issue warnings during this window.<n>We present PaSBench-Video, a 740-video benchmark with 481 risk and 259 no-risk videos across four domains.<n>We find that no model exceeds 20.0% on our strictest metric, and recall is tightly coupled with false-positive rate.
• Score: 54.70033525672316
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: Between the first visible sign of danger and the moment an accident occurs, there is often a window where intervention remains possible. Video-capable multimodal large language models (MLLMs) could serve as always-on safety monitors that issue warnings during this window. Yet current benchmarks do not test this ability: they rely on static inputs, ignore timing precision, and omit false-positive measurement on safe scenes. We present PaSBench-Video, a 740-video benchmark with 481 risk and 259 no-risk videos across four domains: driving, healthcare, daily life, and industrial production. Risk videos are annotated with frame-level risk onset and accident boundaries. A model must observe the video causally and produce a warning that is both temporally calibrated and content-correct. Testing 13 MLLMs, we find that no model exceeds 20.0% on our strictest metric, and recall is tightly coupled with false-positive rate, with Pearson correlation 0.64: higher detection comes only at the cost of triggering warnings on the majority of safe clips. Performance splits sharply by domain: models achieve moderate recall at low false-positive rates in daily life, where risks are inherently anomalous, yet fire indiscriminately in driving, where routine and hazardous scenes look alike. These results indicate that current models rely on scene-level activity cues rather than reasoning about emerging harm.

Related papers

• When Seeing Overrides Knowing: Visual Dominance and Deferral-Based Method for Personalized Safety in VLMs [54.95635713318535]
  Vision-language models (VLMs) are increasingly deployed in high-stakes settings, where a response that is reasonable in general may still be unsafe for a particular user.<n>We study this problem of personalized safety in multimodal systems and introduce MPS-Bench, a benchmark of 5,181 scenarios from 584 real-world images.<n>We find that they almost always respond directly (86-99%) rather than seek missing context, and none exceeds 2.6/5 on personalized safety.<n>Motivated by this mechanism, we propose PRISM, a lightweight input monitor that uses bidirectional cross-modal modulation to predict when a query is likely to require
  arXiv  Detail & Related papers  (2026-09-03T03:58:56Z)
• SafeAtlas-VL: Beyond Binary Multimodal Safety with Large-Scale Data and Guard Models [73.04773649437375]
  Multimodal safety moderation requires distinguishing risks arising from visual content, user intent, and assistant behavior.<n>We introduce SafeAtlas-VL, a dataset of 1.5M training instances that places image-, request-, and response-level judgments on a five-level ordered scale.<n>We also construct SafeAtlas-Bench, a set of 5,000 instances for evaluating five-level predictions and continuous risk scores.
  arXiv  Detail & Related papers  (2026-08-29T07:08:01Z)
• Can We Defend Against AI-Generated Video Attacks on Real-World Crisis Events? A Systematic Evaluation of Detectors, Generators and Social Dissemination [81.00778788481266]
  RA-Bench is a benchmark for AI-generated video detection that uses Real videos as anchors.<n> RA-Bench contains 17,886 videos, comprising 1,830 real-video anchors across 10 social-risk categories.<n>None of the three detector families generalizes consistently across RA-Bench instances.
  arXiv  Detail & Related papers  (2026-08-14T15:32:56Z)
• From Sports to Safety: Benchmarking Proactive Risk Inference in MLLMs [11.582231928519754]
  SPRINT (Sports Proactive Risk INference Testbed) is a benchmark of 2,888 real-world sports videos (2,440 accident, 448 safe controls) spanning 14 sports and 3 environmental settings.<n> Accident videos feature fine-grained annotations of early hazard cues, accident timing, and hierarchical causes; safe videos are manually verified as accident-free and serve to diagnose prompt-induced false alarms.
  arXiv  Detail & Related papers  (2026-08-06T03:25:39Z)
• EgoSafetyBench: A Diagnostic Egocentric Video Benchmark for Evaluating Embodied VLMs as Runtime Safety Guards [4.6400797556929545]
  Vision-language models (VLMs) are proposed as runtime safety guards for embodied agents in homes and factories.<n>We introduce EgoSafetyBench, an egocentric video benchmark of 1,200 robot-view scenarios annotated at half-second granularity.<n>We find that while guards reliably recognize videos containing hazards, they often miss specific hazardous moments.
  arXiv  Detail & Related papers  (2026-06-30T21:50:49Z)
• Kelp: A Streaming Safeguard for Large Models via Latent Dynamics-Guided Risk Detection [29.51645496888383]
  Kelp is a novel plug-in framework that enables streaming risk detection within the LM generation pipeline.<n> Kelp consistently outperforms state-of-the-art post-hoc guardrails and prior plug-in probes.
  arXiv  Detail & Related papers  (2025-10-09T14:42:50Z)
• Read the Scene, Not the Script: Outcome-Aware Safety for LLMs [22.814397376238755]
  We trace both to a common cause: current models reason weakly about links between actions and outcomes.<n>We define this failure mode as Consequence-blindness.<n>Models fine-tuned on CS-Chain-4k show clear gains against semantic-camouflage jailbreaks.
  arXiv  Detail & Related papers  (2025-10-05T18:46:49Z)
• AccidentBench: Benchmarking Multimodal Understanding and Reasoning in Vehicle Accidents and Beyond [101.20320617562321]
  AccidentBench is a large-scale benchmark that combines vehicle accident scenarios with Beyond domains.<n>The benchmark contains approximately 2000 videos and over 19000 human-annotated question-answer pairs.
  arXiv  Detail & Related papers  (2025-09-30T17:59:13Z)
• Video-SafetyBench: A Benchmark for Safety Evaluation of Video LVLMs [45.265397990158846]
  Video-SafetyBench is the first benchmark designed to evaluate the safety of LVLMs under video-text attacks.<n>It comprises 2,264 video-text pairs spanning 48 fine-grained unsafe categories.<n>To generate semantically accurate videos for safety evaluation, we design a controllable pipeline that decomposes video semantics into subject images and motion text.
  arXiv  Detail & Related papers  (2025-05-17T05:06:38Z)
• Safe Vision-Language Models via Unsafe Weights Manipulation [75.04426753720551]
  We revise safety evaluation by introducing Safe-Ground, a new set of metrics that evaluate safety at different levels of granularity.<n>We take a different direction and explore whether it is possible to make a model safer without training, introducing Unsafe Weights Manipulation (UWM)<n>UWM uses a calibration set of safe and unsafe instances to compare activations between safe and unsafe content, identifying the most important parameters for processing the latter.
  arXiv  Detail & Related papers  (2025-03-14T17:00:22Z)
• Sample-Efficient Safety Assurances using Conformal Prediction [57.92013073974406]
  Early warning systems can provide alerts when an unsafe situation is imminent. To reliably improve safety, these warning systems should have a provable false negative rate. We present a framework that combines a statistical inference technique known as conformal prediction with a simulator of robot/environment dynamics.
  arXiv  Detail & Related papers  (2021-09-28T23:00:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.