Towards Detecting LLMs Hallucination via Markov Chain-based Multi-agent Debate Framework

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Sun, Xiaoxi, Li, Jinpeng, Zhong, Yan, Zhao, Dongyan, Yan, Rui
Formato:	Preprint
Publicado:	2024
Materias:	Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

StyleChat: Learning Recitation-Augmented Memory in LLMs for Stylized Dialogue Generation
por: Li, Jinpeng, et al.
Publicado: (2024)

E-Bench: Towards Evaluating the Ease-of-Use of Large Language Models
por: Zhang, Zhenyu, et al.
Publicado: (2024)

Can LLMs Beat Humans in Debating? A Dynamic Multi-agent Framework for Competitive Debate
por: Zhang, Yiqun, et al.
Publicado: (2024)

Chain-of-Discussion: A Multi-Model Framework for Complex Evidence-Based Question Answering
por: Tao, Mingxu, et al.
Publicado: (2024)

Evidence-Enhanced Triplet Generation Framework for Hallucination Alleviation in Generative Question Answering
por: Du, Haowei, et al.
Publicado: (2024)

Counterfactual Debating with Preset Stances for Hallucination Elimination of LLMs
por: Fang, Yi, et al.
Publicado: (2024)

FinRpt: Dataset, Evaluation System and LLM-based Multi-agent Framework for Equity Research Report Generation
por: Jin, Song, et al.
Publicado: (2025)

Stick to Facts: Towards Fidelity-oriented Product Description Generation
por: Chan, Zhangming, et al.
Publicado: (2025)

CodeHalu: Investigating Code Hallucinations in LLMs via Execution-based Verification
por: Tian, Yuchen, et al.
Publicado: (2024)

Multi-Granularity Information Interaction Framework for Incomplete Utterance Rewriting
por: Du, Haowei, et al.
Publicado: (2023)

Internal and External Knowledge Interactive Refinement Framework for Knowledge-Intensive Question Answering
por: Du, Haowei, et al.
Publicado: (2024)

Removal of Hallucination on Hallucination: Debate-Augmented RAG
por: Hu, Wentao, et al.
Publicado: (2025)

Dialectic-Med: Mitigating Diagnostic Hallucinations via Counterfactual Adversarial Multi-Agent Debate
por: Lu, Zhixiang, et al.
Publicado: (2026)

Learning to Edit Knowledge via Instruction-based Chain-of-Thought Prompting
por: Fu, Jinhu, et al.
Publicado: (2026)

HalluClean: A Unified Framework to Combat Hallucinations in LLMs
por: Zhao, Yaxin, et al.
Publicado: (2025)

MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
por: Li, Zhuo, et al.
Publicado: (2026)

FIRP: Faster LLM inference via future intermediate representation prediction
por: Wu, Pengfei, et al.
Publicado: (2024)

Learning to Shop Like Humans: A Review-driven Retrieval-Augmented Recommendation Framework with LLMs
por: Wei, Kaiwen, et al.
Publicado: (2025)

Medico: Towards Hallucination Detection and Correction with Multi-source Evidence Fusion
por: Zhao, Xinping, et al.
Publicado: (2024)

Parallel Decoding via Hidden Transfer for Lossless Large Language Model Acceleration
por: Wu, Pengfei, et al.
Publicado: (2024)

Maintaining Informative Coherence: Migrating Hallucinations in Large Language Models via Absorbing Markov Chains
por: Wu, Jiemin, et al.
Publicado: (2024)

Can LLMs Detect Their Own Hallucinations?
por: Kadotani, Sora, et al.
Publicado: (2025)

Detecting Contextual Hallucinations in LLMs with Frequency-Aware Attention
por: Qi, Siya, et al.
Publicado: (2026)

Debate-Driven Multi-Agent LLMs for Phishing Email Detection
por: Nguyen, Ngoc Tuong Vy, et al.
Publicado: (2025)

Source or It Didn't Happen: A Multi-Agent Framework for Citation Hallucination Detection
por: Li, Mingzhe, et al.
Publicado: (2026)

Can LLMs Produce Faithful Explanations For Fact-checking? Towards Faithful Explainable Fact-Checking via Multi-Agent Debate
por: Kim, Kyungha, et al.
Publicado: (2024)

Generating Effective CoT Traces for Mitigating Causal Hallucination
por: Zhao, Yiheng, et al.
Publicado: (2026)

A Debate-Driven Experiment on LLM Hallucinations and Accuracy
por: Li, Ray, et al.
Publicado: (2024)

Multi-Agent Debate with Memory Masking
por: Tian, Hongduan, et al.
Publicado: (2026)

Phonetic Reconstruction of the Consonant System of Middle Chinese via Mixed Integer Optimization
por: Luo, Xiaoxi, et al.
Publicado: (2025)

Geometric Uncertainty for Detecting and Correcting Hallucinations in LLMs
por: Phillips, Edward, et al.
Publicado: (2025)

Do LLMs Know about Hallucination? An Empirical Investigation of LLM's Hidden States
por: Duan, Hanyu, et al.
Publicado: (2024)

Attention with Markov: A Framework for Principled Analysis of Transformers via Markov Chains
por: Makkuva, Ashok Vardhan, et al.
Publicado: (2024)

Self-Compression of Chain-of-Thought via Multi-Agent Reinforcement Learning
por: Chen, Yiqun, et al.
Publicado: (2026)

LANCET: Neural Intervention via Structural Entropy for Mitigating Faithfulness Hallucinations in LLMs
por: Wang, Chenxu, et al.
Publicado: (2026)

Is LLMs Hallucination Usable? LLM-based Negative Reasoning for Fake News Detection
por: Zhang, Chaowei, et al.
Publicado: (2025)

Detecting Hallucinations in Retrieval-Augmented Generation via Semantic-level Internal Reasoning Graph
por: Hu, Jianpeng, et al.
Publicado: (2026)

RAM-SD: Retrieval-Augmented Multi-agent framework for Sarcasm Detection
por: Zhou, Ziyang, et al.
Publicado: (2026)

Hallucination Detection with the Internal Layers of LLMs
por: Preiß, Martin
Publicado: (2025)

In-Context Learning with Reinforcement Learning for Incomplete Utterance Rewriting
por: Du, Haowei, et al.
Publicado: (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	"es"
VUFIND_SESSION	"b53430b9c79e04b721d2e97ef3b6a924"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790605262.2407 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "b53430b9c79e04b721d2e97ef3b6a924" ]`]
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=es; VUFIND_SESSION=b53430b9c79e04b721d2e97ef3b6a924; 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	"216.73.216.147"
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	"5961"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2406_03075/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2406_03075/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790605262.2362
REQUEST_TIME	1790605262

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler