arXiv:2604. 20329v3 Announce Type: replace-cross Abstract: Recent works show that image and video generators exhibit zero-shot visual understanding behaviors, in a way reminiscent of how LLMs develop emergent capabilities of language understanding and reasoning from generative pretraining.
arXiv:2511. 16107v3 Announce Type: replace-cross Abstract: Visual in-context learning (VICL) solves visual tasks by conditioning on a few input-output demonstrations without any model training.
arXiv:2607. 00684v1 Announce Type: new Abstract: The classification accuracy of pretrained Vision-Language Models (VLMs) relies on the quality of the text prompts.
arXiv:2601. 22108v2 Announce Type: replace-cross Abstract: Continued pretraining is optimized with fixed self-supervised tasks but selected by downstream performance, creating a coarse feedback loop in which practitioners evaluate checkpoints, change data mixtures or objectives, and restart runs, while individual updates remain blind to target capabilities.
arXiv:2607. 02269v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have demonstrated immense promise in Spatio-Temporal Video Grounding (STVG).
arXiv:2607. 09024v1 Announce Type: cross Abstract: Driven by next-token prediction, NLP shifted from task-specific models into powerful generalist foundation models.
ES‑VP introduces Energy‑Shaped Visual Prompting, a method that generates image‑specific prompts through low‑rank initialization and energy‑guided dynamic adaptation. It achieves higher performance than existing single‑prompt and diverse‑prompt approaches while using far fewer parameters. Experiments on five architectures and fifteen datasets show consistent superiority, including a 2.6% accuracy gain over DAM‑VP on CLIP with 590× fewer prompt parameters.
The paper investigates how Vision‑Language‑Action (VLA) models can generalise across different driving environments and camera setups. It introduces a multi‑dataset training strategy and an auxiliary objective called BEV‑Forcing, which injects bird‑eye‑view spatial information into the VLA backbone to improve both in‑distribution and out‑of‑distribution performance on a limited number of camera rigs. The authors observe that while BEV‑Forcing helps when training data is scarce, its advantage diminishes as the number of training embodiments grows, suggesting that scaling diversity may reduce the impact of such auxiliary tasks.
arXiv:2607. 18958v1 Announce Type: cross Abstract: While Large Vision-Language Models (LVLMs), represented by LLaVA and GPT-4V, have demonstrated remarkable capabilities, their visual inputs remain vulnerable to adversarial attacks, posing significant security risks.
arXiv:2602. 23197v2 Announce Type: replace-cross Abstract: Transformer-based large language models exhibit in-context learning, enabling adaptation to downstream tasks via few-shot prompting with demonstrations.
The paper introduces EvoPrompt, a framework for adapting vision‑language models to new tasks with limited data while preventing catastrophic forgetting. EvoPrompt uses a Modality‑Shared Prompt Projector to create hierarchical prompts and an evolutionary training strategy that separates low‑rank updates into directional and magnitude components, preserving learned semantic directions. Experiments show that EvoPrompt achieves state‑of‑the‑art few‑shot performance while maintaining the original zero‑shot capabilities of the pre‑trained models.
arXiv:2607. 07179v1 Announce Type: cross Abstract: Document Visual Question Answering (DocVQA) presents a complex multimodal challenge, requiring models to exploit visual, textual, and layout information from documents.