REVIEW 8 cited by
VARGPT: Unified Understanding and Generation in a Visual Autoregressive Multimodal Large Language Model
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original 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/}
Forward citations
Cited by 8 Pith papers
-
Demystifying Video Reasoning
Video diffusion models reason along the denoising trajectory (Chain-of-Steps), not primarily across frames, and this mechanism can be nudged by ensembling latent trajectories.
-
UniGen-AR: Unifying Visual Generation with Auto-Regressive Modeling
An MLLM-conditioned next-scale VAR decoder handles 15+ unified visual generation tasks with competitive quality and substantially lower latency than diffusion baselines.
-
Unified Audio Intelligence Without Regressing on Text Intelligence
A unified 30B MoE audio-text LLM achieves state-of-the-art audio understanding, generation, and speech tasks while preserving text reasoning comparable to its text-only backbone.
-
Instella-T2I: Pushing the Limits of 1D Discrete Latent Space Image Generation
1D binary image latents reduce a 1024x1024 image to 128 discrete tokens and support text-to-image generation with diffusion and autoregressive models.
-
ComplexBench-Edit: Benchmarking Complex Instruction-Driven Image Editing via Compositional Dependencies
Introduces a benchmark for chain-dependent image editing instructions plus a region-aware consistency metric, and shows a chain-of-thought prompt improves a Gemini-based editor.
-
Not All Tokens and Heads Are Equally Important: Dual-Level Attention Intervention for Hallucination Mitigation
A dual-level attention intervention that boosts salient visual-token attention and suppresses text/system attention during decoding reduces hallucination rates in LLaVA, MiniGPT-4, and mPLUG-Owl2 on POPE and CHAIR.
-
Do we really have to filter out random noise in pre-training data for language models?
Random noise in pre-training data has a surprisingly small effect on language model next-token loss, but can still hurt downstream tasks; a new local gradient matching loss partially counteracts this.
-
Draw ALL Your Imagine: A Holistic Benchmark and Agent Framework for Complex Instruction-based Image Generation
A new benchmark and agent framework for complex text-to-image generation, with an unvalidated AI-judge evaluation and claims that the agent outperforms GPT-4o on the authors' own benchmark.
Discussion (0). Continue with ORCID to comment.