Deception in LLMs: Self-Preservation and Autonomous Goals in Large Language Models

Fuente: arXiv

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Barkur, Sudarshan Kamath, Schacht, Sigurd, Scholl, Johannes
Format:	Preprint
Veröffentlicht:	2025
Schlagworte:	Computation and Language
Online-Zugang:	Acceder al recurso
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Ähnliche Einträge

PHOENIX: Open-Source Language Adaption for Direct Preference Optimization
von: Uhlig, Matthias, et al.
Veröffentlicht: (2024)

Inference Optimizations for Large Language Models: Effects, Challenges, and Practical Considerations
von: Donisch, Leo, et al.
Veröffentlicht: (2024)

A Review of Common Online Speaker Diarization Methods
von: Aperdannier, Roman, et al.
Veröffentlicht: (2024)

Systematic Evaluation of Online Speaker Diarization Systems Regarding their Latency
von: Aperdannier, Roman, et al.
Veröffentlicht: (2024)

An approach to optimize inference of the DIART speaker diarization pipeline
von: Aperdannier, Roman, et al.
Veröffentlicht: (2024)

Do Large Language Models Exhibit Spontaneous Rational Deception?
von: Taylor, Samuel M., et al.
Veröffentlicht: (2025)

Deception Abilities Emerged in Large Language Models
von: Hagendorff, Thilo
Veröffentlicht: (2023)

Scope Ambiguities in Large Language Models
von: Kamath, Gaurav, et al.
Veröffentlicht: (2024)

D-REX: A Benchmark for Detecting Deceptive Reasoning in Large Language Models
von: Krishna, Satyapriya, et al.
Veröffentlicht: (2025)

Self-Steering Optimization: Autonomous Preference Optimization for Large Language Models
von: Xiang, Hao, et al.
Veröffentlicht: (2024)

eDIF: A European Deep Inference Fabric for Remote Interpretability of LLM
von: Guggenberger, Irma Heithoff. Marc, et al.
Veröffentlicht: (2025)

Unmasking the Shadows of AI: Investigating Deceptive Capabilities in Large Language Models
von: Guo, Linge
Veröffentlicht: (2024)

Language Models Largely Exhibit Human-like Constituent Ordering Preferences
von: Tur, Ada Defne, et al.
Veröffentlicht: (2025)

To Tell The Truth: Language of Deception and Language Models
von: Hazra, Sanchaita, et al.
Veröffentlicht: (2023)

Preserving Knowledge in Large Language Model with Model-Agnostic Self-Decompression
von: Zhang, Zilun, et al.
Veröffentlicht: (2024)

From Deception to Detection: The Dual Roles of Large Language Models in Fake News
von: Sallami, Dorsaf, et al.
Veröffentlicht: (2024)

Creating a Fine Grained Entity Type Taxonomy Using LLMs
von: Gunn, Michael, et al.
Veröffentlicht: (2024)

Evaluating the Goal-Directedness of Large Language Models
von: Everitt, Tom, et al.
Veröffentlicht: (2025)

Advancing NLP Security by Leveraging LLMs as Adversarial Engines
von: Srinivasan, Sudarshan, et al.
Veröffentlicht: (2024)

Defending Large Language Models Against Jailbreaking Attacks Through Goal Prioritization
von: Zhang, Zhexin, et al.
Veröffentlicht: (2023)

Goal Hijacking Attack on Large Language Models via Pseudo-Conversation Injection
von: Chen, Zheng, et al.
Veröffentlicht: (2024)

Does Synthetic Data Help Named Entity Recognition for Low-Resource Languages?
von: Kamath, Gaurav, et al.
Veröffentlicht: (2025)

Unmasking Deceptive Visuals: Benchmarking Multimodal Large Language Models on Misleading Chart Question Answering
von: Chen, Zixin, et al.
Veröffentlicht: (2025)

Application of Multimodal Large Language Models in Autonomous Driving
von: Islam, Md Robiul
Veröffentlicht: (2024)

Context-Preserving Tensorial Reconfiguration in Large Language Model Training
von: Tonix, Larin, et al.
Veröffentlicht: (2025)

Self-Prompt Tuning: Enable Autonomous Role-Playing in LLMs
von: Kong, Aobo, et al.
Veröffentlicht: (2024)

Compromising Honesty and Harmlessness in Language Models via Deception Attacks
von: Vaugrante, Laurène, et al.
Veröffentlicht: (2025)

The Perils of Chart Deception: How Misleading Visualizations Affect Vision-Language Models
von: Mahbub, Ridwan, et al.
Veröffentlicht: (2025)

An Assessment of Model-On-Model Deception
von: Heitkoetter, Julius, et al.
Veröffentlicht: (2024)

Hidden in Plain Sight: Evaluation of the Deception Detection Capabilities of LLMs in Multimodal Settings
von: Miah, Md Messal Monem, et al.
Veröffentlicht: (2025)

Plan-Grounded Large Language Models for Dual Goal Conversational Settings
von: Glória-Silva, Diogo, et al.
Veröffentlicht: (2024)

Seamless Deception: Larger Language Models Are Better Knowledge Concealers
von: Ashok, Dhananjay, et al.
Veröffentlicht: (2026)

SelfGoal: Your Language Agents Already Know How to Achieve High-level Goals
von: Yang, Ruihan, et al.
Veröffentlicht: (2024)

Why is "Chicago" Predictive of Deceptive Reviews? Using LLMs to Discover Language Phenomena from Lexical Cues
von: Qu, Jiaming, et al.
Veröffentlicht: (2025)

Robust Utility-Preserving Text Anonymization Based on Large Language Models
von: Yang, Tianyu, et al.
Veröffentlicht: (2024)

Privacy-Preserving Instructions for Aligning Large Language Models
von: Yu, Da, et al.
Veröffentlicht: (2024)

Too Big to Fool: Resisting Deception in Language Models
von: Samsami, Mohammad Reza, et al.
Veröffentlicht: (2024)

Can Large Language Model Summarizers Adapt to Diverse Scientific Communication Goals?
von: Fonseca, Marcio, et al.
Veröffentlicht: (2024)

Towards Goal-oriented Prompt Engineering for Large Language Models: A Survey
von: Li, Haochen, et al.
Veröffentlicht: (2024)

Large Language Models Reveal Information Operation Goals, Tactics, and Narrative Frames
von: Burghardt, Keith, et al.
Veröffentlicht: (2024)

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
language	"de"
VUFIND_SESSION	"983543fdd983936aa04810a886411578"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790983650.2102 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "983543fdd983936aa04810a886411578" ]`]
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	"language=de; VUFIND_SESSION=983543fdd983936aa04810a886411578; 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	"7398"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2501_16513/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2501_16513/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790983650.206
REQUEST_TIME	1790983650

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler