DIF: A Framework for Benchmarking and Verifying Implicit Bias in LLMs

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Yin, Lake, Huang, Fan
Format:	Preprint
Publié:	2025
Sujets:	Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Implicit Bias in LLMs: A Survey
par: Lin, Xinru, et autres
Publié: (2025)

MALIBU Benchmark: Multi-Agent LLM Implicit Bias Uncovered
par: Mirza, Imran, et autres
Publié: (2025)

LLMEval-Logic: A Solver-Verified Chinese Benchmark for Logical Reasoning of LLMs with Adversarial Hardening
par: Zhang, Ming, et autres
Publié: (2026)

Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMs
par: Gupta, Shashank, et autres
Publié: (2023)

Can LLMs Grasp Implicit Cultural Values? Benchmarking LLMs' Cultural Intelligence with CQ-Bench
par: Liu, Ziyi, et autres
Publié: (2025)

Reinforcement Learning with Verifiable Rewards Implicitly Incentivizes Correct Reasoning in Base LLMs
par: Wen, Xumeng, et autres
Publié: (2025)

MIST: Towards Multi-dimensional Implicit BiaS Evaluation of LLMs for Theory of Mind
par: Li, Yanlin, et autres
Publié: (2025)

Finding Words Associated with DIF: Predicting Differential Item Functioning using LLMs and Explainable AI
par: Maeda, Hotaka, et autres
Publié: (2025)

ImplicitBBQ: Benchmarking Implicit Bias in Large Language Models through Characteristic Based Cues
par: Vedula, Bhaskara Hanuma, et autres
Publié: (2026)

BiasAlert: A Plug-and-play Tool for Social Bias Detection in LLMs
par: Fan, Zhiting, et autres
Publié: (2024)

Obscured but Not Erased: Evaluating Nationality Bias in LLMs via Name-Based Bias Benchmarks
par: Pelosio, Giulio, et autres
Publié: (2025)

eDIF: A European Deep Inference Fabric for Remote Interpretability of LLM
par: Guggenberger, Irma Heithoff. Marc, et autres
Publié: (2025)

PRISM: A Unified Framework for Post-Training LLMs Without Verifiable Rewards
par: Ghimire, Mukesh, et autres
Publié: (2026)

Business as Rulesual: A Benchmark and Framework for Business Rule Flow Modeling with LLMs
par: Yang, Chen, et autres
Publié: (2025)

Relative Bias: A Comparative Framework for Quantifying Bias in LLMs
par: Arbabi, Alireza, et autres
Publié: (2025)

Are LLMs Rational Investors? A Study on Detecting and Reducing the Financial Bias in LLMs
par: Zhou, Yuhang, et autres
Publié: (2024)

'Since Lawyers are Males..': Examining Implicit Gender Bias in Hindi Language Generation by LLMs
par: Joshi, Ishika, et autres
Publié: (2024)

Investigating Implicit Bias in Large Language Models: A Large-Scale Study of Over 50 LLMs
par: Kumar, Divyanshu, et autres
Publié: (2024)

M2-Verify: A Large-Scale Multidomain Benchmark for Checking Multimodal Claim Consistency
par: Ansari, Abolfazl, et autres
Publié: (2026)

IndoBias: A Dual Track Culturally Grounded Benchmark for LLMs Bias Evaluation in Indonesian Languages
par: Hanif, Ikhlasul Akmal, et autres
Publié: (2026)

Culturally-Aware Conversations: A Framework & Benchmark for LLMs
par: Havaldar, Shreya, et autres
Publié: (2025)

User-Assistant Bias in LLMs
par: Pan, Xu, et autres
Publié: (2025)

ToolGate: Contract-Grounded and Verified Tool Execution for LLMs
par: Liu, Yanming, et autres
Publié: (2026)

A Scalable Entity-Based Framework for Auditing Bias in LLMs
par: Elbouanani, Akram, et autres
Publié: (2026)

The Bias is in the Details: An Assessment of Cognitive Bias in LLMs
par: Knipper, R. Alexander, et autres
Publié: (2025)

When Sharpening Becomes Collapse: Sampling Bias and Semantic Coupling in RL with Verifiable Rewards
par: Fan, Mingyuan, et autres
Publié: (2026)

Evaluating Gender Bias of LLMs in Making Morality Judgements
par: Bajaj, Divij, et autres
Publié: (2024)

When LLMs Benchmark Themselves: Deconstructing Self-Bias in Automated Evaluation
par: Xu, Wenda, et autres
Publié: (2025)

CompassVerifier: A Unified and Robust Verifier for LLMs Evaluation and Outcome Reward
par: Liu, Shudong, et autres
Publié: (2025)

Evaluating LLMs for Demographic-Targeted Social Bias Detection: A Comprehensive Benchmark Study
par: Majumdar, Ayan, et autres
Publié: (2025)

Measuring Implicit Bias in Explicitly Unbiased Large Language Models
par: Bai, Xuechunzi, et autres
Publié: (2024)

MENTOR: A Metacognition-Driven Self-Evolution Framework for Uncovering and Mitigating Implicit Domain Risks in LLMs
par: Shan, Liang, et autres
Publié: (2025)

PRIME: A Process-Outcome Alignment Benchmark for Verifiable Reasoning in Mathematics and Engineering
par: Wang, Xiangfeng, et autres
Publié: (2026)

MixRea: Benchmarking Explicit-Implicit Reasoning in Large Language Models
par: Cai, Yuanqing, et autres
Publié: (2026)

VisBias: Measuring Explicit and Implicit Social Biases in Vision Language Models
par: Huang, Jen-tse, et autres
Publié: (2025)

How Transformers Learn Regular Language Recognition: A Theoretical Study on Training Dynamics and Implicit Bias
par: Huang, Ruiquan, et autres
Publié: (2025)

Lessons from Training Grounded LLMs with Verifiable Rewards
par: Sim, Shang Hong, et autres
Publié: (2025)

Benchmark Self-Evolving: A Multi-Agent Framework for Dynamic LLM Evaluation
par: Wang, Siyuan, et autres
Publié: (2024)

I Am Aligned, But With Whom? MENA Values Benchmark for Evaluating Cultural Alignment and Multilingual Bias in LLMs
par: Zahraei, Pardis Sadat, et autres
Publié: (2025)

Towards Implicit Bias Detection and Mitigation in Multi-Agent LLM Interactions
par: Borah, Angana, et autres
Publié: (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	"fr"
VUFIND_SESSION	"dba192c8b2d61f8422362782f9dd11f2"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790601072.9572 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "dba192c8b2d61f8422362782f9dd11f2" ]`]
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=fr; VUFIND_SESSION=dba192c8b2d61f8422362782f9dd11f2; 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	"32500"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2505_10013/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2505_10013/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790601072.9527
REQUEST_TIME	1790601072

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler