Deceive, Detect, and Disclose: Large Language Models Play Mini-Mafia

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Costa, Davi Bastos, Vicente, Renato
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Moral Susceptibility and Robustness under Persona Role-Play in Large Language Models
di: Costa, Davi Bastos, et al.
Pubblicazione: (2025)

Persona-Model Collapse in Emergent Misalignment
di: Costa, Davi Bastos, et al.
Pubblicazione: (2026)

Flattering to Deceive: The Impact of Sycophantic Behavior on User Trust in Large Language Model
di: Carro, María Victoria
Pubblicazione: (2024)

Are Your Agents Upward Deceivers?
di: Guo, Dadi, et al.
Pubblicazione: (2025)

Large Language Models can Strategically Deceive their Users when Put Under Pressure
di: Scheurer, Jérémy, et al.
Pubblicazione: (2023)

HoneyTrap: Deceiving Large Language Model Attackers to Honeypot Traps with Resilient Multi-Agent Defense
di: Li, Siyuan, et al.
Pubblicazione: (2026)

Seeing is Deceiving: Exploitation of Visual Pathways in Multi-Modal Language Models
di: Janowczyk, Pete, et al.
Pubblicazione: (2024)

Adversarial Magnification to Deceive Deepfake Detection through Super Resolution
di: Coccomini, Davide Alessandro, et al.
Pubblicazione: (2024)

MiniLLM: On-Policy Distillation of Large Language Models
di: Gu, Yuxian, et al.
Pubblicazione: (2023)

Role-Playing Evaluation for Large Language Models
di: Boudouri, Yassine El, et al.
Pubblicazione: (2025)

MiniGPT-Pancreas: Multimodal Large Language Model for Pancreas Cancer Classification and Detection
di: Moglia, Andrea, et al.
Pubblicazione: (2024)

Can Large Language Models Play Games? A Case Study of A Self-Play Approach
di: Guo, Hongyi, et al.
Pubblicazione: (2024)

Fairness Testing of Large Language Models in Role-Playing
di: Li, Xinyue, et al.
Pubblicazione: (2024)

Super(ficial)-alignment: Strong Models May Deceive Weak Models in Weak-to-Strong Generalization
di: Yang, Wenkai, et al.
Pubblicazione: (2024)

NetDiffuser: Deceiving DNN-Based Network Attack Detection Systems with Diffusion-Generated Adversarial Traffic
di: Kumar, Pratyay, et al.
Pubblicazione: (2026)

Knowledge Graph-enhanced Large Language Model for Incremental Game PlayTesting
di: Mu, Enhong, et al.
Pubblicazione: (2025)

How to Disclose? Strategic AI Disclosure in Crowdfunding
di: Wang, Ning, et al.
Pubblicazione: (2026)

Stop Comparing LLM Agents Without Disclosing the Harness
di: Zhang, Yunbei, et al.
Pubblicazione: (2026)

RSPO: Regularized Self-Play Alignment of Large Language Models
di: Tang, Xiaohang, et al.
Pubblicazione: (2025)

From Code to Play: Benchmarking Program Search for Games Using Large Language Models
di: Eberhardinger, Manuel, et al.
Pubblicazione: (2024)

Deceiving Question-Answering Models: A Hybrid Word-Level Adversarial Approach
di: Li, Jiyao, et al.
Pubblicazione: (2024)

TRAPDOC: Deceiving LLM Users by Injecting Imperceptible Phantom Tokens into Documents
di: Jin, Hyundong, et al.
Pubblicazione: (2025)

RPGBENCH: Evaluating Large Language Models as Role-Playing Game Engines
di: Yu, Pengfei, et al.
Pubblicazione: (2025)

On the Decision-Making Abilities in Role-Playing using Large Language Models
di: Shen, Chenglei, et al.
Pubblicazione: (2024)

Time to Talk: LLM Agents for Asynchronous Group Communication in Mafia Games
di: Eckhaus, Niv, et al.
Pubblicazione: (2025)

Large Language Models Play StarCraft II: Benchmarks and A Chain of Summarization Approach
di: Ma, Weiyu, et al.
Pubblicazione: (2023)

MiniCache: KV Cache Compression in Depth Dimension for Large Language Models
di: Liu, Akide, et al.
Pubblicazione: (2024)

Quantifying Risk Propensities of Large Language Models: Ethical Focus and Bias Detection through Role-Play
di: Zeng, Yifan, et al.
Pubblicazione: (2024)

Credence Calibration Game? Calibrating Large Language Models through Structured Play
di: Fang, Ke, et al.
Pubblicazione: (2025)

Large Language Models as Pokémon Battle Agents: Strategic Play and Content Generation
di: Jain, Daksh, et al.
Pubblicazione: (2025)

Memory Decoder: A Pretrained, Plug-and-Play Memory for Large Language Models
di: Cao, Jiaqi, et al.
Pubblicazione: (2025)

Plug-and-Play Policy Planner for Large Language Model Powered Dialogue Agents
di: Deng, Yang, et al.
Pubblicazione: (2023)

PersonaArena: Dynamic Simulation for Evaluating and Enhancing Persona-Level Role-Playing in Large Language Models
di: Shi, Wenlong, et al.
Pubblicazione: (2026)

What if Othello-Playing Language Models Could See?
di: Chen, Xinyi, et al.
Pubblicazione: (2025)

Deceptive Risk Minimization: Out-of-Distribution Generalization by Deceiving Distribution Shift Detectors
di: Majumdar, Anirudha
Pubblicazione: (2025)

Don't Deceive Me: Mitigating Gaslighting through Attention Reallocation in LMMs
di: Jiao, Pengkun, et al.
Pubblicazione: (2025)

SeRL: Self-Play Reinforcement Learning for Large Language Models with Limited Data
di: Fang, Wenkai, et al.
Pubblicazione: (2025)

Orca: Enhancing Role-Playing Abilities of Large Language Models by Integrating Personality Traits
di: Huang, Yuxuan
Pubblicazione: (2024)

RoleLLM: Benchmarking, Eliciting, and Enhancing Role-Playing Abilities of Large Language Models
di: Wang, Zekun Moore, et al.
Pubblicazione: (2023)

Large Language Models Playing Mixed Strategy Nash Equilibrium Games
di: Silva, Alonso
Pubblicazione: (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	"it"
VUFIND_SESSION	"f1bafde2e46560047ad120df861fd063"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790897291.9515 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "f1bafde2e46560047ad120df861fd063" ]`]
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=it; VUFIND_SESSION=f1bafde2e46560047ad120df861fd063; 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	"37760"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2509_23023/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2509_23023/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790897291.9472
REQUEST_TIME	1790897291

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler