@inproceedings{luo-etal-2026-riskcuebench,
title = "{R}isk{C}ue{B}ench: Evaluating Vision Language Models for Anticipatory Risk Reasoning in Videos",
author = "Luo, Sha and
Prabhu, Yogesh and
Ossowski, Timothy and
Chen, Kaiping and
Hu, Junjie",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.2082/",
pages = "47417--47449",
ISBN = "979-8-89176-459-0",
abstract = "Anticipating risky events before accidents unfold is essential for safety-critical AI systems and increasingly relevant in video-centric social media. However, existing video risk assessment benchmarks largely evaluate post-hoc understanding, assuming access to complete videos that include the incident itself. We introduce RiskCueBench, a benchmark spanning four risk domains that evaluates whether vision-language models (VLMs) can anticipate emerging safety concerns from incomplete visual cues. Each video is annotated with a risk signal clip, defined as the earliest segment containing sufficient cues for humans to infer elevated risk prior to incident onset. To assess not only prediction accuracy but also reasoning quality, we adopt a Question-Reasoning-Answer (QRA) framework that evaluates grounded reasoning, temporal sensitivity, and self-correction behavior. Experiments across 16 state-of-the-art VLMs reveal a substantial gap between current systems (best F1 0.69) and human performance (F1 0.98), with particularly pronounced failures on socially embedded risks such as protests and arson. These results show persistent limitations in anticipatory video reasoning and establish RiskCueBench as a testbed for safety-oriented video understanding."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="luo-etal-2026-riskcuebench">
<titleInfo>
<title>RiskCueBench: Evaluating Vision Language Models for Anticipatory Risk Reasoning in Videos</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sha</namePart>
<namePart type="family">Luo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yogesh</namePart>
<namePart type="family">Prabhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Timothy</namePart>
<namePart type="family">Ossowski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kaiping</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junjie</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Anticipating risky events before accidents unfold is essential for safety-critical AI systems and increasingly relevant in video-centric social media. However, existing video risk assessment benchmarks largely evaluate post-hoc understanding, assuming access to complete videos that include the incident itself. We introduce RiskCueBench, a benchmark spanning four risk domains that evaluates whether vision-language models (VLMs) can anticipate emerging safety concerns from incomplete visual cues. Each video is annotated with a risk signal clip, defined as the earliest segment containing sufficient cues for humans to infer elevated risk prior to incident onset. To assess not only prediction accuracy but also reasoning quality, we adopt a Question-Reasoning-Answer (QRA) framework that evaluates grounded reasoning, temporal sensitivity, and self-correction behavior. Experiments across 16 state-of-the-art VLMs reveal a substantial gap between current systems (best F1 0.69) and human performance (F1 0.98), with particularly pronounced failures on socially embedded risks such as protests and arson. These results show persistent limitations in anticipatory video reasoning and establish RiskCueBench as a testbed for safety-oriented video understanding.</abstract>
<identifier type="citekey">luo-etal-2026-riskcuebench</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2082/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>47417</start>
<end>47449</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T RiskCueBench: Evaluating Vision Language Models for Anticipatory Risk Reasoning in Videos
%A Luo, Sha
%A Prabhu, Yogesh
%A Ossowski, Timothy
%A Chen, Kaiping
%A Hu, Junjie
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F luo-etal-2026-riskcuebench
%X Anticipating risky events before accidents unfold is essential for safety-critical AI systems and increasingly relevant in video-centric social media. However, existing video risk assessment benchmarks largely evaluate post-hoc understanding, assuming access to complete videos that include the incident itself. We introduce RiskCueBench, a benchmark spanning four risk domains that evaluates whether vision-language models (VLMs) can anticipate emerging safety concerns from incomplete visual cues. Each video is annotated with a risk signal clip, defined as the earliest segment containing sufficient cues for humans to infer elevated risk prior to incident onset. To assess not only prediction accuracy but also reasoning quality, we adopt a Question-Reasoning-Answer (QRA) framework that evaluates grounded reasoning, temporal sensitivity, and self-correction behavior. Experiments across 16 state-of-the-art VLMs reveal a substantial gap between current systems (best F1 0.69) and human performance (F1 0.98), with particularly pronounced failures on socially embedded risks such as protests and arson. These results show persistent limitations in anticipatory video reasoning and establish RiskCueBench as a testbed for safety-oriented video understanding.
%U https://aclanthology.org/2026.emnlp-main.2082/
%P 47417-47449
Markdown (Informal)
[RiskCueBench: Evaluating Vision Language Models for Anticipatory Risk Reasoning in Videos](https://aclanthology.org/2026.emnlp-main.2082/) (Luo et al., EMNLP 2026)
ACL