SemShareKV: Efficient KVCache Sharing for Semantically Similar Prompts via Token-Level LSH Matching

Fuente: arXiv

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Zhao, Xinye, Mastorakis, Spyridon
Format:	Preprint
Veröffentlicht:	2025
Schlagworte:	Computation and Language Artificial Intelligence
Online-Zugang:	Acceder al recurso
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Ähnliche Einträge

Helping Large Language Models Protect Themselves: An Enhanced Filtering and Summarization System
von: Muhaimin, Sheikh Samit, et al.
Veröffentlicht: (2025)

KVSharer: Efficient Inference via Layer-Wise Dissimilar KV Cache Sharing
von: Yang, Yifei, et al.
Veröffentlicht: (2024)

SemToken: Semantic-Aware Tokenization for Efficient Long-Context Language Modeling
von: Liu, Dong, et al.
Veröffentlicht: (2025)

HySparse: A Hybrid Sparse Attention Architecture with Oracle Token Selection and KV Cache Sharing
von: Gao, Yizhao, et al.
Veröffentlicht: (2026)

Beyond KV Caching: Shared Attention for Efficient LLMs
von: Liao, Bingli, et al.
Veröffentlicht: (2024)

SentenceKV: Efficient LLM Inference via Sentence-Level Semantic KV Caching
von: Zhu, Yuxuan, et al.
Veröffentlicht: (2025)

Krul: Efficient State Restoration for Multi-turn Conversations with Dynamic Cross-layer KV Sharing
von: Wen, Junyi, et al.
Veröffentlicht: (2025)

ShareLoRA: Parameter Efficient and Robust Large Language Model Fine-tuning via Shared Low-Rank Adaptation
von: Song, Yurun, et al.
Veröffentlicht: (2024)

TokenRatio: Principled Token-Level Preference Optimization via Ratio Matching
von: Nguyen, Truong, et al.
Veröffentlicht: (2026)

DeltaKV: Residual-Based KV Cache Compression via Long-Range Similarity
von: Hao, Jitai, et al.
Veröffentlicht: (2026)

TokenSelect: Efficient Long-Context Inference and Length Extrapolation for LLMs via Dynamic Token-Level KV Cache Selection
von: Wu, Wei, et al.
Veröffentlicht: (2024)

Shared Path: Unraveling Memorization in Multilingual LLMs through Language Similarities
von: Luo, Xiaoyu, et al.
Veröffentlicht: (2025)

CompressKV: Semantic Retrieval Heads Know What Tokens are Not Important Before Generation
von: Lin, Xiaolin, et al.
Veröffentlicht: (2025)

Team Ryu's Submission to SIGMORPHON 2024 Shared Task on Subword Tokenization
von: Li, Zilong
Veröffentlicht: (2024)

BERT-LSH: Reducing Absolute Compute For Attention
von: Li, Zezheng, et al.
Veröffentlicht: (2024)

Leveraging KV Similarity for Online Structured Pruning in LLMs
von: Lee, Jungmin, et al.
Veröffentlicht: (2025)

Sem-DPO: Mitigating Semantic Inconsistency in Preference Optimization for Prompt Engineering
von: Mohamed, Anas, et al.
Veröffentlicht: (2025)

KV Prediction for Improved Time to First Token
von: Horton, Maxwell, et al.
Veröffentlicht: (2024)

1 Trillion Token (1TT) Platform: A Novel Framework for Efficient Data Sharing and Compensation in Large Language Models
von: Park, Chanjun, et al.
Veröffentlicht: (2024)

PQCache: Product Quantization-based KVCache for Long Context LLM Inference
von: Zhang, Hailin, et al.
Veröffentlicht: (2024)

CROP: Token-Efficient Reasoning in Large Language Models via Regularized Prompt Optimization
von: Shah, Deep, et al.
Veröffentlicht: (2026)

Metric-Fair Prompting: Treating Similar Samples Similarly
von: Wang, Jing, et al.
Veröffentlicht: (2025)

Token Alignment via Character Matching for Subword Completion
von: Athiwaratkun, Ben, et al.
Veröffentlicht: (2024)

Share Your Attention: Transformer Weight Sharing via Matrix-based Dictionary Learning
von: Zhussip, Magauiya, et al.
Veröffentlicht: (2025)

DroidSpeak: KV Cache Sharing for Cross-LLM Communication and Multi-LLM Serving
von: Liu, Yuhan, et al.
Veröffentlicht: (2024)

KVReviver: Reversible KV Cache Compression with Sketch-Based Token Reconstruction
von: Yuan, Aomufei, et al.
Veröffentlicht: (2025)

DynSplit-KV: Dynamic Semantic Splitting for KVCache Compression in Efficient Long-Context LLM Inference
von: Ye, Jiancai, et al.
Veröffentlicht: (2026)

Sharing State Between Prompts and Programs
von: Cheng, Ellie Y., et al.
Veröffentlicht: (2025)

SemBench: A Universal Semantic Framework for LLM Evaluation
von: Zubillaga, Mikel, et al.
Veröffentlicht: (2026)

FreeKV: Boosting KV Cache Retrieval for Efficient LLM Inference
von: Liu, Guangda, et al.
Veröffentlicht: (2025)

Rethinking Word Similarity: Semantic Similarity through Classification Confusion
von: Zhou, Kaitlyn, et al.
Veröffentlicht: (2025)

MiSS: Revisiting the Trade-off in LoRA with an Efficient Shard-Sharing Structure
von: Kang, Jiale, et al.
Veröffentlicht: (2024)

Token-Guard: Towards Token-Level Hallucination Control via Self-Checking Decoding
von: Zhu, Yifan, et al.
Veröffentlicht: (2026)

Semantic Caching of Contextual Summaries for Efficient Question-Answering with Language Models
von: Couturier, Camille, et al.
Veröffentlicht: (2025)

Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference
von: Feng, Yuan, et al.
Veröffentlicht: (2024)

Tracing the Roots of Facts in Multilingual Language Models: Independent, Shared, and Transferred Knowledge
von: Zhao, Xin, et al.
Veröffentlicht: (2024)

Efficient and Effective Prompt Tuning via Prompt Decomposition and Compressed Outer Product
von: Lan, Pengxiang, et al.
Veröffentlicht: (2025)

IntactKV: Improving Large Language Model Quantization by Keeping Pivot Tokens Intact
von: Liu, Ruikang, et al.
Veröffentlicht: (2024)

Generative Caching for Structurally Similar Prompts and Responses
von: Chakraborty, Sarthak, et al.
Veröffentlicht: (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	"de"
VUFIND_SESSION	"8db470839888287d2de8f50ab0d9ae28"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790839229.8707 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "8db470839888287d2de8f50ab0d9ae28" ]`]
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=de; VUFIND_SESSION=8db470839888287d2de8f50ab0d9ae28; 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	"1820"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2509_24832/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2509_24832/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790839229.8662
REQUEST_TIME	1790839229

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler