Bio‑VQA is a new, large-scale visual question answering benchmark for biology, comprising 1.86 million questions over 217 k images from 11 datasets. It tests four key aspects of biological visual understanding—taxonomic reasoning, attribute grounding, spatial reasoning, and individual re‑identification—using a graded scoring system that rewards near‑misses more than distant ones. Evaluation of 18 open‑weight vision‑language models shows that general visual capabilities do not transfer well to biological tasks, with markedly lower accuracy on counting, taxonomy, trait measurement, and re‑identification compared to everyday object benchmarks.
arXiv:2609. 11916v1 Announce Type: new Abstract: Camera traps often run in the field on edge hardware with limited or no connectivity, making small, locally-deployable vision-language models (VLMs) -- not frontier-scale ones -- the practically relevant class to evaluate for species identification.
arXiv:2508. 17117v3 Announce Type: replace-cross Abstract: Existing plant-disease datasets target classification and detection, leaving vision-language models unable to support interactive, reasoning-based diagnosis.
arXiv:2609.09417v1 Announce Type: new Abstract: Vision-language models (VLMs) show promise for agricultural classification, but zero-shot performance on disease, pest, damage, quality, and species id...
arXiv:2608. 08727v1 Announce Type: cross Abstract: To address this gap, we introduce TomaMMU, a large-scale Tomato leaf disease MultiModal Understanding dataset, alongside TomaBench, a benchmark for evaluating VLMs on tomato disease understanding.
arXiv:2609.32352v2 Announce Type: replace-cross Abstract: Vision-language models (VLMs) have shown increasing potential for medical image understanding, yet their capabilities in ophthalmic imaging r...
To address this gap, we introduce TomaMMU, a large-scale Tomato leaf disease MultiModal Understanding dataset, alongside TomaBench, a benchmark for evaluating VLMs on tomato disease understanding. TomaMMU comprises 28,808 high-quality images spanning 15 categories and 213,119 human-annotated visual question-answer pairs, generated through a three-stage pipeline comprising Data Collection, Human Annotation, and Question-Answer Generation.
FieldGuideRAG is a training‑free GraphRAG approach that uses a frozen vision‑language model to recognize fine‑grained species by comparing a query image to photographs stored on each species node in a knowledge graph. The graph links species to relatives, geographic range, and ecological facts, and the system fuses verification scores with range edges to answer organism‑related questions. On datasets such as CUB‑200 and iNaturalist, FieldGuideRAG’s use of node photographs improves verification over species names, surpasses a SigLIP retriever by up to 16 points, and achieves up to 89.5% accuracy on ecological questions.
The paper introduces PixelJev, a native-image decision interface that combines an image, a task instruction, and a runtime candidate set into a structured choice and candidate-conditioned probabilities using small open multimodal models. It unifies recognition and multiple-choice visual question answering via a language-model readout, offering options for frozen inference, language-side adaptation, and held-out calibration. Across seven benchmarks, 64-shot source adaptation significantly boosts Pets accuracy from 60.13% to 92.40%, and the system supports both VQA tasks with frozen inference, while also highlighting areas for improvement such as schema robustness and cross-family transfer.
The paper investigates whether open‑weight multimodal large language models (MLLMs) can perform simple visual comparisons, specifically measuring body‑part sizes in synthetic ant images. It finds that while some body parts are compared accurately, others fail despite the task being visually trivial, indicating a grounding failure linked to the models’ inability to recognize certain structures. The study evaluates various in‑context learning strategies and shows that recognition‑driven grounding limits even basic visual comparisons that do not inherently require semantic understanding.
The paper introduces PixelJev, a native-image decision interface that combines an image, a task instruction, and a runtime candidate set to produce a structured choice and candidate-conditioned probabilities using small open multimodal models. It unifies recognition and multiple-choice visual question answering via a language-model readout, offering options for frozen inference, language-side adaptation, and held-out calibration. Across seven benchmarks, 64-shot source adaptation significantly boosts Pets accuracy from 60.13% to 92.40%, and the model supports VQA tasks without target fitting, though calibration and cross-family transfer remain challenges.
arXiv:2512. 15748v2 Announce Type: replace Abstract: Visual Species Recognition (VSR) is a fundamental task in scientific disciplines that require species-level identification, including ecology, palynology, evolutionary biology, systematics, and phylogenetics.
arXiv:2606. 07861v1 Announce Type: cross Abstract: Recent vision-language models (VLMs) excel at multimodal understanding and reasoning, yet their fine-grained visual perception remains underexplored.