Meaningless is better: hashing bias-inducing words in LLM prompts improves performance in logical reasoning and statistical learning

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Chadimová, Milena, Jurášek, Eduard, Kliegr, Tomáš
Format:	Preprint
Publié:	2024
Sujets:	Computation and Language Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

The Outputs of Large Language Models are Meaningless
par: Hattiangadi, Anandi, et autres
Publié: (2025)

CoT-Self-Instruct: Building high-quality synthetic prompts for reasoning and non-reasoning tasks
par: Yu, Ping, et autres
Publié: (2025)

Fooling LLM graders into giving better grades through neural activity guided adversarial prompting
par: Yamamura, Atsushi, et autres
Publié: (2024)

Is continuous CoT better suited for multi-lingual reasoning?
par: Bashir, Ali Hamza, et autres
Publié: (2026)

Language hooks: a modular framework for augmenting LLM reasoning that decouples tool usage from the model and its prompt
par: de Mijolla, Damien, et autres
Publié: (2024)

Traceable LLM-based validation of statements in knowledge graphs
par: Adam, Daniel, et autres
Publié: (2024)

Can LLMs perform structured graph reasoning?
par: Agrawal, Palaash, et autres
Publié: (2024)

A word association network methodology for evaluating implicit biases in LLMs compared to humans
par: Abramski, Katherine, et autres
Publié: (2025)

Can formal argumentative reasoning enhance LLMs performances?
par: Castagna, Federico, et autres
Publié: (2024)

Certainty robustness: Evaluating LLM stability under self-challenging prompts
par: Saadat, Mohammadreza, et autres
Publié: (2026)

Alignment-Weighted DPO: A principled reasoning approach to improve safety alignment
par: Hu, Mengxuan, et autres
Publié: (2026)

Critical-Questions-of-Thought: Steering LLM reasoning with Argumentative Querying
par: Castagna, Federico, et autres
Publié: (2024)

Superhuman performance of a large language model on the reasoning tasks of a physician
par: Brodeur, Peter G., et autres
Publié: (2024)

LLM-based feature generation from text for interpretable machine learning
par: Balek, Vojtěch, et autres
Publié: (2024)

MARS: toward more efficient multi-agent collaboration for LLM reasoning
par: Wang, Xiao, et autres
Publié: (2025)

Reinforcement learning fine-tuning of language model for instruction following and math reasoning
par: Han, Yifu, et autres
Publié: (2025)

Scaling few-shot spoken word classification with generative meta-continual learning
par: Beyers, Louise, et autres
Publié: (2026)

Prompt-prompted Adaptive Structured Pruning for Efficient LLM Generation
par: Dong, Harry, et autres
Publié: (2024)

DeepSeek performs better than other Large Language Models in Dental Cases
par: Zhang, Hexian, et autres
Publié: (2025)

Auto prompting without training labels: An LLM cascade for product quality assessment in e-commerce catalogs
par: Satyadharma, Soham, et autres
Publié: (2025)

Retrieval-augmented reasoning with lean language models
par: Chan, Ryan Sze-Yin, et autres
Publié: (2025)

Temperature-scaling surprisal estimates improve fit to human reading times -- but does it do so for the "right reasons"?
par: Liu, Tong, et autres
Publié: (2023)

Does language matter for spoken word classification? A multilingual generative meta-learning approach
par: Ziki, Batsirayi Mupamhi, et autres
Publié: (2026)

Towards better Human-Agent Alignment: Assessing Task Utility in LLM-Powered Applications
par: Arabzadeh, Negar, et autres
Publié: (2024)

Cognitive bias in LLM reasoning compromises interpretation of clinical oncology notes
par: Kenaston, Matthew W., et autres
Publié: (2025)

How word semantics and phonology affect handwriting of Alzheimer's patients: a machine learning based analysis
par: Cilia, Nicole Dalia, et autres
Publié: (2023)

Child vs. machine language learning: Can the logical structure of human language unleash LLMs?
par: Sauerland, Uli, et autres
Publié: (2025)

Multi-step retrieval and reasoning improves radiology question answering with large language models
par: Wind, Sebastian, et autres
Publié: (2025)

Knowledge prompt chaining for semantic modeling
par: Ding, Ning Pei, et autres
Publié: (2025)

Fine-Grained Bias Detection in LLM: Enhancing detection mechanisms for nuanced biases
par: Mohanty, Suvendu
Publié: (2025)

IoT-LLM: a framework for enhancing Large Language Model reasoning from real-world sensor data
par: An, Tuo, et autres
Publié: (2024)

StateAct: Enhancing LLM Base Agents via Self-prompting and State-tracking
par: Rozanov, Nikolai, et autres
Publié: (2024)

Phi-4-reasoning Technical Report
par: Abdin, Marah, et autres
Publié: (2025)

MOSLIM:Align with diverse preferences in prompts through reward classification
par: Zhang, Yu, et autres
Publié: (2025)

Preference learning in shades of gray: Interpretable and bias-aware reward modeling for human preferences
par: Oprea, Simona-Vasilica, et autres
Publié: (2026)

Mitigating LLM biases toward spurious social contexts using direct preference optimization
par: Nam, Hyunji, et autres
Publié: (2026)

Is your LLM trapped in a Mental Set? Investigative study on how mental sets affect the reasoning capabilities of LLMs
par: Haq, Saiful, et autres
Publié: (2025)

Explaining word embeddings with perfect fidelity: Case study in research impact prediction
par: Dvorackova, Lucie, et autres
Publié: (2024)

Temporal reasoning for timeline summarisation in social media
par: Song, Jiayu, et autres
Publié: (2024)

Mis-prompt: Benchmarking Large Language Models for Proactive Error Handling
par: Zeng, Jiayi, et autres
Publié: (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
language	"fr"
VUFIND_SESSION	"f10c210fd8ec5bb23b168695dd64cef6"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790894858.7858 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "f10c210fd8ec5bb23b168695dd64cef6" ]`]
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=f10c210fd8ec5bb23b168695dd64cef6; 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	"35265"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2411_17304/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2411_17304/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790894858.7794
REQUEST_TIME	1790894858

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler