Enhanced Language Model Truthfulness with Learnable Intervention and Uncertainty Expression

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Bayat, Farima Fatahi, Liu, Xin, Jagadish, H. V., Wang, Lu
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Enhancing Language Model Factuality via Activation-Based Confidence Calibration and Guided Decoding
di: Liu, Xin, et al.
Pubblicazione: (2024)

FactBench: A Dynamic Benchmark for In-the-Wild Language Model Factuality Evaluation
di: Bayat, Farima Fatahi, et al.
Pubblicazione: (2024)

From Proof to Program: Characterizing Tool-Induced Reasoning Hallucinations in Large Language Models
di: Bayat, Farima Fatahi, et al.
Pubblicazione: (2025)

Logit Arithmetic Elicits Long Reasoning Capabilities Without Training
di: Zhang, Yunxiang, et al.
Pubblicazione: (2025)

Unconditional Truthfulness: Learning Unconditional Uncertainty of Large Language Models
di: Vazhentsev, Artem, et al.
Pubblicazione: (2024)

Truth Forest: Toward Multi-Scale Truthfulness in Large Language Models through Intervention without Tuning
di: Chen, Zhongzhi, et al.
Pubblicazione: (2023)

APE: Active Learning-based Tooling for Finding Informative Few-shot Examples for LLM-based Entity Matching
di: Qian, Kun, et al.
Pubblicazione: (2024)

Towards Reliable Truth-Aligned Uncertainty Estimation in Large Language Models
di: Srey, Ponhvoan, et al.
Pubblicazione: (2026)

Token-Level Density-Based Uncertainty Quantification Methods for Eliciting Truthfulness of Large Language Models
di: Vazhentsev, Artem, et al.
Pubblicazione: (2025)

Finetuning Language Models to Emit Linguistic Expressions of Uncertainty
di: Chaudhry, Arslan, et al.
Pubblicazione: (2024)

Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
di: Li, Kenneth, et al.
Pubblicazione: (2023)

TruthPrInt: Mitigating Large Vision-Language Models Object Hallucination Via Latent Truthful-Guided Pre-Intervention
di: Duan, Jinhao, et al.
Pubblicazione: (2025)

ARREST: Adversarial Resilient Regulation Enhancing Safety and Truth in Large Language Models
di: Dasgupta, Sharanya, et al.
Pubblicazione: (2026)

Think Through Uncertainty: Improving Long-Form Generation Factuality via Reasoning Calibration
di: Liu, Xin, et al.
Pubblicazione: (2026)

Offline Training of Language Model Agents with Functions as Learnable Weights
di: Zhang, Shaokun, et al.
Pubblicazione: (2024)

DNACHUNKER: Learnable Tokenization for DNA Language Models
di: Kim, Taewon, et al.
Pubblicazione: (2026)

Enhancing Confidence Expression in Large Language Models Through Learning from Past Experience
di: Han, Haixia, et al.
Pubblicazione: (2024)

Emergence of Linear Truth Encodings in Language Models
di: Ravfogel, Shauli, et al.
Pubblicazione: (2025)

On the Learnability of Watermarks for Language Models
di: Gu, Chenchen, et al.
Pubblicazione: (2023)

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

MetaFaith: Faithful Natural Language Uncertainty Expression in LLMs
di: Liu, Gabrielle Kaili-May, et al.
Pubblicazione: (2025)

Syntactic Learnability of Echo State Neural Language Models at Scale
di: Ueda, Ryo, et al.
Pubblicazione: (2025)

Representational and Behavioral Stability of Truth in Large Language Models
di: Dies, Samantha, et al.
Pubblicazione: (2025)

When Truth Is Overridden: Uncovering the Internal Origins of Sycophancy in Large Language Models
di: Wang, Keyu, et al.
Pubblicazione: (2025)

Truth Knows No Language: Evaluating Truthfulness Beyond English
di: Figueras, Blanca Calvo, et al.
Pubblicazione: (2025)

ToolMem: Enhancing Multimodal Agents with Learnable Tool Capability Memory
di: Xiao, Yunzhong, et al.
Pubblicazione: (2025)

From Yes-Men to Truth-Tellers: Addressing Sycophancy in Large Language Models with Pinpoint Tuning
di: Chen, Wei, et al.
Pubblicazione: (2024)

Non-Linear Inference Time Intervention: Improving LLM Truthfulness
di: Hoscilowicz, Jakub, et al.
Pubblicazione: (2024)

Learnable Privacy Neurons Localization in Language Models
di: Chen, Ruizhe, et al.
Pubblicazione: (2024)

KatotohananQA: Evaluating Truthfulness of Large Language Models in Filipino
di: Nery, Lorenzo Alfred, et al.
Pubblicazione: (2025)

Self-Evolutionary Large Language Models through Uncertainty-Enhanced Preference Optimization
di: Wang, Jianing, et al.
Pubblicazione: (2024)

LitCab: Lightweight Language Model Calibration over Short- and Long-form Responses
di: Liu, Xin, et al.
Pubblicazione: (2023)

Learnable Assessment Skills for LLM-based Automated Scoring: Rubric Construction via Iterative Optimization
di: Wang, Yun, et al.
Pubblicazione: (2026)

Enhancing Fact Retrieval in PLMs through Truthfulness
di: Youssef, Paul, et al.
Pubblicazione: (2024)

Uncertainty is Fragile: Manipulating Uncertainty in Large Language Models
di: Zeng, Qingcheng, et al.
Pubblicazione: (2024)

Entropy2Vec: Crosslingual Language Modeling Entropy as End-to-End Learnable Language Representations
di: Irawan, Patrick Amadeus, et al.
Pubblicazione: (2025)

Structure-Learnable Adapter Fine-Tuning for Parameter-Efficient Large Language Models
di: Gong, Ming, et al.
Pubblicazione: (2025)

Query-Aware Learnable Graph Pooling Tokens as Prompt for Large Language Models
di: Kim, Wooyoung, et al.
Pubblicazione: (2025)

Characterizing Truthfulness in Large Language Model Generations with Local Intrinsic Dimension
di: Yin, Fan, et al.
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
VUFIND_SESSION	"74cbe74f4ffa95a2e096cd3c146b3f49"
language	"it"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790867577.5255 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "74cbe74f4ffa95a2e096cd3c146b3f49" ]`]
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	"VUFIND_SESSION=74cbe74f4ffa95a2e096cd3c146b3f49; language=it; 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	"43386"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2405_00301/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2405_00301/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790867577.5194
REQUEST_TIME	1790867577

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler