arXiv:2608. 05000v1 Announce Type: cross Abstract: Vision offers a critical axis for advancing foundation models, driving a shift towards natively unified multimodal pretraining.
arXiv:2606. 20077v1 Announce Type: cross Abstract: Visual tokens enter Large Language Models (LLMs) as raw, foreign signals.
The paper introduces AttWarp, a lightweight technique that uses a multimodal large language model’s cross‑modal attention to perform rectilinear warping of input images at test time. By reallocating spatial resolution toward query‑relevant regions without altering model weights or architecture, AttWarp preserves global context while making small objects and subtle relationships easier for the model to read. Experiments on five benchmarks and four MLLMs show consistent accuracy gains, improved compositional reasoning, and reduced hallucinations compared to baseline image‑manipulation methods.
VIVAS is a new Vision‑Language Model pre‑training framework that addresses the lack of fine‑grained visual perception in existing VLMs. It introduces a unified token space and a dense‑structural‑semantic vision tokenizer that expands the textual vocabulary with visual tokens, enabling vision‑language unified autoregressive supervision over both visual details and linguistic content. Trained on 12.4 T tokens, VIVAS achieves state‑of‑the‑art results on 7 tasks and 39 multimodal benchmarks.
arXiv:2607. 18615v1 Announce Type: cross Abstract: Machine unlearning for vision-language models (VLMs) remains underexplored.
arXiv:2607. 03738v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) generate responses autoregressively, integrating visual and linguistic information in an evolving context.
arXiv:2607. 06445v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) are increasingly utilized as the conditioning backbone for diffusion-based image editing due to their remarkable multimodal reasoning capabilities.
The paper investigates attention sinks—tokens or positions that attract disproportionate attention—in text-to-image transformers, comparing bidirectional diffusion models and causal autoregressive models. It finds that both model types exhibit attention sinks, but with distinct patterns: autoregressive models focus on the first and global tokens, while diffusion models develop sinks near modality boundaries, within text regions, and at spatial corners. Perturbation experiments show that sink-related heads influence generation more than non-sink heads, and introducing register tokens can redistribute attention without harming performance.
Modern computer vision pipelines remain fragmented, with tasks such as text-to-image generation, editing, restoration, and classical perception handled by separate models. We study Unified Visual Generation (UVG), where a single model produces diverse image-valued outputs through a unified multimodal interface.
arXiv:2507. 18043v2 Announce Type: replace-cross Abstract: Inference-time steering methods offer a lightweight alternative to fine-tuning large language models (LLMs) and vision-language models (VLMs) by modifying internal activations at test time without updating model weights.
The paper introduces Mask Fine‑Tuning (MFT), a new approach for adapting Vision‑Language Models that avoids modifying backbone weights. MFT learns masks to selectively route information through existing pretrained connections, dynamically uncovering subnetworks that better align with downstream tasks. Experiments demonstrate that MFT consistently outperforms both Full Fine‑Tuning and Parameter‑Efficient Fine‑Tuning across multiple benchmarks, while also offering insights into how pretrained VLMs reorganize their internal pathways during adaptation.
arXiv:2607. 24354v1 Announce Type: new Abstract: Automatic prompt optimization (APO) has been widely adopted to adapt vision-language models (VLMs) to downstream tasks without weight updates, yielding promising results.