LF-Steering: Latent Feature Activation Steering for Enhancing Semantic Consistency in Large Language Models

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Yang, Jingyuan, Li, Rongjun, Wang, Weixuan, Zhou, Ziyu, Feng, Zhiyong, Peng, Wei
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Semantics-Adaptive Activation Intervention for LLMs via Dynamic Steering Vectors
di: Wang, Weixuan, et al.
Pubblicazione: (2024)

Enhancing Semantic Consistency of Large Language Models through Model Editing: An Interpretability-Oriented Approach
di: Yang, Jingyuan, et al.
Pubblicazione: (2025)

Gradient Co-occurrence Analysis for Detecting Unsafe Prompts in Large Language Models
di: Yang, Jingyuan, et al.
Pubblicazione: (2025)

Feature Extraction and Steering for Enhanced Chain-of-Thought Reasoning in Language Models
di: Li, Zihao, et al.
Pubblicazione: (2025)

Enhancing Cross-task Transfer of Large Language Models via Activation Steering
di: Tang, Xinyu, et al.
Pubblicazione: (2025)

Fine-Grained Activation Steering: Steering Less, Achieving More
di: Feng, Zijian, et al.
Pubblicazione: (2026)

Interpretable Steering of Large Language Models with Feature Guided Activation Additions
di: Soo, Samuel, et al.
Pubblicazione: (2025)

Controlling Large Language Model Agents with Entropic Activation Steering
di: Rahn, Nate, et al.
Pubblicazione: (2024)

Beyond Static Personas: Situational Personality Steering for Large Language Models
di: Wei, Zesheng, et al.
Pubblicazione: (2026)

Steer2Edit: From Activation Steering to Component-Level Editing
di: Sun, Chung-En, et al.
Pubblicazione: (2026)

Exploring the Personality Traits of LLMs through Latent Features Steering
di: Yang, Shu, et al.
Pubblicazione: (2024)

Steering Language Models With Activation Engineering
di: Turner, Alexander Matt, et al.
Pubblicazione: (2023)

Activation Steering for Masked Diffusion Language Models
di: Shnaidman, Adi, et al.
Pubblicazione: (2025)

Contextual Linear Activation Steering of Language Models
di: Hsu, Brandon, et al.
Pubblicazione: (2026)

On Effects of Steering Latent Representation for Large Language Model Unlearning
di: Huu-Tien, Dang, et al.
Pubblicazione: (2024)

ExpertSteer: Intervening in LLMs through Expert Knowledge
di: Wang, Weixuan, et al.
Pubblicazione: (2025)

ASRU: Activation Steering Meets Reinforcement Unlearning for Multimodal Large Language Models
di: Guang, Jiahui, et al.
Pubblicazione: (2026)

Steering When Necessary: Flexible Steering Large Language Models with Backtracking
di: Cheng, Zifeng, et al.
Pubblicazione: (2025)

Compositional Steering of Large Language Models with Steering Tokens
di: Radevski, Gorjan, et al.
Pubblicazione: (2026)

Evaluating the Effectiveness of Black-Box Prompt Optimization as the Scale of LLMs Continues to Grow
di: Zhou, Ziyu, et al.
Pubblicazione: (2025)

Analyze Feature Flow to Enhance Interpretation and Steering in Language Models
di: Laptev, Daniil, et al.
Pubblicazione: (2025)

Psychological Steering of Large Language Models
di: Blas, Leonardo, et al.
Pubblicazione: (2026)

Activation Scaling for Steering and Interpreting Language Models
di: Stoehr, Niklas, et al.
Pubblicazione: (2024)

CogSteer: Cognition-Inspired Selective Layer Intervention for Efficiently Steering Large Language Models
di: Wang, Xintong, et al.
Pubblicazione: (2024)

FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steering
di: Li, Yichen, et al.
Pubblicazione: (2025)

Prototype-Based Dynamic Steering for Large Language Models
di: Kayan, Ceyhun Efe, et al.
Pubblicazione: (2025)

WorkTeam: Constructing Workflows from Natural Language with Multi-Agents
di: Liu, Hanchao, et al.
Pubblicazione: (2025)

Steering Awareness: Detecting Activation Steering from Within
di: Rivera, Joshua Fonseca, et al.
Pubblicazione: (2025)

Endogenous Resistance to Activation Steering in Language Models
di: McKenzie, Alex, et al.
Pubblicazione: (2026)

Decomposing and Steering Functional Metacognition in Large Language Models
di: Li, Yanshi, et al.
Pubblicazione: (2026)

Cross-Lingual Activation Steering for Multilingual Language Models
di: Pokharel, Rhitabrat, et al.
Pubblicazione: (2026)

HyperSteer: Activation Steering at Scale with Hypernetworks
di: Sun, Jiuding, et al.
Pubblicazione: (2025)

Steering off Course: Reliability Challenges in Steering Language Models
di: Da Silva, Patrick Queiroz, et al.
Pubblicazione: (2025)

Multi-property Steering of Large Language Models with Dynamic Activation Composition
di: Scalena, Daniel, et al.
Pubblicazione: (2024)

ContextFocus: Activation Steering for Contextual Faithfulness in Large Language Models
di: Anand, Nikhil, et al.
Pubblicazione: (2026)

Corpus-Steered Query Expansion with Large Language Models
di: Lei, Yibin, et al.
Pubblicazione: (2024)

GeoSteer: Faithful Chain-of-Thought Steering via Latent Manifold Gradients
di: Kazama, Kentaro, et al.
Pubblicazione: (2026)

Trojan Activation Attack: Red-Teaming Large Language Models using Activation Steering for Safety-Alignment
di: Wang, Haoran, et al.
Pubblicazione: (2023)

SteerX: Disentangled Steering for LLM Personalization
di: Zhao, Xiaoyan, et al.
Pubblicazione: (2025)

REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering
di: Zhan, Li-Ming, et al.
Pubblicazione: (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	"it"
VUFIND_SESSION	"50ea4ff255ea2d9235d8dca6dfb44691"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790359045.1915 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "50ea4ff255ea2d9235d8dca6dfb44691" ]`]
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=50ea4ff255ea2d9235d8dca6dfb44691; 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	"14241"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2501_11036/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2501_11036/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790359045.1758
REQUEST_TIME	1790359045

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler