ILLUME+: Illuminating Unified MLLM with Dual Visual Tokenization and Diffusion Refinement

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Huang, Runhui, Wang, Chunwei, Yang, Junwei, Lu, Guansong, Yuan, Yunlong, Han, Jianhua, Hou, Lu, Zhang, Wei, Hong, Lanqing, Zhao, Hengshuang, Xu, Hang
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Computer Vision and Pattern Recognition
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

ILLUME: Illuminating Your LLMs to See, Draw, and Self-Enhance
di: Wang, Chunwei, et al.
Pubblicazione: (2024)

HiRes-LLaVA: Restoring Fragmentation Input in High-Resolution Large Vision-Language Models
di: Huang, Runhui, et al.
Pubblicazione: (2024)

DrivePI: Spatial-aware 4D MLLM for Unified Autonomous Driving Understanding, Perception, Prediction and Planning
di: Liu, Zhe, et al.
Pubblicazione: (2025)

SemHiTok: A Unified Image Tokenizer via Semantic-Guided Hierarchical Codebook for Multimodal Understanding and Generation
di: Chen, Zisheng, et al.
Pubblicazione: (2025)

LayerDiff: Exploring Text-guided Multi-layered Composable Image Synthesis via Layer-Collaborative Diffusion Model
di: Huang, Runhui, et al.
Pubblicazione: (2024)

UNIT: Unifying Image and Text Recognition in One Vision Encoder
di: Zhu, Yi, et al.
Pubblicazione: (2024)

Getting More Juice Out of Your Data: Hard Pair Refinement Enhances Visual-Language Models Without Extra Data
di: Wang, Haonan, et al.
Pubblicazione: (2023)

Brick-Diffusion: Generating Long Videos with Brick-to-Wall Denoising
di: Yuan, Yunlong, et al.
Pubblicazione: (2025)

Towards Unified Multimodal Interleaved Generation via Group Relative Policy Optimization
di: Nie, Ming, et al.
Pubblicazione: (2026)

DualToken: Towards Unifying Visual Understanding and Generation with Dual Visual Vocabularies
di: Song, Wei, et al.
Pubblicazione: (2025)

AlphaGRPO: Unlocking Self-Reflective Multimodal Generation in UMMs via Decompositional Verifiable Reward
di: Huang, Runhui, et al.
Pubblicazione: (2026)

FreqPrior: Improving Video Diffusion Models with Frequency Filtering Gaussian Noise
di: Yuan, Yunlong, et al.
Pubblicazione: (2025)

RealignDiff: Boosting Text-to-Image Diffusion Model with Coarse-to-fine Semantic Re-alignment
di: Jiang, Zutao, et al.
Pubblicazione: (2023)

From Summary to Action: Enhancing Large Language Models for Complex Tasks with Open World APIs
di: Liu, Yulong, et al.
Pubblicazione: (2024)

PanGu-Draw: Advancing Resource-Efficient Text-to-Image Synthesis with Time-Decoupled Training and Reusable Coop-Diffusion
di: Lu, Guansong, et al.
Pubblicazione: (2023)

Efficient Multi-modal Large Language Models via Visual Token Grouping
di: Huang, Minbin, et al.
Pubblicazione: (2024)

PaMi-VDPO: Mitigating Video Hallucinations by Prompt-Aware Multi-Instance Video Preference Learning
di: Ding, Xinpeng, et al.
Pubblicazione: (2025)

SaDiT: Efficient Protein Backbone Design via Latent Structural Tokenization and Diffusion Transformers
di: Mo, Shentong, et al.
Pubblicazione: (2026)

Implicit Concept Removal of Diffusion Models
di: Liu, Zhili, et al.
Pubblicazione: (2023)

FocalClick-XL: Towards Unified and High-quality Interactive Segmentation
di: Chen, Xi, et al.
Pubblicazione: (2025)

EDEN: Enhanced Diffusion for High-quality Large-motion Video Frame Interpolation
di: Zhang, Zihao, et al.
Pubblicazione: (2025)

Revisiting MLLM Token Technology through the Lens of Classical Visual Coding
di: Liu, Jinming, et al.
Pubblicazione: (2025)

HaploVL: A Single-Transformer Baseline for Multi-Modal Understanding
di: Yang, Rui, et al.
Pubblicazione: (2025)

InstructX: Towards Unified Visual Editing with MLLM Guidance
di: Mou, Chong, et al.
Pubblicazione: (2025)

PUMA: Empowering Unified MLLM with Multi-granular Visual Generation
di: Fang, Rongyao, et al.
Pubblicazione: (2024)

Task-customized Masked AutoEncoder via Mixture of Cluster-conditional Experts
di: Liu, Zhili, et al.
Pubblicazione: (2024)

MedVeriSeg: Teaching MLLM-Based Medical Segmentation Models to Verify Query Validity Without Extra Training
di: Lu, Ziqian, et al.
Pubblicazione: (2026)

Slot-MLLM: Object-Centric Visual Tokenization for Multimodal LLM
di: Chi, Donghwan, et al.
Pubblicazione: (2025)

CAD-MLLM: Unifying Multimodality-Conditioned CAD Generation With MLLM
di: Xu, Jingwei, et al.
Pubblicazione: (2024)

AtomThink: Multimodal Slow Thinking with Atomic Step Reasoning
di: Xiang, Kun, et al.
Pubblicazione: (2024)

Saliency-R1: Incentivizing Unified Saliency Reasoning Capability in MLLM with Confidence-Guided Reinforcement Learning
di: Li, Long, et al.
Pubblicazione: (2025)

RADAR: Revealing Asymmetric Development of Abilities in MLLM Pre-training
di: Nie, Yunshuang, et al.
Pubblicazione: (2026)

VisionTrim: Unified Vision Token Compression for Training-Free MLLM Acceleration
di: Yu, Hanxun, et al.
Pubblicazione: (2026)

Boosting Chart-to-Code Generation in MLLM via Dual Preference-Guided Refinement
di: Zhang, Zhihan, et al.
Pubblicazione: (2025)

EMOVA: Empowering Language Models to See, Hear and Speak with Vivid Emotions
di: Chen, Kai, et al.
Pubblicazione: (2024)

Can Atomic Step Decomposition Enhance the Self-structured Reasoning of Multimodal Large Models?
di: Xiang, Kun, et al.
Pubblicazione: (2025)

Treat Visual Tokens as Text? But Your MLLM Only Needs Fewer Efforts to See
di: Zhang, Zeliang, et al.
Pubblicazione: (2024)

MergeTok: Unified Continuous and Discrete Visual Tokenization via Token Merging
di: Zhang, Luyuan, et al.
Pubblicazione: (2026)

Seg-VAR: Image Segmentation with Visual Autoregressive Modeling
di: Zheng, Rongkun, et al.
Pubblicazione: (2025)

Exploring MLLM-Diffusion Information Transfer with MetaCanvas
di: Lin, Han, 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
VUFIND_SESSION	"60b2b6c871811559aafcaaf59c609694"
language	"it"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790712626.6452 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "60b2b6c871811559aafcaaf59c609694" ]`]
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=60b2b6c871811559aafcaaf59c609694; 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	"56483"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2504_01934/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2504_01934/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790712626.64
REQUEST_TIME	1790712626

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler