# image recognition

Image recognition is a computer vision and machine learning technique for identifying and classifying objects, scenes, or patterns in digital images or video.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Backbone Matters: Benchmarking Encoder-Decoder Networks for Semantic Segmentation of Driving Scenes on BDD100K

DevFeed: [Backbone Matters: Benchmarking Encoder-Decoder Networks for Semantic Segmentation of Driving Scenes on BDD100K](<https://devfeed.tech/articles/backbone-matters-benchmarking-encoder-decoder-networks-for-semantic-segmentation-of-driving-scenes-on-bdd100k-145151.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26191>)

Author: Ehsan Rassekh

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [autonomous vehicles](<https://devfeed.tech/topics/autonomous-vehicles.md>), [Machine Perception](<https://devfeed.tech/topics/machine-perception.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [decoder](<https://devfeed.tech/tags/decoder.md>), [driving](<https://devfeed.tech/tags/driving.md>), [encoder](<https://devfeed.tech/tags/encoder.md>), [encoder-decoder](<https://devfeed.tech/tags/encoder-decoder.md>), [networks](<https://devfeed.tech/tags/networks.md>), [semantic](<https://devfeed.tech/tags/semantic.md>)

### AI overview

On BDD100K, decoder choice strongly changes how a backbone performs: EfficientNet-B4 scores 62.6% mIoU with FPN but 42.6% with PSPNet. The study trained 52 combinations of four decoders (U-Net, FPN, LinkNet and PSPNet) and 13 backbones across six families using a common protocol. FPN with EfficientNet-B4 was best, reaching 62.6% mIoU, 89.8% accuracy and 0.092 loss; only two FPN-EfficientNet combinations exceeded the cited BDD100K benchmark of 56.9% mIoU. The best-to-worst backbone gap varied by decoder, from 12.0 points for U-Net to 42.5 for FPN. These are reported results for this dataset and protocol; the supplied text does not provide further methodological details or broader validation.

### Source excerpt

arXiv:2611.26191v1 Announce Type: cross Abstract: Semantic segmentation gives an autonomous vehicle a dense, pixel-level description of its surroundings and is a core part of scene understanding. Most segmentation networks consist of an encoder (the backbone) and a decoder, yet the backbone is often chosen without a systematic comparison. This work studies this choice on the BDD100K driving dataset. Four widely used decoders (U-Net, FPN, LinkNet and PSPNet) are combined with 13 backbones from six families (ResNet, VGG, DenseNet, Inception, MobileNet and EfficientNet), and all 52 models are trained with one common protocol. The best model, an FPN decoder with an EfficientNet-B4 backbone, reaches 62.6% mean intersection over union (mIoU), 89.8% accuracy and a loss of 0.092. Only two of the 52 models, both FPN with EfficientNet, exceed the 56.9% mIoU reported with the BDD100K benchmark. The effect of the backbone depends strongly on the decoder: the same EfficientNet-B4 gives 42.6% mIoU with PSPNet but 62.6% with FPN, and the gap between the best and the worst backbone ranges from 12.0 points (U-Net) to 42.5 points (FPN). The results and the scope of this study are discussed.

## Online Mistake Detection in Instruction-Conditioned Procedural Video

DevFeed: [Online Mistake Detection in Instruction-Conditioned Procedural Video](<https://devfeed.tech/articles/online-mistake-detection-in-instruction-conditioned-procedural-video-145538.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25567>)

Author: Amit Feldman, Ilan Shimshoni, Rotem Dror

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Video Analytics](<https://devfeed.tech/topics/video-analytics.md>), [multimodal-ai](<https://devfeed.tech/topics/multimodal-ai.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [actions](<https://devfeed.tech/tags/actions.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [detection](<https://devfeed.tech/tags/detection.md>), [error](<https://devfeed.tech/tags/error.md>), [error-propagation](<https://devfeed.tech/tags/error-propagation.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [execution](<https://devfeed.tech/tags/execution.md>), [f1](<https://devfeed.tech/tags/f1.md>), [instructions](<https://devfeed.tech/tags/instructions.md>), [localization](<https://devfeed.tech/tags/localization.md>), [mistakes](<https://devfeed.tech/tags/mistakes.md>), [model](<https://devfeed.tech/tags/model.md>), [modular](<https://devfeed.tech/tags/modular.md>), [monitoring](<https://devfeed.tech/tags/monitoring.md>), [offline](<https://devfeed.tech/tags/offline.md>), [online](<https://devfeed.tech/tags/online.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [predictions](<https://devfeed.tech/tags/predictions.md>), [procedural](<https://devfeed.tech/tags/procedural.md>), [protocol](<https://devfeed.tech/tags/protocol.md>), [video](<https://devfeed.tech/tags/video.md>)

### AI overview

The paper studies online mistake detection in egocentric procedural video, conditioning predictions on task instructions, observed video and prior predictions. The system identifies and temporally localizes completed action attempts, then classifies them as correct, modified, slips, corrections or additions; omissions are assessed after the video ends. It compares a modular pipeline for step identification, localization and execution verification with a joint model predicting all outputs at once. After repartitioning EgoPER to include erroneous videos in every split, both approaches achieve similar step-segmentation F1@0.5 (93.6% modular; 92.1% joint). The joint model scores higher on binary mistake detection (84.1% vs. 80.7%), especially for slips (56.4% vs. 37.5%). Component analysis suggests stage-to-stage error propagation contributes to the modular model's lower detection score. On test videos containing mistakes, both approaches outperform released AEM and ESTANet checkpoints under the paper's evaluation protocol.

### Source excerpt

arXiv:2611.25567v1 Announce Type: cross Abstract: An assistant monitoring a person performing a procedure is most useful when it can detect mistakes while there is still time to recover. This requires judging actions against task instructions as the video unfolds, identifying the attempted step, locating it in time, and assessing its execution. Existing online detectors do not use the instruction text, while instruction-conditioned methods operate offline. In this paper, we study online, instruction-conditioned mistake detection in egocentric videos. Using only observed video, task instructions, and prior predictions, our model identifies and localizes completed action attempts and classifies them as correct, modified, slip, corrections, or additions; omitted steps are determined after the video ends. We compare a modular pipeline, with separate step identification, temporal localization, and execution verification, to a joint model that predicts all outputs in a single step. To support supervised learning from mistakes, we repartition EgoPER to include erroneous videos in all splits. The approaches achieve similar step-segmentation F1@0.5 (93.6% and 92.1%), while the joint model achieves higher binary mistake-detection F1@0.5 (84.1% vs. 80.7%), with its largest gain on slips (56.4% vs. 37.5%). Component-level analysis suggests that error propagation between stages contributes to the modular approach's lower detection score. On test videos containing mistakes, both designs outperform released AEM and ESTANet checkpoints under our evaluation protocol.

## Digitization of Analogue Gauges using Computer Vision

DevFeed: [Digitization of Analogue Gauges using Computer Vision](<https://devfeed.tech/articles/digitization-of-analogue-gauges-using-computer-vision-145647.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26262>)

Author: Erico Kayumba Mutonkole, Olasupo Ajayi, Omowunmi Isafiade

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Computer vision](<https://devfeed.tech/topics/computer-vision.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [OpenCV](<https://devfeed.tech/topics/opencv.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [Image processing](<https://devfeed.tech/topics/image-processing.md>)

Tags: [automation](<https://devfeed.tech/tags/automation.md>), [computer-vision](<https://devfeed.tech/tags/computer-vision.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [gauge](<https://devfeed.tech/tags/gauge.md>), [manufacturing](<https://devfeed.tech/tags/manufacturing.md>), [opencv](<https://devfeed.tech/tags/opencv.md>), [residual](<https://devfeed.tech/tags/residual.md>), [web-scraping](<https://devfeed.tech/tags/web-scraping.md>)

### AI overview

The study reports that OpenCV with ResNet18 achieved 97.3% accuracy in reading analogue pressure gauges, compared with 64.4% for OpenCV with VGG19 and 63.8% for OpenCV with Scikit-Image. The comparison used 1,848 gauge images covering 0-10 BAR, assembled through web scraping, synthetic generation, and augmentation. These results favor the ResNet18 approach within the reported evaluation; the supplied abstract does not describe deployment performance.

### Source excerpt

arXiv:2611.26262v1 Announce Type: cross Abstract: Analogue gauges remain prevalent in manufacturing industries for measuring physical quantities, yet manual monitoring by operators introduces human error and operational inefficiency. This research developed and evaluated an automated gauge reading system using computer vision and deep learning techniques. Three hybrid approaches were implemented and compared: OpenCV with Scikit-Image (traditional computer vision), OpenCV with ResNet18, and OpenCV with VGG19 (deep learning approaches). A comprehensive dataset of 1,848 pressure gauge images (0-10 BAR range) was curated through web scraping, synthetic generation, and augmentation techniques. The experimental results demonstrate that ResNet18 significantly outperforms the alternatives, achieving an accuracy of 97.3%, compared to VGG19's 64.4% and Scikit-Image's 63.8%. The study validates the effectiveness of residual learning architectures for industrial gauge digitization, providing a robust foundation for Industry 5.0 automation applications.

## C-LiTE: A Lightweight Foundation Model for Cellular feature learning in Histopathology

DevFeed: [C-LiTE: A Lightweight Foundation Model for Cellular feature learning in Histopathology](<https://devfeed.tech/articles/c-lite-a-lightweight-foundation-model-for-cellular-feature-learning-in-histopathology-145653.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26316>)

Author: Esha Sadia Nasir, Jiaqi Lv, Gozde N. Gunesli, Shan E Ahmed Raza

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [Training AI Models](<https://devfeed.tech/topics/training-ai-models.md>), [Dataset development](<https://devfeed.tech/topics/dataset-development.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>)

Tags: [cellular](<https://devfeed.tech/tags/cellular.md>), [classification](<https://devfeed.tech/tags/classification.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [embedding](<https://devfeed.tech/tags/embedding.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [f1](<https://devfeed.tech/tags/f1.md>), [features](<https://devfeed.tech/tags/features.md>), [fewer](<https://devfeed.tech/tags/fewer.md>), [figure](<https://devfeed.tech/tags/figure.md>), [foundation-model](<https://devfeed.tech/tags/foundation-model.md>), [lightweight](<https://devfeed.tech/tags/lightweight.md>), [native](<https://devfeed.tech/tags/native.md>), [public](<https://devfeed.tech/tags/public.md>), [representation-learning](<https://devfeed.tech/tags/representation-learning.md>), [token](<https://devfeed.tech/tags/token.md>), [training-dataset](<https://devfeed.tech/tags/training-dataset.md>), [transformation](<https://devfeed.tech/tags/transformation.md>), [type](<https://devfeed.tech/tags/type.md>), [vit](<https://devfeed.tech/tags/vit.md>)

### AI overview

C-LiTE is a 22-million-parameter model for learning cellular features from H&E histopathology images. It processes 64x64-pixel cell-centred crops at native resolution, using 4x4-pixel patches to form a 16x16 token grid without upsampling; its patch embedding is initialized by adapting weights from a pre-trained ViT-5-S with a pseudoinverse-based transformation. The model is pre-trained on 60 million TCGA cell crops balanced across 19 cancer types and 16 tissue types. Across seven public datasets for mitotic figure recognition, nucleus and cell-type classification, and immune-cell phenotyping, frozen-feature evaluation yields mean macro-F1 0.670, close to CytoFormer's 0.682 with about 30 times fewer parameters. These findings are limited to the reported evaluation setting.

### Source excerpt

arXiv:2611.26316v1 Announce Type: cross Abstract: Characterising individual cells is fundamental to computational pathology, as cellular morphology informs tasks such as tumour grading, mitotic assessment and immune profiling. However, most pathology foundation models are trained on tissue patches and capture cellular features only implicitly, while existing cell-level models are either restricted to specific cell populations or rely on computationally expensive architectures with hundreds of millions of parameters. Applying ViTs directly to small cellular crops also presents an architectural challenge: upsampling increases computational cost without adding morphological information and may introduce interpolation artefacts. To address these limitations, we introduce C-LiTE, a lightweight foundation model for cellular representation learning in H&E-stained histopathology, with only 22 million parameters. C-LiTE employs a native-resolution tokenisation strategy tailored to cellular inputs, processing 64 x 64 px cell-centred crops using a 4 x 4 patch embedding to produce a 16 x 16 token grid without image upsampling. To transfer visual knowledge from large-scale pre-training, we adapt the patch embedding weights of a pre-trained ViT-5-S using a pseudoinverse-based transformation, enabling weight initialisation at the finer patch resolution. We further assemble a pre-training dataset of 60 million cell-centred crops from TCGA, balanced across 19 cancer types and 16 tissue types. To our knowledge, this represents the largest public dataset for self-supervised pre-training on H&E images at the cellular level. We evaluate C-LiTE on seven public datasets spanning mitotic figure recognition, nucleus and cell-type classification, and immune cell phenotyping. Under frozen-feature evaluation, C-LiTE achieves a mean macro-F1 of 0.670, approaching CytoFormer (0.682) with approximately 30x fewer parameters.

## Seeing Beyond the Obvious: Complementary Semantic Guidance for Salient Single-Positive Multi-Label Learning

DevFeed: [Seeing Beyond the Obvious: Complementary Semantic Guidance for Salient Single-Positive Multi-Label Learning](<https://devfeed.tech/articles/seeing-beyond-the-obvious-complementary-semantic-guidance-for-salient-single-positive-multi-label-learning-145124.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25986>)

Author: Tong Jin, Meng-Long Wang, Xingyu Zhao, Sheng-Jun Huang

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [Image](<https://devfeed.tech/topics/image.md>)

Tags: [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [framework](<https://devfeed.tech/tags/framework.md>), [guidance](<https://devfeed.tech/tags/guidance.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [prototypes](<https://devfeed.tech/tags/prototypes.md>), [ranking](<https://devfeed.tech/tags/ranking.md>), [semantic](<https://devfeed.tech/tags/semantic.md>), [support](<https://devfeed.tech/tags/support.md>), [target](<https://devfeed.tech/tags/target.md>)

### AI overview

Salient Single-Positive Multi-Label Learning records only each image's most salient positive label, so prototypes built from observed positives may miss the visual diversity of unlabeled positives. CoST supplements those Salient Prototypes with class-wise Coverage Prototypes: it selects semantically relevant candidates among unknown classes using task-learned category information and image-text correspondence, then aggregates them by class. A ranking-based selective refinement step strengthens soft targets only for classes with stronger Coverage Prototype support, leaving other base soft targets unchanged. The abstract reports strong gains over existing baselines across multiple benchmarks, but gives no benchmark names or quantitative results.

### Source excerpt

arXiv:2611.25986v1 Announce Type: new Abstract: Salient Single-Positive Multi-Label Learning (SalSPML) assigns each image its most salient positive label, leaving other classes unlabeled. Existing methods construct class-level Salient Prototypes from observed positives to guide the recognition of unobserved classes. However, since salient annotations favor visually prominent instances, these prototypes may not fully capture the visual diversity of hidden positives, resulting in insufficient positive supervision. To address this limitation, we propose CoST, a Coverage-guided Complementary Soft-Target Refinement framework that supplements Salient Prototypes with complementary class-semantic information. Specifically, CoST identifies semantically relevant candidates from unknown classes by leveraging task-learned category information and image-text semantic correspondence. The selected candidates are then aggregated into class-wise Coverage Prototypes to complement the semantic representations provided by Salient Prototypes. Furthermore, we introduce a ranking-based selective refinement strategy that identifies classes receiving stronger semantic support from Coverage Prototypes and selectively enhances their soft targets while preserving the base soft targets for the rest. Experiments on multiple benchmarks demonstrate the effectiveness of CoST, with strong performance gains over existing baselines.

## A Wolf in Sheep's Pixels: Visual Camouflage Attacks via HDR Gain Maps

DevFeed: [A Wolf in Sheep's Pixels: Visual Camouflage Attacks via HDR Gain Maps](<https://devfeed.tech/articles/a-wolf-in-sheep-s-pixels-visual-camouflage-attacks-via-hdr-gain-maps-145517.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25469>)

Author: Qi Shen, Yujia Gong, Yue Zhao, Tong Liu

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Image processing](<https://devfeed.tech/topics/image-processing.md>), [Image](<https://devfeed.tech/topics/image.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [LLM security](<https://devfeed.tech/topics/llm-security.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [attacks](<https://devfeed.tech/tags/attacks.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [exploits](<https://devfeed.tech/tags/exploits.md>), [hdr](<https://devfeed.tech/tags/hdr.md>), [maps](<https://devfeed.tech/tags/maps.md>), [processing](<https://devfeed.tech/tags/processing.md>), [sdr](<https://devfeed.tech/tags/sdr.md>)

### AI overview

DRJanus exploits the difference between SDR image processing and HDR display: automated systems may inspect an SDR base while viewers see a reconstruction that also uses an auxiliary gain map. It embeds two independently chosen color images in one JPEG without querying or optimizing against the target model, preserving the SDR base exactly and achieving an average HDR reconstruction PSNR of 43.03 dB across 400 images. In tests, it evaded all six commercial moderation services and nine vision-language models on a 100-image sexual-content benchmark. Across five agents, tested on 120 task cases per threat setting, attack success ranged from 31.67% to 100% for moderation evasion and 95% to 100% for task manipulation. The authors also explore cross-view consistency checks as a mitigation; the supplied abstract does not report their effectiveness.

### Source excerpt

arXiv:2611.25469v1 Announce Type: cross Abstract: An image can pass automated safety checks while showing harmful content to its human viewer. We introduce DRJanus, a visual camouflage attack that exploits the gap between standard dynamic range (SDR) processing and high dynamic range (HDR) display. Automated systems often read only an image's SDR base, whereas HDR viewers reconstruct its appearance using an auxiliary gain map. DRJanus inverts this reconstruction to embed two independently chosen color images in a single JPEG, without querying or optimizing against the target model. It preserves the chosen SDR base exactly and reconstructs HDR targets at an average PSNR of 43.03 dB across 400 images. On a benchmark of 100 sexual-content images, DRJanus evades detection by all six tested commercial moderation services and nine vision-language models. Across five agents, each evaluated on 120 task cases per threat setting, attack success ranges from 31.67% to 100% for moderation evasion and from 95% to 100% for task manipulation. We also explore cross-view consistency checks as a mitigation. These findings expose a gap in visual trust: securing what a machine reads does not secure what a user sees. A demo is available at https://qshen0629.github.io/DRJanus/

## Beyond Entity Recognition: Vision-Centric Knowledge Visual Question Answering

DevFeed: [Beyond Entity Recognition: Vision-Centric Knowledge Visual Question Answering](<https://devfeed.tech/articles/beyond-entity-recognition-vision-centric-knowledge-visual-question-answering-145570.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25680>)

Author: Jiatong Ma, Yuchen Liu, Longteng Guo, Yichen Yuan, Yuze Wu, Dongze Hao, Ru Zhen, Yanhao Zhang, Haonan Lu, Jing Liu

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [vlm](<https://devfeed.tech/topics/vlm.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [Computer vision](<https://devfeed.tech/topics/computer-vision.md>), [knowledge-engineering](<https://devfeed.tech/topics/knowledge-engineering.md>)

Tags: [acquisition](<https://devfeed.tech/tags/acquisition.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-retrieval](<https://devfeed.tech/tags/agentic-retrieval.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [bottleneck](<https://devfeed.tech/tags/bottleneck.md>), [complex](<https://devfeed.tech/tags/complex.md>), [cross](<https://devfeed.tech/tags/cross.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [external](<https://devfeed.tech/tags/external.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [knowledge-base](<https://devfeed.tech/tags/knowledge-base.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [large-language-models](<https://devfeed.tech/tags/large-language-models.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [open-web](<https://devfeed.tech/tags/open-web.md>), [question](<https://devfeed.tech/tags/question.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [retrieval](<https://devfeed.tech/tags/retrieval.md>), [scale](<https://devfeed.tech/tags/scale.md>), [vision](<https://devfeed.tech/tags/vision.md>)

### AI overview

VC-VQA benchmarks vision-centric knowledge reasoning in multimodal large language models, including how they acquire and use visual evidence across multiple reasoning stages. It distinguishes internal from external images and pre-retrieval from post-retrieval visual information; tasks include retrieving external visual evidence, spotting clues before retrieval, and localizing, verifying, or matching retrieved knowledge to an input image. Its controlled multimodal knowledge base provides traceable textual and visual evidence. The authors evaluate 21 models across four settings: no external knowledge, provided evidence, knowledge-base retrieval, and open-web search. They report that models still struggle, with fine-grained visual knowledge acquisition the main bottleneck, while agentic retrieval points to the value of multi-step planning and cross-modal integration.

### Source excerpt

arXiv:2611.25680v1 Announce Type: cross Abstract: We propose VC-VQA, a large-scale benchmark for vision-centric knowledge reasoning, aiming to systematically evaluate the capabilities of Multimodal Large Language Models (MLLMs) in fine-grained visual knowledge acquisition and complex multi-stage reasoning. Unlike existing benchmarks that primarily treat visual information as entity recognition and rely on text-based external knowledge, VC-VQA explicitly models the role of visual information by considering both its source (internal vs. external images) and stage of involvement (pre-retrieval vs. post-retrieval). Specifically, VC-VQA requires models to retrieve external visual evidence, identify visual clues before retrieval, and localize, verify, or match retrieved knowledge against the internal image. We further construct a controlled multimodal knowledge base with traceable textual and visual evidence for each sample. We evaluate 21 representative MLLMs under four settings: no external knowledge, with provided evidence, knowledge base retrieval, and open-web search. Experimental results lead to three key findings: current models still struggle with vision-centric knowledge reasoning; fine-grained visual knowledge acquisition remains the primary bottleneck; and agentic retrieval highlights the value of multi-step planning and cross-modal knowledge integration. Overall, VC-VQA provides a more challenging benchmark for evaluating vision-centric knowledge reasoning in MLLMs. Our code and dataset are available at https://github.com/CASIA-IVA-Lab/VC-VQA.

## Where Global Context and Weight Sharing Work in Compact Vision Backbones

DevFeed: [Where Global Context and Weight Sharing Work in Compact Vision Backbones](<https://devfeed.tech/articles/where-global-context-and-weight-sharing-work-in-compact-vision-backbones-142656.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23446>)

Author: Michael Schroeder

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [ImageNet](<https://devfeed.tech/topics/imagenet.md>)

Tags: [attention](<https://devfeed.tech/tags/attention.md>), [batchnorm](<https://devfeed.tech/tags/batchnorm.md>), [block](<https://devfeed.tech/tags/block.md>), [blocks](<https://devfeed.tech/tags/blocks.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [classification](<https://devfeed.tech/tags/classification.md>), [computer-vision](<https://devfeed.tech/tags/computer-vision.md>), [convolution](<https://devfeed.tech/tags/convolution.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [efficientnet](<https://devfeed.tech/tags/efficientnet.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [global](<https://devfeed.tech/tags/global.md>), [global-context](<https://devfeed.tech/tags/global-context.md>), [lightweight](<https://devfeed.tech/tags/lightweight.md>), [scale](<https://devfeed.tech/tags/scale.md>), [self-attention](<https://devfeed.tech/tags/self-attention.md>), [sharing](<https://devfeed.tech/tags/sharing.md>), [stage](<https://devfeed.tech/tags/stage.md>), [vision](<https://devfeed.tech/tags/vision.md>), [weight](<https://devfeed.tech/tags/weight.md>), [work](<https://devfeed.tech/tags/work.md>)

### AI overview

Controlled EfficientNet-B0 experiments on ImageNet-1k suggest that global context helps when added to an intact MBConv trunk: zero-gated residual attention adapters at P3 (windowed) and P4/P5 (global) scored 73.88% versus 72.92% for a matched plain-B0 control, with attention amplitude increasing at deeper stages. A BiFPN wrapper alone scored 69.99-71.81%, below plain B0's 72.86%; adding stage attention raised it to 73.15%. For weight sharing, averaging kernels collapsed accuracy to 17.03%, while tying one template across a stage reduced parameters 27.6% but reached only 66.10%. Alternating two shared templates in Stage 6 cut parameters 10.9% (582K weights) and reached 70.92% after one recovery epoch, 1.84 points below the unshared control. Findings are limited to the described experiments and training schedule.

### Source excerpt

arXiv:2611.23446v1 Announce Type: cross Abstract: Compact convolutional backbones such as EfficientNet achieve strong parameter efficiency through local depthwise separable convolutions, yet two structural questions remain central to hybrid design: where does global self-attention improve recognition over pure convolution, and can repeated blocks share weights without losing representational capacity? We investigate both questions on ImageNet-1k (50,000-image validation) using controlled EfficientNet-B0 experiments. First, we show that wrapping a classification backbone in multi-scale feature pyramids (BiFPN) fails to outperform a plain MBConv trunk (69.99% - 71.81% vs. 72.86% plain B0) until paired with stage attention (73.15%). In contrast, keeping the MBConv trunk intact and inserting lightweight zero-gated residual self-attention adapters at stages P3 (28x28 windowed), P4 (14x14 global), and P5 (7x7 global) reaches 74.02% standalone and outperforms a bit-for-bit matched plain-B0 control by +0.954 percentage points (73.88% vs. 72.92%) across an 88-epoch stateless replay schedule, with learned attention amplitude rising monotonically with semantic depth (27.2% at P3 -> 45.4% at P4 -> 66.3% at P5). Second, we evaluate recurrent weight sharing across repeated MBConv blocks with pass-specific BatchNorm statistics (+1,536 parameters per pass). While post-hoc kernel averaging degrades accuracy to 17.03% and single-template stage tying (-27.6% parameters) plateaus at 66.10%, alternating two shared block templates ([B1, A, B, A]) in Stage 6 cuts total model parameters by 10.9% (582K weights) and recovers by +10.82 points in a single epoch to 70.92% (-1.84 points vs. unshared control).

## PANC: Prior-Aware Normalized Cut for Training-Free In-Context Segmentation

DevFeed: [PANC: Prior-Aware Normalized Cut for Training-Free In-Context Segmentation](<https://devfeed.tech/articles/panc-prior-aware-normalized-cut-for-training-free-in-context-segmentation-142505.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2602.06912>)

Author: Juan Guti\\'errez, Victor Guti\\'errez-Garc\\'ia, Jos\\'e Luis Blanco-Murillo

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [Image](<https://devfeed.tech/topics/image.md>)

Tags: [affinity](<https://devfeed.tech/tags/affinity.md>), [anchor](<https://devfeed.tech/tags/anchor.md>), [clicks](<https://devfeed.tech/tags/clicks.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [external](<https://devfeed.tech/tags/external.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [graph](<https://devfeed.tech/tags/graph.md>), [image](<https://devfeed.tech/tags/image.md>), [masks](<https://devfeed.tech/tags/masks.md>), [nodes](<https://devfeed.tech/tags/nodes.md>), [object](<https://devfeed.tech/tags/object.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [semantic](<https://devfeed.tech/tags/semantic.md>), [sign](<https://devfeed.tech/tags/sign.md>), [stabilization](<https://devfeed.tech/tags/stabilization.md>), [structure](<https://devfeed.tech/tags/structure.md>), [study](<https://devfeed.tech/tags/study.md>), [target](<https://devfeed.tech/tags/target.md>), [token](<https://devfeed.tech/tags/token.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

PANC uses a small set of foreground and background tokens from annotated context images to guide training-free segmentation of a target concept, without fine-tuning or target-image clicks. It adds two labeled anchor nodes to a frozen ViT affinity graph; these softly bias the Normalized Cut eigenspace without clamping target tokens or replacing their affinity structure. Prior-aware sign stabilization and thresholding produce deterministic masks. The abstract reports gains over architecture-matched no-prior token-graph baselines across saliency and low-semantic domains, with the largest gains when class-agnostic saliency is ambiguous or weak.

### Source excerpt

arXiv:2602.06912v3 Announce Type: replace-cross Abstract: We study training-free in-context segmentation with externally annotated context priors: given a target image and a small set of foreground/background tokens extracted from annotated context images, the goal is to segment the corresponding target concept without training, fine-tuning, or target-image clicks. Existing ViT-token spectral methods such as TokenCut are class-agnostic and can select the wrong object in multi-object scenes, or become unstable in low-semantic texture-dominated domains. We propose PANC, a prior-aware spectral segmentation pipeline that injects external labeled tokens into a frozen ViT affinity graph through two foreground/background anchor nodes. The anchors do not clamp target-image tokens; instead, they softly bias the Normalized Cut eigenspace while preserving the target-image affinity structure. PANC further uses prior-aware sign stabilization and prior-driven thresholding to obtain deterministic masks. Across heterogeneous saliency datasets and homogeneous low-semantic domains, PANC improves over architecture-matched no-prior token-graph baselines, with the largest gains in settings where class-agnostic saliency is ambiguous or weak.

## ProtoBridge-CLIP: Observation-Masked Finding Prototypes for Medical Image--Report Representation Learning

DevFeed: [ProtoBridge-CLIP: Observation-Masked Finding Prototypes for Medical Image--Report Representation Learning](<https://devfeed.tech/articles/protobridge-clip-observation-masked-finding-prototypes-for-medical-image-report-representation-learning-142738.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24002>)

Author: Jinrui Wu, Yun Peng

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Medical imaging](<https://devfeed.tech/topics/medical-imaging.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [vlm](<https://devfeed.tech/topics/vlm.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [alignment](<https://devfeed.tech/tags/alignment.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [clip](<https://devfeed.tech/tags/clip.md>), [component](<https://devfeed.tech/tags/component.md>), [concepts](<https://devfeed.tech/tags/concepts.md>), [contrastive-learning](<https://devfeed.tech/tags/contrastive-learning.md>), [controls](<https://devfeed.tech/tags/controls.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [framework](<https://devfeed.tech/tags/framework.md>), [global](<https://devfeed.tech/tags/global.md>), [image](<https://devfeed.tech/tags/image.md>), [inspect](<https://devfeed.tech/tags/inspect.md>), [learning](<https://devfeed.tech/tags/learning.md>), [local](<https://devfeed.tech/tags/local.md>), [localization](<https://devfeed.tech/tags/localization.md>), [medical](<https://devfeed.tech/tags/medical.md>), [objectives](<https://devfeed.tech/tags/objectives.md>), [prototypes](<https://devfeed.tech/tags/prototypes.md>), [report](<https://devfeed.tech/tags/report.md>), [representation-learning](<https://devfeed.tech/tags/representation-learning.md>)

### AI overview

ProtoBridge-CLIP separates affirmed, negated, uncertain, and unmentioned report concepts during medical image-report learning. It combines symmetric global image-report alignment with finding supervision that aligns text-derived finding prototypes to local visual tokens, excluding uncertain and unmentioned concepts from direct finding labels, plus a soft relation objective over affirmed concepts. Across three CT and five chest-radiograph benchmarks, it reports macro-AUROCs ranging from 66.5% to 91.4%; the findings and relation objectives consistently improve on matched global-only controls. The abstract also reports ablations and retrieval, localization, robustness, calibration, and per-finding analyses, but does not provide their detailed results.

### Source excerpt

arXiv:2611.24002v1 Announce Type: cross Abstract: Radiology reports distinguish affirmed findings, explicit negations, and unmentioned concepts, but conventional global image--report contrastive learning often compresses these different evidence states into a single study-level target. We introduce \model{}, which combines symmetric global image--report alignment with observation-masked finding supervision and a soft relation objective over affirmed report concepts. Text-derived finding prototypes are aligned with local visual tokens, while uncertain and unmentioned concepts are excluded from direct finding labels. We evaluate the framework on three CT benchmarks and five chest-radiograph benchmarks, together with component ablations, paired and clinical retrieval, localization, perturbation robustness, calibration, and per-finding analyses. \model{} achieves macro-AUROC values of 73.6%, 66.5\%, and 68.8% on CT-RATE, RAD-ChestCT, and INSPECT, and 80.2%, 91.4%, 78.5%, 86.5%, and 90.8% on ChestXray14, CheXpert, PadChest, RSNA, and SIIM-ACR, respectively. Relative to the matched global-only controls, the findings and relation objectives yield consistent gains across the evaluated datasets. These results demonstrate the practical value of treating observed and unobserved report concepts differently during medical image--text representation learning.

## DetectViT: Test-time Backdoor Detection for Vision Transformers via Inter-Head Attention Discrepancy

DevFeed: [DetectViT: Test-time Backdoor Detection for Vision Transformers via Inter-Head Attention Discrepancy](<https://devfeed.tech/articles/detectvit-test-time-backdoor-detection-for-vision-transformers-via-inter-head-attention-discrepancy-143135.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23770>)

Author: Siquan Huang, Yijiang Li, Xingfu Yan, Ningzhi Gao, Leyu Shi, Ying Gao

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Transformer](<https://devfeed.tech/topics/transformer.md>), [vlm](<https://devfeed.tech/topics/vlm.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [attention](<https://devfeed.tech/tags/attention.md>), [backdoor](<https://devfeed.tech/tags/backdoor.md>), [classification](<https://devfeed.tech/tags/classification.md>), [clip](<https://devfeed.tech/tags/clip.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [deit](<https://devfeed.tech/tags/deit.md>), [detection](<https://devfeed.tech/tags/detection.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [entropy](<https://devfeed.tech/tags/entropy.md>), [false-positive](<https://devfeed.tech/tags/false-positive.md>), [false-positive-rate](<https://devfeed.tech/tags/false-positive-rate.md>), [hijacking](<https://devfeed.tech/tags/hijacking.md>), [image](<https://devfeed.tech/tags/image.md>), [image-classification](<https://devfeed.tech/tags/image-classification.md>), [inference](<https://devfeed.tech/tags/inference.md>), [llava](<https://devfeed.tech/tags/llava.md>), [model](<https://devfeed.tech/tags/model.md>), [test](<https://devfeed.tech/tags/test.md>), [transformers](<https://devfeed.tech/tags/transformers.md>), [vision](<https://devfeed.tech/tags/vision.md>), [vision-transformers](<https://devfeed.tech/tags/vision-transformers.md>)

### AI overview

DetectViT detects test-time backdoors in Vision Transformers by measuring two attention anomalies: unusually high spatial overlap between heads and unusually variable attention concentration across heads. It combines these as inter-head consistency and entropy-variance scores, setting thresholds with a small OOD calibration set. The method needs no training data, model outputs, or prior trigger knowledge, and reuses attention weights already computed at inference. The abstract reports results across four attacks, DeiT, CLIP, and LLaVA, and classification and captioning tasks; with 64-128 calibration samples, it reports 100% true positive and 0.17% false positive rates on BadCLIP. These are abstract-reported results, not evidence of performance beyond the stated settings.

### Source excerpt

arXiv:2611.23770v1 Announce Type: cross Abstract: Vision Transformers (ViTs) now underpin the visual encoders of most multimodal models, but this success comes at a cost: relying on third-party pretrained checkpoints leaves these systems open to backdoor attacks. Existing test-time detection methods rely on either the predicted label or intermediate embeddings, limiting their applicability across diverse tasks and incurring substantial computational overhead. In this work, we study how backdoor triggers affect the multi-head attention mechanism of ViTs and identify two distinct anomalous patterns: $\textit{complete attention hijacking}$, causing the attention to exhibit abnormally high inter-head spatial overlap, and $\textit{partial attention hijacking}$, causing an abnormally large variance in the concentration of the attention distribution. Motivated by these findings, we propose $\textbf{DetectViT}$, a test-time backdoor detection method that requires neither training data, model outputs, nor any prior knowledge of the trigger. DetectViT quantifies the two hijacking phenomena via an inter-head consistency score and an inter-head entropy variance score, with thresholds estimated from a small out-of-distribution (OOD) calibration set drawn from an arbitrary source. Across four representative attacks, diverse architectures (DeiT, CLIP, and LLaVA), and downstream tasks (image classification and captioning), DetectViT substantially outperforms all baselines with as few as 64--128 OOD calibration samples, achieving a true positive rate of $100\%$ and a false positive rate of $0.17\%$ on BadCLIP. Notably, it introduces negligible additional overhead by relying solely on attention weights already computed during inference. Code is available at: https://github.com/siquanhuang/DetectViT.

## AdaS-ViT: Test-Time Adaptive Routing for Spiking Vision Transformers with Ultra-High Energy Efficiency

DevFeed: [AdaS-ViT: Test-Time Adaptive Routing for Spiking Vision Transformers with Ultra-High Energy Efficiency](<https://devfeed.tech/articles/adas-vit-test-time-adaptive-routing-for-spiking-vision-transformers-with-ultra-high-energy-efficiency-142696.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23728>)

Author: Zecheng Hao, Zihan Huang, Shenghao Xie, Jiyuan Zhang, Zhaofei Yu, Tiejun Huang

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Transformer](<https://devfeed.tech/topics/transformer.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [Green Software](<https://devfeed.tech/topics/green-software.md>), [Transformers](<https://devfeed.tech/topics/transformers.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [adaptive](<https://devfeed.tech/tags/adaptive.md>), [all](<https://devfeed.tech/tags/all.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [attention](<https://devfeed.tech/tags/attention.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [boundary](<https://devfeed.tech/tags/boundary.md>), [building](<https://devfeed.tech/tags/building.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-ne](<https://devfeed.tech/tags/cs-ne.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [energy](<https://devfeed.tech/tags/energy.md>), [energy-efficiency](<https://devfeed.tech/tags/energy-efficiency.md>), [routing](<https://devfeed.tech/tags/routing.md>), [test](<https://devfeed.tech/tags/test.md>), [time](<https://devfeed.tech/tags/time.md>), [transformers](<https://devfeed.tech/tags/transformers.md>), [vision](<https://devfeed.tech/tags/vision.md>), [vision-transformers](<https://devfeed.tech/tags/vision-transformers.md>), [vit](<https://devfeed.tech/tags/vit.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

AdaS-ViT reduces redundant spiking computation by adapting firing precision at test time across temporal, spatial and channel dimensions. Its AdaE-IF module uses attention scores to route salient regions to fine-grained firing patterns and the rest to coarser, lower-energy patterns; S-GSA and S-LGU add selective spike emission at token and channel granularity. A progressive spatio-temporal training curriculum is used to stabilize the routing masks and firing precision. The abstract reports competitive results across benchmarks and 1.02-2.45 mJ per image on ImageNet-1k, with an accuracy-per-energy ratio of 32x-72x.

### Source excerpt

arXiv:2611.23728v1 Announce Type: cross Abstract: Spiking Neural Networks (SNNs) promise event-driven and energy-efficient vision, yet existing spiking vision transformers (S-ViTs) largely confine spiking tokens to fixed and uniform firing schemes along the temporal, spatial and channel dimensions, whereas suppressing the computational redundancy thus incurred calls for adaptive firing mechanisms that operate jointly across all three axes at test time. To this end, we propose the Adaptive Efficient Integrate-and-Fire (AdaE-IF) model, which consults the attention score of each spiking token and routes salient regions to fine-grained firing patterns while assigning energy-frugal coarse patterns to the remaining regions. Building on AdaE-IF, we further introduce Spiking Gated Self-Attention (S-GSA) and Spiking Local-Gated Unit (S-LGU), which enforce selective spike emission at token and channel granularity. We integrate the aforementioned mechanisms and modules into AdaS-ViT, a test-time adaptive-routing architecture trained with a progressive spatio-temporal curriculum that stabilizes the co-adaptation of firing precision and routing masks. Extensive experiments across benchmarks spanning diverse data scales and modalities demonstrate that AdaS-ViT consistently achieves highly competitive performance. Notably, on ImageNet-1k it consumes only 1.02 to 2.45 millijoules per image and attains an accuracy-per-energy ratio of 32$\times$ to 72$\times$, operating in an economical inference regime and thereby pushing the energy-efficiency boundary further.

## Enhancing Active Learning Performance with Initial Data Selection Based on Foundation Models

DevFeed: [Enhancing Active Learning Performance with Initial Data Selection Based on Foundation Models](<https://devfeed.tech/articles/enhancing-active-learning-performance-with-initial-data-selection-based-on-foundation-models-142789.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24384>)

Author: Mohammad Mirzaee, Mehrshad Fallah Estalkhzir, Mohammad Reza Mohammadi

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Model Development](<https://devfeed.tech/topics/model-development.md>), [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [clustering](<https://devfeed.tech/topics/clustering.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [clustering](<https://devfeed.tech/tags/clustering.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [data](<https://devfeed.tech/tags/data.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [detection](<https://devfeed.tech/tags/detection.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [foundation-model](<https://devfeed.tech/tags/foundation-model.md>), [foundation-models](<https://devfeed.tech/tags/foundation-models.md>), [framework](<https://devfeed.tech/tags/framework.md>), [image](<https://devfeed.tech/tags/image.md>), [labeling](<https://devfeed.tech/tags/labeling.md>), [learning](<https://devfeed.tech/tags/learning.md>), [model-performance](<https://devfeed.tech/tags/model-performance.md>), [object-detection](<https://devfeed.tech/tags/object-detection.md>), [performance](<https://devfeed.tech/tags/performance.md>), [pool](<https://devfeed.tech/tags/pool.md>), [resource](<https://devfeed.tech/tags/resource.md>), [results](<https://devfeed.tech/tags/results.md>), [robustness](<https://devfeed.tech/tags/robustness.md>), [sampling](<https://devfeed.tech/tags/sampling.md>)

### AI overview

The framework uses YOLO-World feature clustering and objectness scores to select and refine the initial labeled pool for active object detection, instead of uncertainty sampling in the first cycle; later cycles use consistency-based sampling. Evaluated on Aquarium and NWPU VHR-10, it reports an average 6.6% increase in image-based mAP50 on NWPU, while image refinement improves object-based evaluation by approximately 11%. The abstract also reports gains in model performance and labeling time, but does not quantify the latter.

### Source excerpt

arXiv:2611.24384v1 Announce Type: cross Abstract: In active object detection, selecting the initial labeled pool is essential, especially in complex datasets with expensive labeling and low resource budget. This study presents an active learning framework designed to optimize the initial training cycle of active learning for the object detection task. Our approach diverges from traditional methods by utilizing feature-based clustering from the YOLO-World model and refining images based on objectness scores rather than relying on uncertainty sampling in the initial cycle. This approach focuses on finding samples with various features and emphasizes training in areas where the foundation model is most confident. Doing so helps the model better understand different types of objects. For subsequent cycles, we employ a consistency-based sampling technique to maintain the robustness of the training process. To evaluate our method, we use two complex datasets called Aquarium and NWPU VHR-10, which are well known for their diverse object sizes and challenging conditions. In the NWPU dataset, our method increases mAP50 by an average of 6.6% in image-based evaluation. Moreover, the image refinement method shows an improvement of approximately 11% in object-based evaluation. The results show improvements in model performance and labeling time, demonstrating the advantages of using foundation models in early cycles of active learning. An implementation of the algorithm is available at https://github.com/Mhrshd-fllh/ALBoost-FM

## VideoACE: Benchmarking Agentic Video Understanding through High-Frame-Rate Actions

DevFeed: [VideoACE: Benchmarking Agentic Video Understanding through High-Frame-Rate Actions](<https://devfeed.tech/articles/videoace-benchmarking-agentic-video-understanding-through-high-frame-rate-actions-142763.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24184>)

Author: Luoyuan Zhang, Xuanyu Zheng, Yancheng Long, Haoyang Xu, Bin Wen, Tingting Gao

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [actions](<https://devfeed.tech/tags/actions.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-mm](<https://devfeed.tech/tags/cs-mm.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [interaction](<https://devfeed.tech/tags/interaction.md>), [spatial](<https://devfeed.tech/tags/spatial.md>), [temporal](<https://devfeed.tech/tags/temporal.md>), [testbed](<https://devfeed.tech/tags/testbed.md>), [tools](<https://devfeed.tech/tags/tools.md>), [video](<https://devfeed.tech/tags/video.md>)

### AI overview

VideoACE evaluates whether models improve video understanding by interacting with high-frame-rate action videos, where rapid, subtle changes can make targeted observation valuable. It includes 1,000 multiple-choice questions across six fine-grained action categories, with temporal evidence annotations and user hints screened for answer leakage. Its environment lets models adjust temporal coverage, sampling density and spatial detail, revisit evidence, and revise judgments with simulated user feedback. The study compares single-pass observation, autonomous tool use, and tool use with feedback across five models, finding distinct evidence-gathering and interaction patterns. The abstract presents VideoACE as a testbed beyond fixed-observation evaluation; it does not report specific model scores or identify which patterns performed best.

### Source excerpt

arXiv:2611.24184v1 Announce Type: cross Abstract: Agentic video understanding requires models to actively interact with video environments to gather evidence and form judgments. Evaluating this capability therefore requires assessing whether models can translate opportunities for interaction into better understanding. We introduce VideoACE, a benchmark for assessing video understanding under conditions that support sustained interaction. We focus on high-frame-rate action videos, where rapid, subtle changes make targeted observation especially valuable. The benchmark comprises 1,000 multiple-choice questions spanning six categories of fine-grained action understanding, accompanied by temporal evidence annotations and user hints screened for answer leakage. Its evaluation environment combines video inspection tools with a user simulator, allowing models to adjust temporal coverage, sampling density, and spatial detail, and to revisit evidence and revise their judgments under simulated user feedback. We evaluate single-pass observation, autonomous tool use, and tool use combined with user feedback to examine how effectively models exploit these opportunities. Evaluation of five models reveals distinct patterns in how they acquire evidence and respond to opportunities for further interaction. These findings establish VideoACE as a testbed for examining agentic video understanding beyond performance under a fixed observation setting.

## Inherited Adapter Experts for Replay-Free Domain-Incremental Learning

DevFeed: [Inherited Adapter Experts for Replay-Free Domain-Incremental Learning](<https://devfeed.tech/articles/inherited-adapter-experts-for-replay-free-domain-incremental-learning-142758.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24145>)

Author: Gyutae Oh, Hyunmin Kong, Jitae Shin

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Mixture of Experts (MoE)](<https://devfeed.tech/topics/mixture-of-experts-moe.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Training AI Models](<https://devfeed.tech/topics/training-ai-models.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [adapter](<https://devfeed.tech/tags/adapter.md>), [adapters](<https://devfeed.tech/tags/adapters.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [computer-vision](<https://devfeed.tech/tags/computer-vision.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [deepfake](<https://devfeed.tech/tags/deepfake.md>), [domain](<https://devfeed.tech/tags/domain.md>), [domains](<https://devfeed.tech/tags/domains.md>), [experts](<https://devfeed.tech/tags/experts.md>), [free](<https://devfeed.tech/tags/free.md>), [incremental](<https://devfeed.tech/tags/incremental.md>), [learning](<https://devfeed.tech/tags/learning.md>), [points](<https://devfeed.tech/tags/points.md>), [router](<https://devfeed.tech/tags/router.md>), [vit](<https://devfeed.tech/tags/vit.md>)

### AI overview

Lamarck addresses replay-free domain-incremental learning by strengthening per-domain experts rather than training a router: the abstract reports that a training-free Mahalanobis router costs the analyzed learners under 2.7 points versus always choosing the own-domain expert. Lamarck adds adapters after each deep ViT block, improving final accuracy by 4.4 points on CDDB-Hard, and initializes each new expert from its predecessor at no extra cost. This is intended to reduce the cost when an image is routed to a later expert, since domains share classes. Against twelve frozen-backbone baselines under one protocol, Lamarck ranks first on CDDB-Hard by 5.2 points, second on DomainNet and OfficeHome, and highest in mean accuracy.

### Source excerpt

arXiv:2611.24145v1 Announce Type: cross Abstract: Replay-free domain-incremental learning trains a model on the same classes in successive domains without keeping earlier images. Per-domain learners freeze a pre-trained ViT and train an additional expert for each new domain: the small set of parameters and the classification head learned on that domain. Test images carry no domain label, so a router selects one expert for each image. Recent work trains this router. However, a training-free Mahalanobis router costs the analyzed learners under 2.7 points of final accuracy relative to always selecting each image's own-domain expert. The experts, not this router, limit accuracy. Our learner, Lamarck, therefore uses this router and strengthens the experts. Its adapters, small trainable layers that correct every image token after each deep ViT block, add 4.4 points of final accuracy on the deepfake benchmark CDDB-Hard. In addition, each expert is initialized from its predecessor at no extra cost: the domains share their classes, so an image misrouted to a later expert loses less. Against twelve frozen-backbone baselines under one protocol, Lamarck is first on CDDB-Hard by 5.2 points, second on DomainNet and OfficeHome, and highest in mean accuracy.

## MonitorVLM:A Vision Language Framework for Safety Violation Detection in Mining Operations

DevFeed: [MonitorVLM:A Vision Language Framework for Safety Violation Detection in Mining Operations](<https://devfeed.tech/articles/monitorvlm-a-vision-language-framework-for-safety-violation-detection-in-mining-operations-142499.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2510.03666>)

Author: Jiang Wu, Sichao Wu, Yinsong Ma, Guangyuan Yu, Haoyuan Xu, Lifang Zheng, Jingliang Duan

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Video Analytics](<https://devfeed.tech/topics/video-analytics.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [action-recognition](<https://devfeed.tech/tags/action-recognition.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [detection](<https://devfeed.tech/tags/detection.md>), [experimental](<https://devfeed.tech/tags/experimental.md>), [filter](<https://devfeed.tech/tags/filter.md>), [framework](<https://devfeed.tech/tags/framework.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [inference](<https://devfeed.tech/tags/inference.md>), [inspection](<https://devfeed.tech/tags/inspection.md>), [interface](<https://devfeed.tech/tags/interface.md>), [language](<https://devfeed.tech/tags/language.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [large](<https://devfeed.tech/tags/large.md>), [latency](<https://devfeed.tech/tags/latency.md>), [lightweight](<https://devfeed.tech/tags/lightweight.md>), [mining](<https://devfeed.tech/tags/mining.md>), [models](<https://devfeed.tech/tags/models.md>), [module](<https://devfeed.tech/tags/module.md>), [safety](<https://devfeed.tech/tags/safety.md>), [vision](<https://devfeed.tech/tags/vision.md>)

### AI overview

MonitorVLM detects mining safety violations from surveillance video using a domain-specific dataset of 9,000 VQA samples covering 40 regulations. Its clause filter selects relevant rules and reportedly cuts inference latency by 13.56% while maintaining accuracy; its behavior magnifier enlarges worker regions, adding 3.45% precision and 8.62% recall. Against an unfine-tuned 72B vision-language baseline, the framework reports gains of 22.01% in precision, 34.22% in recall, and 28.37% in F1. A web interface supports automatic violation reports with video timestamps. These are reported abstract results; the excerpt provides no further evaluation details.

### Source excerpt

arXiv:2510.03666v3 Announce Type: replace-cross Abstract: Industrial accidents, particularly in high-risk domains such as surface and underground mining, are frequently caused by unsafe worker behaviors. Traditional manual inspection remains labor-intensive, error-prone, and insufficient for large-scale, dynamic environments, highlighting the urgent need for intelligent and automated safety monitoring. In this paper, we present MonitorVLM, a novel vision--language framework designed to detect safety violations directly from surveillance video streams. MonitorVLM introduces three key innovations: (1) a domain-specific violation dataset comprising 9,000 vision--question--answer (VQA) samples across 40 high-frequency mining regulations, enriched with augmentation and auxiliary detection cues; (2) a clause filter (CF) module that dynamically selects the Top-$K$ most relevant clauses, reducing inference latency by 13.56% while maintaining accuracy; and (3) a behavior magnifier (BM) module that enhances worker regions to improve fine-grained action recognition, yielding additional gains of 3.45% in precision and 8.62% in recall. Experimental results demonstrate that MonitorVLM significantly outperforms baseline vision--language models, achieving improvements of 22.01% in precision, 34.22\% in recall, and 28.37% in F1 score over the 72B unfine-tuned baseline. A lightweight web-based interface further integrates MonitorVLM into practical workflows, enabling automatic violation reporting with video timestamping. This study highlights the potential of multimodal large models to enhance occupational safety monitoring in mining and beyond.

## Action Recognition vs. Detection + Tracking: When to Use Which

DevFeed: [Action Recognition vs. Detection + Tracking: When to Use Which](<https://devfeed.tech/articles/action-recognition-vs-detection-tracking-when-to-use-which-143831.md>)

Original publisher: [Read original article](<https://roboflow.com/blog/when-to-use-action-recognition>)

Author: Erik Kokalj

Published: 2026-11-18T22:22:42Z

Content type: comparison

Language: en

Sources: [Roboflow](<https://devfeed.tech/sources/roboflow-blog.md>)

Topics: [Video Analytics](<https://devfeed.tech/topics/video-analytics.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [rule-engine](<https://devfeed.tech/topics/rule-engine.md>)

Tags: [action-recognition](<https://devfeed.tech/tags/action-recognition.md>), [basketball](<https://devfeed.tech/tags/basketball.md>), [bytetrack](<https://devfeed.tech/tags/bytetrack.md>), [cosmos](<https://devfeed.tech/tags/cosmos.md>), [cosmos-3](<https://devfeed.tech/tags/cosmos-3.md>), [detection](<https://devfeed.tech/tags/detection.md>), [f1](<https://devfeed.tech/tags/f1.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [fps](<https://devfeed.tech/tags/fps.md>), [motion](<https://devfeed.tech/tags/motion.md>), [tracking](<https://devfeed.tech/tags/tracking.md>), [video](<https://devfeed.tech/tags/video.md>), [workout](<https://devfeed.tech/tags/workout.md>)

### AI overview

The best approach depends on what the video question asks. Action recognition is suited to brief, subtle actions that require several frames: in the F1 test, Cosmos 3 Edge initially missed a quarter-second jack lift at 4 fps, but at 8 fps it became the strongest pit-stop system. In basketball, it caught all five dribbles, while a detector-based dribble rule caught one; it learned made shots from 13 examples. Detection plus ByteTrack is useful for visible objects and locations: it tracked tire removal better than action recognition, but missed jack lifts because the car barely moved and the crew obscured the jack. For body-pose tasks, a pretrained RF-DETR keypoint model plus simple geometry counted workout reps without training; action recognition often identified the exercise but merged repeated movements into one segment. The excerpt recommends enough labeled clips for action recognition, higher sampling rates for actions lasting a second or less, and adding a counter for reps. Detection and tracking depend on reliable detections, and pose distinctions such as squats versus lunges need keypoints rather than plain boxes.

### Source excerpt

We built action recognition and detection or keypoints + tracking systems for six video tasks, from F1 pit stops to workout reps. Here is when each approach wins.

## Know When to Zoom: How Small Is Too Small for Vision-Language Models?

DevFeed: [Know When to Zoom: How Small Is Too Small for Vision-Language Models?](<https://devfeed.tech/articles/know-when-to-zoom-how-small-is-too-small-for-vision-language-models-140070.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22709>)

Author: Siddharth Vohra, Min Xu

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [vlm](<https://devfeed.tech/topics/vlm.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [computer-vision](<https://devfeed.tech/tags/computer-vision.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [vision](<https://devfeed.tech/tags/vision.md>), [vision-language-models](<https://devfeed.tech/tags/vision-language-models.md>), [zoom](<https://devfeed.tech/tags/zoom.md>)

### AI overview

Online vision-language services can shrink or reject large images before the model sees them, making small details unreadable. Across seven models from four vendors, the study finds that the smallest text read half the time rises from 6 pixels or less on small images to a predicted 10-13 pixels on large images at default settings. Tiling images into up to nine crops improves Claude and Gemini accuracy by 11.1-14.5 percentage points and raises GPT-6.1 Sol from 50.7% to 83.7% by avoiding rejections. A two-step zoom, where the model identifies a region before inspecting it, matches or beats tiling on Gemini and comes within 2.5 points on Claude using about 40% of the image tokens. Claude and Gemini both often admit uncertainty when detail is lost, but Gemini is more likely to answer confidently and incorrectly. On Claude, tiling only after an unsure initial reply nearly matches always tiling while using 62-70% of its image tokens.

### Source excerpt

arXiv:2611.22709v1 Announce Type: cross Abstract: Online vision-language models often miss small details in large images, because the service shrinks the image, or rejects it, before the model ever sees it. A car 20 pixels wide in an 8000 by 6000 drone photo reaches Claude about 6 pixels wide, and OpenAI's default rejects the photo outright. We measure how much detail survives for seven models from four vendors, using an eye exam of printed codes and 601 questions about small objects in real photos. The smallest text a model reads half the time grows from 6 pixels or less on a small image to a predicted 10 to 13 pixels on a 7680 by 4320 photo at default settings. Zooming wins much of the detail back. Splitting each photo into up to nine tiles raises accuracy by 11.1 to 14.5 percentage points for Claude and Gemini, and by avoiding rejections it lifts GPT-6.1 Sol from 50.7% to 83.7%. A two-step zoom, in which the model first names the region it needs, matches or beats tiles on Gemini and comes within 2.5 points of them on Claude, with about 40% of their image tokens. When a detail is lost, Claude and Gemini both say they cannot tell about half the time, but they fail differently in the rest: Gemini gives a confident wrong answer in 43% to 48% of these cases, Claude in only 5% to 9%. A check based on detail size, run before the call, misses most of the gain from tiles on Claude and Gemini, but Claude's own reply is a reliable trigger: tiling only after an unsure reply comes within 1.4 points of always tiling with 62% to 70% of the image tokens. We turn these findings into five rules for when an agent should zoom, and release the eye exam, every model answer and code at https://siddvoh.com/know-when-to-zoom.

## Benchmarking Deep Learning Models for Wheelchair Configuration Classification in Accessible Automated Vehicle Applications

DevFeed: [Benchmarking Deep Learning Models for Wheelchair Configuration Classification in Accessible Automated Vehicle Applications](<https://devfeed.tech/articles/benchmarking-deep-learning-models-for-wheelchair-configuration-classification-in-accessible-automated-vehicle-applications-139932.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21945>)

Author: Samir Ahmed, Shaaz Saleem, Devin Dhami, Wesley Maia, Angel Martinez-Sanchez, Parthib Roy, Jannatul Ferdous Srabonee, Ahmed Sabbir Arif, Ross Greer

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Dataset development](<https://devfeed.tech/topics/dataset-development.md>)

Tags: [accessible](<https://devfeed.tech/tags/accessible.md>), [automated](<https://devfeed.tech/tags/automated.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [classification](<https://devfeed.tech/tags/classification.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [deep-learning-models](<https://devfeed.tech/tags/deep-learning-models.md>), [models](<https://devfeed.tech/tags/models.md>), [vehicle](<https://devfeed.tech/tags/vehicle.md>), [wheelchair](<https://devfeed.tech/tags/wheelchair.md>)

### AI overview

The Wheelchair Configuration Image Dataset (WCID) provides 467 face-blurred images for evaluating wheelchair configuration classification, addressing datasets that label wheelchair presence without describing chair type. It includes manual or powered labels, chair subtype, and annotations for occupancy, viewpoint, setting, lighting, occlusion, propulsion, and control; its 273-image validation and 194-image test splits are used solely for evaluation. Three convolutional backbones and a frozen vision-language probe, trained on an external image dataset, are assessed for cross-dataset generalization and Gaussian-blur robustness. Convolutional models perform well on clean images, but strong blur reduces performance, leaving a gap before deployment. The abstract says code, labels, split assignments, trained weights, and face-blurred images are available.

### Source excerpt

arXiv:2611.21945v1 Announce Type: cross Abstract: Making automated vehicles accessible requires consideration of vulnerable road users, including wheelchair users. To offer a more tailored experience that supports independence for this group, vehicle design needs to accommodate the wide range of wheelchair configurations. However, wheelchairs are individualized, varying in frame geometry, propulsion, and visual appearance. Moreover, occupants or accessories may obscure visual cues. Existing mobility-aid datasets only label the presence of a wheelchair, rather than describing its specific configuration. To fill this gap, this paper introduces the Wheelchair Configuration Image Dataset (WCID), comprising 467 face-blurred images across 192 capture groups solely for evaluation. Each image is annotated with (i) a two-class label of manual or powered, (ii) a chair subtype, (iii) occupancy, viewpoint, setting, lighting, occlusion, propulsion, and control, with a 273-image validation split and 194-image test split. Three convolutional neural network backbones and a frozen vision-language probe are trained on an external image dataset and evaluated on WCID to measure cross-dataset generalization, while also testing robustness using Gaussian blur. Convolutional models do well on clean images, but performance drops under strong blur, leaving a gap before deployment. Code, split assignments, labels, trained weights, and the face-blurred images is available at: https://github.com/Mi3-Lab/MAID

## TAR: Temporal Anchor-Constrained Reasoning for Video Temporal Grounding

DevFeed: [TAR: Temporal Anchor-Constrained Reasoning for Video Temporal Grounding](<https://devfeed.tech/articles/tar-temporal-anchor-constrained-reasoning-for-video-temporal-grounding-139746.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2508.07683>)

Author: Chaohong Guo, Xun Mo, Yongwei Nie, Fei Ma, Xuemiao Xu, Chengjiang Long

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [Chain-of-thought](<https://devfeed.tech/topics/chain-of-thought.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Hallucination detection](<https://devfeed.tech/topics/hallucination-detection.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [anchor](<https://devfeed.tech/tags/anchor.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [grounding](<https://devfeed.tech/tags/grounding.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [temporal](<https://devfeed.tech/tags/temporal.md>), [video](<https://devfeed.tech/tags/video.md>)

### AI overview

TAR (Temporal Anchor-Constrained Reasoning) adds an auditable temporal anchor checkpoint to video temporal grounding. It uses the anchor to progressively refine chain-of-thought reasoning against visual evidence and recalibrate temporal predictions, aiming to keep intermediate reasoning aligned with the final segment prediction. The paper says this improves reasoning faithfulness and accuracy. It also describes bootstrapping high-quality reasoning data with a standard 7B model, avoiding reliance on ultra-large models. The abstract reports state-of-the-art performance, but provides no benchmark details or quantitative results.

### Source excerpt

arXiv:2508.07683v3 Announce Type: replace-cross Abstract: Video Temporal Grounding (VTG) aims to localize specific video segments corresponding to natural language queries. While recent Large Vision-Language Models (LVLMs) employ Reinforcement Learning to generate Chains-of-Thought (CoT), they typically rely solely on outcome-based supervision. Consequently, this often leads to hallucinations, where the reasoning process becomes disconnected from the visual content and the final prediction. Existing attempts to mitigate this by relying on external supervision from larger models or separate reward models are computationally expensive and prone to rigid patterns. To address these challenges, we propose TAR (Temporal Anchor-Constrained Reasoning), a framework that introduces the temporal anchor (T-anchor) as a transparent and auditable checkpoint mechanism. T-anchor enforces progressive refinement within the CoT, compelling the model to continuously ground its intermediate thoughts in visual evidence and iteratively calibrate temporal predictions, thereby significantly enhancing the faithfulness and autonomy of the reasoning process and final accuracy. Furthermore, we introduce a bootstrapping paradigm that automatically harvests high-quality CoT data using only a standard 7B model, eliminating the dependency on ultra-large models. Extensive experiments demonstrate that TAR achieves state-of-the-art performance and generates faithful, autonomous, and progressively refined reasoning traces.

## How Much of SAR Vehicle Recognition Is the Vehicle? Background Shortcuts, Angle Gaps and Partial Libraries on MSTAR and ATRNet-STAR

DevFeed: [How Much of SAR Vehicle Recognition Is the Vehicle? Background Shortcuts, Angle Gaps and Partial Libraries on MSTAR and ATRNet-STAR](<https://devfeed.tech/articles/how-much-of-sar-vehicle-recognition-is-the-vehicle-background-shortcuts-angle-gaps-and-partial-libraries-on-mstar-and-atrnet-star-139914.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21827>)

Author: Mousa Shaya

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [background](<https://devfeed.tech/tags/background.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [change](<https://devfeed.tech/tags/change.md>), [controls](<https://devfeed.tech/tags/controls.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [foundation](<https://devfeed.tech/tags/foundation.md>), [foundation-model](<https://devfeed.tech/tags/foundation-model.md>), [gap](<https://devfeed.tech/tags/gap.md>), [recognition](<https://devfeed.tech/tags/recognition.md>), [shortcuts](<https://devfeed.tech/tags/shortcuts.md>), [split](<https://devfeed.tech/tags/split.md>), [training](<https://devfeed.tech/tags/training.md>), [training-data](<https://devfeed.tech/tags/training-data.md>), [vehicle](<https://devfeed.tech/tags/vehicle.md>)

### AI overview

The reported accuracy on SAR vehicle recognition can depend heavily on background cues and the gap between training and test depression angles. With ConvNeXt-Small, removing the central vehicle region leaves 73.3% accuracy on MSTAR and 63.3% on ATRNet-STAR's 40-class SOC split. On ATRNet-STAR's scene-change split, full chips reach 22.8%, while vehicle-only inputs reach 30.8%; SARATR-X relies even more on background cues and does not resolve this scene-change result. With training data held fixed, accuracy falls from 57% at a 15° angle gap to 39% at 30° and 26% at 45°, and adding training angles does not help. At a 45° gap, classes missing training images at the test angle reach 9.0% when other classes have them, versus 22.4% when none do; most such images are assigned to known classes. The authors recommend reporting background-only and target-only controls alongside the angle gap.

### Source excerpt

arXiv:2611.21827v1 Announce Type: new Abstract: Deep networks score near 99% on the MSTAR vehicle benchmark, and the newer ATRNet-STAR benchmark was built to be harder. We ask how much of these scores comes from the vehicle itself. With a modern network (ConvNeXt-Small, 3 seeds) we first reproduce the known MSTAR result: with the central 96x96 pixels zeroed out, accuracy is still 73.3% (chance 10%). The same shortcut is present in ATRNet-STAR, which was collected across several scenes: with the vehicle removed, accuracy is 63.3% on the 40-class SOC split (chance 2.5%). Under the official scene-change split, the background stops helping and starts hurting. Accuracy is 22.8% on full chips and 30.8% when only the vehicle is shown. A SAR foundation model, SARATR-X, leans on the background more than ConvNeXt does (+4 to +6 points with the vehicle removed, in every seed) and does not solve the scene change. We then hold the amount of training data fixed and vary the depression angles used for training. Accuracy is set mainly by the gap to the nearest training angle: 57% at 15 degrees, 39% at 30 degrees and 26% at 45 degrees. Adding more training angles does not help. Finally, when most classes have training images at the test angle and a few do not, the missing classes do worse than when no class has them: 9.0% instead of 22.4% at a 45-degree gap. Most of their images are assigned to known classes. We recommend reporting background-only and target-only controls, and the angle gap, next to any accuracy figure. Code is public.

## Video recordings and play behaviour annotations of group-housed dairy calves on two Dutch farms

DevFeed: [Video recordings and play behaviour annotations of group-housed dairy calves on two Dutch farms](<https://devfeed.tech/articles/video-recordings-and-play-behaviour-annotations-of-group-housed-dairy-calves-on-two-dutch-farms-141671.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21706>)

Author: Haiyu Yang, Heidi Lesscher, Enhong Liu, Miel Hostens

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [human feedback](<https://devfeed.tech/topics/human-feedback.md>)

Tags: [annotations](<https://devfeed.tech/tags/annotations.md>), [behaviour](<https://devfeed.tech/tags/behaviour.md>), [computer](<https://devfeed.tech/tags/computer.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [play](<https://devfeed.tech/tags/play.md>), [recordings](<https://devfeed.tech/tags/recordings.md>), [video](<https://devfeed.tech/tags/video.md>)

### AI overview

The dataset provides video and individual calf behaviour annotations from group pens on two Dutch farms: 1,173 files spanning 651 hours across four roughly week-long camera deployments. Trained observers continuously coded an 18-item ethogram over 104 hours on eight observation days; every calf in the pen was coded for 90 of those hours. The release also includes 1.1 million decoded frames from seven annotated sessions, camera-clock timestamps extracted by optical character recognition, per-calf segmentation tracks and frame-level play labels. These materials support calf behaviour and welfare research and the development and evaluation of automated behaviour recognition.

### Source excerpt

arXiv:2611.21706v1 Announce Type: cross Abstract: Play is used as an indicator of positive welfare in dairy calves, but openly available video of group-housed calves with individual behaviour annotations is scarce. We describe video from four camera deployments of about one week each in group pens of dairy calves on two farms in the Netherlands: a research and teaching dairy farm (2024 and 2025) and a commercial dairy farm (2024). The recordings comprise 1,173 video files spanning 651 h. For eight observation days, trained observers coded calf behaviour continuously in Noldus The Observer XT with an 18-item ethogram of play and other behaviours, covering 104 h of video, 90 h of which have every calf in the pen coded. The release also contains 1.1 million decoded frames from seven annotated sessions, a per-frame timestamp read by optical character recognition from the camera clock burned into each frame, per-calf segmentation tracks and per-calf frame-level play labels. The data support studies of calf behaviour and welfare and the development and evaluation of automated behaviour recognition.

## On the Feasibility of Image-Based Representations for Air Traffic Complexity Estimation

DevFeed: [On the Feasibility of Image-Based Representations for Air Traffic Complexity Estimation](<https://devfeed.tech/articles/on-the-feasibility-of-image-based-representations-for-air-traffic-complexity-estimation-141593.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2608.11810>)

Author: Hyewook Kim, Byul Kang, Seokbin Yoon, Keumjin Lee

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [model architecture](<https://devfeed.tech/topics/model-architecture.md>), [Image](<https://devfeed.tech/topics/image.md>)

Tags: [air-traffic](<https://devfeed.tech/tags/air-traffic.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [computer-vision](<https://devfeed.tech/tags/computer-vision.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [vision-transformer](<https://devfeed.tech/tags/vision-transformer.md>), [vit](<https://devfeed.tech/tags/vit.md>)

### AI overview

A Vision Transformer can estimate four intrinsic air traffic complexity components from a sparse radar-style image representation. Each traffic situation is encoded as a position image with five additional channels for aircraft state variables, including heading, speed, and altitude; the target components are derived from pairwise geometric relations among aircraft. On test data, the model achieved R² > 0.96 for all four components. In a study removing one aircraft at a time, estimate changes tracked the removed aircraft's contribution to sector complexity rather than being uniform. The results support this representation as a viable input format, though the supplied excerpt does not establish broader operational performance.

### Source excerpt

arXiv:2608.11810v3 Announce Type: replace-cross Abstract: Air traffic controllers perceive traffic complexity through the radar display, suggesting that a computer vision model operating on the same imagery may provide a natural architecture for modeling controller-perceived complexity; however, whether radar imagery is a viable input format for deep learning vision models remains unclear. Unlike natural images, radar images are extremely sparse and self-similar, consisting primarily of a black background and a few visually identical aircraft blobs, while small changes in aircraft positions can substantially alter sector-level complexity. To test whether a vision model can capture these operationally important differences, we encode each traffic situation as a position image supplemented by five channels representing aircraft state variables, including heading, speed, and altitude, and train a Vision Transformer (ViT) to regress four intrinsic complexity components derived from pairwise geometric relations among aircraft. The model achieves $R^2 > 0.96$ for all four components on test data, and a one-aircraft removal perturbation study shows that the change in the model's estimate is proportional to how much the removed aircraft contributed to sector complexity, rather than the same for every removal. These results demonstrate that, despite its atypical visual characteristics, the proposed image-based representation is a viable input format for air traffic complexity modeling.

## Nanorobot queue: Cooperative treatment of cancer based on team member communication and image processing

DevFeed: [Nanorobot queue: Cooperative treatment of cancer based on team member communication and image processing](<https://devfeed.tech/articles/nanorobot-queue-cooperative-treatment-of-cancer-based-on-team-member-communication-and-image-processing-141512.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2111.11236>)

Author: Xinyu Zhou

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Image processing](<https://devfeed.tech/topics/image-processing.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>), [Image](<https://devfeed.tech/topics/image.md>), [Distributed Systems & Parallel Computing](<https://devfeed.tech/topics/distributed-systems-parallel-computing.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>)

Tags: [cancer](<https://devfeed.tech/tags/cancer.md>), [classification](<https://devfeed.tech/tags/classification.md>), [communication](<https://devfeed.tech/tags/communication.md>), [computer-vision](<https://devfeed.tech/tags/computer-vision.md>), [control](<https://devfeed.tech/tags/control.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [detection](<https://devfeed.tech/tags/detection.md>), [image](<https://devfeed.tech/tags/image.md>), [image-processing](<https://devfeed.tech/tags/image-processing.md>), [model](<https://devfeed.tech/tags/model.md>), [nano](<https://devfeed.tech/tags/nano.md>), [queue](<https://devfeed.tech/tags/queue.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [robot](<https://devfeed.tech/tags/robot.md>), [target](<https://devfeed.tech/tags/target.md>), [team](<https://devfeed.tech/tags/team.md>)

### AI overview

The paper proposes an idealized cancer-treatment model in which a queue of nanorobots cooperates through communication among team members and computer-vision image classification or target detection. It motivates the proposal by describing existing nanorobot applications and control approaches as limited by immature technology, low efficiency or accuracy, or difficulty with mass production, while identifying chemotherapy and radiotherapy as the most effective current treatments. The supplied abstract does not explain how the model would be implemented or demonstrate that it can cure cancer.

### Source excerpt

arXiv:2111.11236v4 Announce Type: replace-cross Abstract: Although nanorobots have been used as clinical prescriptions for work such as gastroscopy, and even photoacoustic tomography technology has been proposed to control nanorobots to deliver drugs at designated delivery points in real time, and there are cases of eliminating "superbacteria" in blood through nanorobots, most technologies are immature, either with low efficiency or low accuracy, Either it can not be mass produced, so the most effective way to treat cancer diseases at this stage is through chemotherapy and radiotherapy. Patients are suffering and can not be cured. Therefore, this paper proposes an ideal model of a treatment method that can completely cure cancer, a cooperative treatment method based on nano robot queue through team member communication and computer vision image classification (target detection).

[Next page](<https://devfeed.tech/topics/image-recognition.md?cursor=WyIyMDI2LTExLTE4VDA1OjAwOjAwKzAwOjAwIiwgIjAyMzFiYmY3LTU2MjctNDg2OC1hZWUwLWY5ZmUwYjYyZjliYSJd>)