LPS-Bench: Benchmarking Safety Awareness of Computer-Use Agents in Long-Horizon Planning under Benign and Adversarial Scenarios

Fuente: arXiv

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Chen, Tianyu, Hu, Chujia, Gao, Ge, Liu, Dongrui, Hu, Xia, Wang, Wenjie
Format:	Preprint
Veröffentlicht:	2026
Schlagworte:	Artificial Intelligence
Online-Zugang:	Acceder al recurso
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Ähnliche Einträge

A Trajectory-Based Safety Audit of Clawdbot (OpenClaw)
von: Chen, Tianyu, et al.
Veröffentlicht: (2026)

UltraHorizon: Benchmarking Agent Capabilities in Ultra Long-Horizon Scenarios
von: Luo, Haotian, et al.
Veröffentlicht: (2025)

TRIP-Bench: A Benchmark for Long-Horizon Interactive Agents in Real-World Scenarios
von: Shen, Yuanzhe, et al.
Veröffentlicht: (2026)

CookBench: A Long-Horizon Embodied Planning Benchmark for Complex Cooking Scenarios
von: Cai, Muzhen, et al.
Veröffentlicht: (2025)

RiOSWorld: Benchmarking the Risk of Multimodal Computer-Use Agents
von: Yang, Jingyi, et al.
Veröffentlicht: (2025)

OS-Marathon: Benchmarking Computer-Use Agents on Long-Horizon Repetitive Tasks
von: Wu, Jing, et al.
Veröffentlicht: (2026)

EgoPlan-Bench2: A Benchmark for Multimodal Large Language Model Planning in Real-World Scenarios
von: Qiu, Lu, et al.
Veröffentlicht: (2024)

IS-Bench: Evaluating Interactive Safety of VLM-Driven Embodied Agents in Daily Household Tasks
von: Lu, Xiaoya, et al.
Veröffentlicht: (2025)

WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluation
von: Ding, Shuangrui, et al.
Veröffentlicht: (2026)

MobilityBench: A Benchmark for Evaluating Route-Planning Agents in Real-World Mobility Scenarios
von: Song, Zhiheng, et al.
Veröffentlicht: (2026)

SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks
von: Orlanski, Gabriel, et al.
Veröffentlicht: (2026)

SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces
von: Jin, Chang, et al.
Veröffentlicht: (2026)

HeroBench: A Benchmark for Long-Horizon Planning and Structured Reasoning in Virtual Worlds
von: Anokhin, Petr, et al.
Veröffentlicht: (2025)

Adversarial Artifact Detection in EEG-Based Brain-Computer Interfaces
von: Chen, Xiaoqing, et al.
Veröffentlicht: (2022)

LongGenBench: Long-context Generation Benchmark
von: Liu, Xiang, et al.
Veröffentlicht: (2024)

TRACE: Trajectory Risk-Aware Compression for Long-Horizon Agent Safety
von: Hong, Zhepei, et al.
Veröffentlicht: (2026)

The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents
von: Ding, Xuwei, et al.
Veröffentlicht: (2026)

ELHPlan: Efficient Long-Horizon Task Planning for Multi-Agent Collaboration
von: Ling, Shaobin, et al.
Veröffentlicht: (2025)

ColorBench: Benchmarking Mobile Agents with Graph-Structured Framework for Complex Long-Horizon Tasks
von: Song, Yuanyi, et al.
Veröffentlicht: (2025)

GroupTravelBench: Benchmarking LLM Agents on Multi-Person Travel Planning
von: Cheng, Xiang, et al.
Veröffentlicht: (2026)

LifeBench: A Benchmark for Long-Horizon Multi-Source Memory
von: Cheng, Zihao, et al.
Veröffentlicht: (2026)

DecisionBench: A Benchmark for Emergent Delegation in Long-Horizon Agentic Workflows
von: Gao, Yuxuan, et al.
Veröffentlicht: (2026)

Synthetic Computers at Scale for Long-Horizon Productivity Simulation
von: Ge, Tao, et al.
Veröffentlicht: (2026)

AD-Bench: A Real-World, Trajectory-Aware Advertising Analytics Benchmark for LLM Agents
von: Hu, Lingxiang, et al.
Veröffentlicht: (2026)

NL2Repo-Bench: Towards Long-Horizon Repository Generation Evaluation of Coding Agents
von: Ding, Jingzhe, et al.
Veröffentlicht: (2025)

ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis
von: Li, Yu, et al.
Veröffentlicht: (2026)

ComplexFuncBench: Exploring Multi-Step and Constrained Function Calling under Long-Context Scenario
von: Zhong, Lucen, et al.
Veröffentlicht: (2025)

HorizonBench: Long-Horizon Personalization with Evolving Preferences
von: Li, Shuyue Stella, et al.
Veröffentlicht: (2026)

Empowering LLMs with Parameterized Skills for Adversarial Long-Horizon Planning
von: Cui, Sijia, et al.
Veröffentlicht: (2025)

Enhancing Incipient Fault Detection for Interface Converter Sensors through Signal Correlation Analysis
von: Chujia Guo, et al.
Veröffentlicht: (2024)

Benign Samples Matter! Fine-tuning On Outlier Benign Samples Severely Breaks Safety
von: Guan, Zihan, et al.
Veröffentlicht: (2025)

Agent-RewardBench: Towards a Unified Benchmark for Reward Modeling across Perception, Planning, and Safety in Real-World Multimodal Agents
von: Men, Tianyi, et al.
Veröffentlicht: (2025)

KellyBench: A Benchmark for Long-Horizon Sequential Decision Making
von: Grady, Thomas, et al.
Veröffentlicht: (2026)

EgoPlan-Bench: Benchmarking Multimodal Large Language Models for Human-Level Planning
von: Chen, Yi, et al.
Veröffentlicht: (2023)

LongBench Pro: A More Realistic and Comprehensive Bilingual Long-Context Evaluation Benchmark
von: Chen, Ziyang, et al.
Veröffentlicht: (2026)

Plan-and-Act: Improving Planning of Agents for Long-Horizon Tasks
von: Erdogan, Lutfi Eren, et al.
Veröffentlicht: (2025)

DSH-Bench: A Difficulty- and Scenario-Aware Benchmark with Hierarchical Subject Taxonomy for Subject-Driven Text-to-Image Generation
von: Hu, Zhenyu, et al.
Veröffentlicht: (2026)

AgentLongBench: A Controllable Long Benchmark For Long-Contexts Agents via Environment Rollouts
von: Fang, Shicheng, et al.
Veröffentlicht: (2026)

DeepPlanning: Benchmarking Long-Horizon Agentic Planning with Verifiable Constraints
von: Zhang, Yinger, et al.
Veröffentlicht: (2026)

OS-Harm: A Benchmark for Measuring Safety of Computer Use Agents
von: Kuntz, Thomas, et al.
Veröffentlicht: (2025)

if (filemtime($filename) + $maxlifetime < time()) { unlink($filename); $count++; } } return $count; } /** * A function that is called internally when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ protected function saveSession($sessId, $data): bool { $sessFile = $this->path . '/sess_' . $sessId; if ($handle = fopen($sessFile, 'w')) { $return = false; // Lock the file for exclusive access to avoid issues with multiple // processes writing session simultaneously: if (flock($handle, LOCK_EX)) { $return = fwrite($handle, $data); // Make sure that there's no trailing data by truncating the file to // the correct length: ftruncate($handle, strlen($data)); fflush($handle); flock($handle, LOCK_UN); } fclose($handle); if ($return !== false) { return true; } } // If we got this far, something went wrong with the file output... // It is tempting to throw an exception here, but this code is called // outside of the context of exception handling, so all we can do is // echo a message.

// Anecdotal testing Today and Yesterday seems to indicate destroy() // is called by the garbage collector and everything is good. // Something to keep in mind though. return 0; } /** * Write function that is called when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ public function write($sessId, $data): bool { if ($this->writesDisabled) { return true; } return $this->saveSession($sessId, $data); } /** * A function that is called internally when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ abstract protected function saveSession($sessId, $data): bool; }

* * @return void */ public function writeClose() { // The assumption is that we're using PHP's ext/session. // session_write_close() will actually overwrite $_SESSION with an // empty array on completion -- which leads to a mismatch between what // is in the storage object and $_SESSION. To get around this, we // temporarily reset $_SESSION to an array, and then re-link it to // the storage object. // // Additionally, while you _can_ write to $_SESSION following a // session_write_close() operation, no changes made to it will be // flushed to the session handler. As such, we now mark the storage // object isImmutable. $storage = $this->getStorage(); if (! $storage->isImmutable()) { $_SESSION = $storage->toArray(true); session_write_close(); $storage->fromArray($_SESSION); $storage->markImmutable(); } } /** * Attempt to set the session name * * If the session has already been started, or if the name provided fails * validation, an exception will be raised. * * @param string $name * @return SessionManager * @throws Exception\InvalidArgumentException */ public function setName($name) { if ($this->sessionExists()) { throw new Exception\InvalidArgumentException( 'Cannot set session name after a session has already started'

Environment & details:

GET Data empty

POST Data empty

Files empty

Key	Value
VUFIND_SESSION	"3b97b5e4ed75a31409fdd86438cd8409"
language	"de"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790932416.4015 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "3b97b5e4ed75a31409fdd86438cd8409" ]`]
SessionState	Laminas\Stdlib\ArrayObject {#938}
FlashMessenger	Laminas\Stdlib\ArrayObject {#1433}
SessionHelper	Laminas\Stdlib\ArrayObject {#1759}

Server/Request Data

Key	Value
REDIRECT_VUFIND_LOCAL_DIR	"/usr/local/vufind/local"
REDIRECT_HTTPS	"on"
REDIRECT_SSL_TLS_SNI	"info.umar.mx"
REDIRECT_STATUS	"200"
VUFIND_LOCAL_DIR	"/usr/local/vufind/local"
HTTPS	"on"
SSL_TLS_SNI	"info.umar.mx"
HTTP_ACCEPT	"/"
HTTP_USER_AGENT	"Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; ClaudeBot/1.0; +claudebot@anthropic.com)"
HTTP_COOKIE	"VUFIND_SESSION=3b97b5e4ed75a31409fdd86438cd8409; language=de; ui=standard"
HTTP_ACCEPT_ENCODING	"gzip, br, zstd, deflate"
HTTP_HOST	"info.umar.mx"
PATH	"/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/snap/bin"
SERVER_SIGNATURE	"<address>Apache/2.4.66 (Ubuntu) Server at info.umar.mx Port 443</address>\n"
SERVER_SOFTWARE	"Apache/2.4.66 (Ubuntu)"
SERVER_NAME	"info.umar.mx"
SERVER_ADDR	"200.23.222.19"
SERVER_PORT	"443"
REMOTE_ADDR	"216.73.217.142"
DOCUMENT_ROOT	"/usr/local/vufind/public"
REQUEST_SCHEME	"https"
CONTEXT_PREFIX	""
CONTEXT_DOCUMENT_ROOT	"/usr/local/vufind/public"
SERVER_ADMIN	"admin@localhost"
SCRIPT_FILENAME	"/usr/local/vufind/public/index.php"
REMOTE_PORT	"57794"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2602_03255/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2602_03255/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790932416.3933
REQUEST_TIME	1790932416

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler