CompoSET is a new vision‑language benchmark that tests compositionality by presenting each image‑caption pair with a foil that differs by only a single controlled edit—such as color, count, spatial relation, or object state—while keeping the overall scene largely unchanged. The dataset contains 1,776 instances across 80 scenes, 16 edit types in four semantic categories, and captions at three verbosity levels, with an average of only 8% pixel change between base and foil images. Evaluation of 22 dual‑encoder VLMs shows that all models perform at or below chance on CompoSET, contrasting with their strong performance on prior benchmarks that conflate global image differences with compositional reasoning.
By Laure Delisle, Pietro Perona
Flow-CLIP introduces a method to improve compositional image-text matching by refining global image-text alignment with image-conditioned changes in token–patch evidence. It combines Global Compatibility Calibration (GCC) for sentence-level similarity, Image-Conditioned Evidence Refinement (ICER) to incorporate visual patch context, and Cross-Image Margin Ranking (CIMR) to enhance ranking margins when images support correct descriptions. Experiments demonstrate that Flow-CLIP outperforms pretrained CLIP and other global alignment approaches on binding accuracy and reduces reliance on language priors.
By Minghong Wei, Enguang Zhu, Pu Cao, Lu Yang, Qing Song
arXiv:2610.33798v1 Announce Type: cross
Abstract: An image is rarely described only once. Retrieval benchmarks and many collections attach several captions to each picture, yet dual encoders still ju...
By Khang Nhat Hoang Vo, Belal Shoer, Abhirama Subramanyam Penamakuri
Global vision--language similarities compress an image and a caption into one vector, preserving semantics but not which word corresponds to which region or how those regions are arranged; a model can...
arXiv:2609.23717v1 Announce Type: new
Abstract: Global vision--language similarities compress an image and a caption into one vector, preserving semantics but not which word corresponds to which regi...
By Liuyang Song, Yi Zhang, Zhongyi Deng, Daqian Yang, Hongbo Zhang
The paper introduces PRAISE, a test‑time adaptation framework for vision‑language model based person re‑identification. PRAISE constructs an Attribute‑Conditioned Subspace (ACS) from a bank of attribute prompts, reshaping query‑gallery affinities without requiring identity or attribute labels. Experiments on image‑to‑image and text‑to‑image ReID benchmarks show that PRAISE consistently improves retrieval performance over existing backbones and TTA baselines.
By Rajarshi Bhattacharya, Aryan Shukla, Shakeeb Murtaza, Christian Desrosiers, Jose Dolz, Maguelonne Heritier, Eric Granger
CogCanvas is a new benchmark for multi-subject reference-based image generation, featuring 1,952 curated reference images of 100 celebrities, 115 objects/fashion items, and 29 real-world backgrounds. It generates 1,361 compositional prompts with 2–5 people, using a pipeline that includes DINOv2 deduplication, aesthetic filtering, and automated graph derivation for interaction and positioning. The benchmark evaluates three tasks—reference-based multi-human-object generation, text-to-image compositional generation, and reference retrieval—under a six-axis protocol, and introduces BG‑Sim and Attr‑VQA metrics to assess background fidelity and attribute binding.
By Long-Bao Nguyen, Quang-Khai Le, Tam V. Nguyen, Minh-Triet Tran, Trung-Nghia Le
UniFGVC introduces a training‑free framework for few‑shot fine‑grained visual classification that reframes the task as a multimodal retrieval problem. It uses a Category‑Discriminative Visual Captioner to generate attribute‑aware textual descriptions, then matches image‑description pairs with category templates via pre‑trained encoders. Experiments on 12 datasets show that UniFGVC consistently outperforms existing CLIP‑based and fully‑supervised MLLM methods, demonstrating the effectiveness of the retrieval paradigm.
By Haiyun Guo, Hongyu Guo, Xiangzhao Hao, Jiarui Guo, Jinqiao Wang, Tat-Seng Chua
DEPICT is a new training‑free metric for evaluating text‑to‑image alignment. It replaces fixed reference answers with an agreement rule that compares image‑based and caption‑only responses, weighting questions by how decisively the caption determines them. By merging this agreement score with a holistic score, DEPICT improves negation accuracy dramatically and outperforms existing training‑free metrics while matching or exceeding fine‑tuned evaluators on several benchmarks.
By Vasco Ramos, Sandra Godinho Silva, Joao Magalhaes, Ricardo Rei, Pedro Henrique Martins
MINER is a training‑free inference framework that enhances frozen dual‑encoder models for text‑to‑image retrieval when queries refer to small, visually subordinate objects in cluttered scenes. It augments the global image embedding with a bank of region‑level embeddings and applies hubness‑correcting similarity rescoring, thereby recovering visual evidence that global pooling underweights. The authors introduce ROCS, a benchmark derived from Flickr30K and MS COCO, and demonstrate that MINER improves retrieval performance across CLIP, SigLIP, and SigLIP 2 backbones on both ROCS and standard splits, attributing gains mainly to broader spatial coverage rather than precise crop placement.
By Abdulmalik Alquwayfili, Faisal AlMeshal, Jumanah Almajnouni, Huda Abdulhadi Alamri, Muhammad Kamran J Khan
Intelligent systems that act in the world require image understanding that is both comprehensive and spatially grounded. Current vision-language models (VLMs) can generate fluent and detailed image ca...
The paper introduces Preserve-and-Compose Training (PACT) for composed image retrieval, a task where a query image is modified by a textual instruction while preserving visual content from a reference image. PACT learns from image–text–text triplets, using target captions for supervision and visual evidence from the source image to maintain relevant details, without requiring target images or gallery updates. The authors also propose Chord scoring, which blends target similarity with source-relative directional agreement in a frozen image space, and demonstrate that this combined approach yields strong retrieval performance across multiple zero-shot CIR benchmarks and various backbones.
By Sehyun Kwon