arXiv Computer Vision August 28, 2026 By Wei Liu, Xing Deng, Haijian Shao

When Semantics Saturate or Emerge: Adaptation-Conditional Semantic Utility in Source-Free Cross-Domain Few-Shot Learning

Read the original on arXiv Computer Vision →

The paper investigates whether language prompts selected by zero‑shot accuracy remain effective after visual adaptation in source‑free cross‑domain few‑shot learning. Using a paired protocol, the authors compare generic class‑name templates with detailed class descriptions before and after Low‑Rank Adaptation (LoRA) on datasets such as EuroSAT, CropDisease, ISIC, and ChestX. They identify two regimes: semantic saturation, where detailed prompts are already useful before adaptation, and semantic emergence, where detailed prompts become more useful only after visual representation updates, driven by changes in prediction patterns.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

arXiv AI

Jul 22

Attributes Should Come from Images, Not Class Names: Distribution-Conditioned Attribute Selection for Vision-Language Models

arXiv:2607. 18695v1 Announce Type: cross Abstract: A popular route to interpretable zero-shot classification asks a large language model (LLM) to describe each class name and prompts CLIP with the resulting descriptors.

By Gautam Rajendrakumar Gare, Jia Shi, Zhiqiu Lin, Deepak Pathak, John Galeotti, Deva Ramanan

llms rag multimodal

arXiv Computation and Language

Sep 11

Augustinian BabyLM: What Ostensive Definition Can and Cannot Teach a Small Language Model

The paper introduces an experiment where a small masked language model (DeBERTa) is initialized with visual embeddings for tokens that correspond to image regions, following St. Augustine’s ostensive definition of word learning. The visual initialization leaves a measurable imprint that persists through training, yet it does not improve performance on most BabyLM benchmarks that test abstract grammatical knowledge. However, the seeded models show a consistent advantage in zero‑shot object‑property tasks and in a custom Visual‑Property Swap benchmark that probes color, material, size, and shape knowledge, with the advantage confined to the seeded words and transferable to newly seeded words.

By Lisa Bylinina

llms rag benchmarks

arXiv Computer Vision

4d ago

Do Words Matter? What Text Contributes to Few and Full-Shot Vision-Language Anomaly Detection

arXiv:2610.34610v1 Announce Type: new Abstract: Vision-language anomaly detectors describe normal and abnormal appearance with text, and their gains are commonly credited to the semantic content of t...

By Arda Baris Basaran, Mesut Ceylan, Elgar Fleisch, Filipe Barata

rag multimodal benchmarks

arXiv AI

Aug 5

In-Context Collapse in Vision-Language Models and How to Mitigate it?

arXiv:2608. 02830v1 Announce Type: cross Abstract: Many-shot in-context learning (ICL) lets vision-language models (VLMs) adapt from image--label demonstrations without weight updates, and is widely assumed to improve as more demonstrations are supplied.

By Mohammad Rostami

llms computer-vision multimodal benchmarks

arXiv Machine Learning

Aug 3

Visual Distribution Anchoring for Efficient Prompt Tuning

arXiv:2607. 28967v1 Announce Type: cross Abstract: Prompt tuning adapts vision--language models with few trainable parameters, but existing approaches trade off efficiency and adaptation: static textual prompts can overfit source classes, image-conditioned prompts add per-instance computation, and multimodal tuning modifies the visual branch.

By Pouya Parsa, Raoof Zare Moayedi, Seongjin Choi

llms multimodal

arXiv Computation and Language

Sep 25

Free the Language Model From the Vision Encoder: Semantic Serialization as a Perception Interface for Small Language Models

The paper introduces a perception interface that separates vision from language in vision‑language models. A frozen perception stack detects objects, a deterministic semantic serializer converts the perceived state into text, and a standard text‑only large language model (LLM) answers questions. Experiments on a campus‑robot benchmark show that this serialized interface outperforms a zero‑shot VLM of the same language‑model size, especially as the language model shrinks, and that the advantage persists under paraphrase and different supervision regimes.

By Cong Xu, Ravi Sankar

llms robotics fine-tuning multimodal benchmarks