HACK++ is a head-aware KV cache compression framework for VAR models that decouples current-scale attention from historical cache under adaptive per-head budgets to achieve near-lossless generation at 30% attention and 10% cache budgets.
V ARGPT: unified understanding and generation in a visual autoregres- IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 21 sive multimodal large language model
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
abstract
We present VARGPT, a novel multimodal large language model (MLLM) that unifies visual understanding and generation within a single autoregressive framework. VARGPT employs a next-token prediction paradigm for visual understanding and a next-scale prediction paradigm for visual autoregressive generation. VARGPT innovatively extends the LLaVA architecture, achieving efficient scale-wise autoregressive visual generation within MLLMs while seamlessly accommodating mixed-modal input and output within a single model framework. Our VARGPT undergoes a three-stage unified training process on specially curated datasets, comprising a pre-training phase and two mixed visual instruction-tuning phases. The unified training strategy are designed to achieve alignment between visual and textual features, enhance instruction following for both understanding and generation, and improve visual generation quality, respectively. Despite its LLAVA-based architecture for multimodel understanding, VARGPT significantly outperforms LLaVA-1.5 across various vision-centric benchmarks, such as visual question-answering and reasoning tasks. Notably, VARGPT naturally supports capabilities in autoregressive visual generation and instruction-to-image synthesis, showcasing its versatility in both visual understanding and generation tasks. Project page is at: \url{https://vargpt-1.github.io/}
citation-role summary
background 1
baseline 1
citation-polarity summary
representative citing papers
Audex unifies audio understanding and generation on a strong text MoE backbone with multi-stage SFT plus text-only Cascade RL, matching open SOTA audio scores while mostly retaining text capability.
A self-evolving framework with proposer-solver-generator roles, Solver Token Entropy, and multi-scale internal evaluation improves unified LMMs on understanding and generation tasks using only self-derived consistency signals.
MMaDA is a unified multimodal diffusion model using mixed chain-of-thought fine-tuning and a new UniGRPO reinforcement learning algorithm that outperforms specialized models in reasoning, understanding, and text-to-image tasks.
This is the first survey on vision-language-action models, providing a taxonomy across three lines, plus summaries of datasets, simulators, benchmarks, challenges, and future directions in embodied AI.
MEPA adds token-routed MoE and residual self-supervised feature alignment to VAR models, reporting better FID on ImageNet 256x256 with half the training epochs and fewer parameters than dense baselines.
Semantic Generative Tuning applies segmentation-based generative proxies during post-training to align and improve both understanding and generation in unified multimodal models.
WinTok is a hybrid visual tokenizer that supplements pixel tokens with learnable semantic tokens distilled asymmetrically from foundation models to improve reconstruction, understanding, and generation.
The paper supplies a unified definition based on data flow and dynamic interaction plus a systematic taxonomy to organize fragmented work on streaming large language models.
citing papers explorer
-
HACK++: Towards More Effective Head-Aware Key-Value Compression for Efficient Visual Autoregressive Modeling
HACK++ is a head-aware KV cache compression framework for VAR models that decouples current-scale attention from historical cache under adaptive per-head budgets to achieve near-lossless generation at 30% attention and 10% cache budgets.
-
Unified Audio Intelligence Without Regressing on Text Intelligence
Audex unifies audio understanding and generation on a strong text MoE backbone with multi-stage SFT plus text-only Cascade RL, matching open SOTA audio scores while mostly retaining text capability.
-
Ask, Solve, Generate: Self-Evolving Unified Multimodal Understanding and Generation via Self-Consistency Rewards
A self-evolving framework with proposer-solver-generator roles, Solver Token Entropy, and multi-scale internal evaluation improves unified LMMs on understanding and generation tasks using only self-derived consistency signals.
-
MMaDA: Multimodal Large Diffusion Language Models
MMaDA is a unified multimodal diffusion model using mixed chain-of-thought fine-tuning and a new UniGRPO reinforcement learning algorithm that outperforms specialized models in reasoning, understanding, and text-to-image tasks.
-
A Survey on Vision-Language-Action Models for Embodied AI
This is the first survey on vision-language-action models, providing a taxonomy across three lines, plus summaries of datasets, simulators, benchmarks, challenges, and future directions in embodied AI.
-
MEPA: Multi-Scale Representation Alignment for Visual Autoregressive Modeling with Mixture of Experts
MEPA adds token-routed MoE and residual self-supervised feature alignment to VAR models, reporting better FID on ImageNet 256x256 with half the training epochs and fewer parameters than dense baselines.
-
Semantic Generative Tuning for Unified Multimodal Models
Semantic Generative Tuning applies segmentation-based generative proxies during post-training to align and improve both understanding and generation in unified multimodal models.
-
WinTok: A Win-Win Hybrid Tokenizer via Decomposing Visual Understanding and Generation with Transferable Tokens
WinTok is a hybrid visual tokenizer that supplements pixel tokens with learnable semantic tokens distilled asymmetrically from foundation models to improve reconstruction, understanding, and generation.
-
From Static Inference to Dynamic Interaction: A Survey of Streaming Large Language Models
The paper supplies a unified definition based on data flow and dynamic interaction plus a systematic taxonomy to organize fragmented work on streaming large language models.