The DOI matches its Crossref record; 154 of the 154 references it cites were found in a registry or an index.
Flamingo: A Visual Language Model for Few-Shot Learning
Abstract
Building models that can be rapidly adapted to novel tasks using only a handful of annotated examples is an open challenge for multimodal machine learning research. We introduce Flamingo, a family of Visual Language Models (VLM) with this ability. We propose key architectural innovations to: (i) bridge powerful pretrained vision-only and language-only models, (ii) handle sequences of arbitrarily interleaved visual and textual data, and (iii) seamlessly ingest images or videos as inputs. Thanks to their flexibility, Flamingo models can be trained on large-scale multimodal web corpora containing arbitrarily interleaved text and images, which is key to endow them with in-context few-shot learning capabilities. We perform a thorough evaluation of our models, exploring and measuring their ability to rapidly adapt to a variety of image and video tasks. These include open-ended tasks such as visual question-answering, where the model is prompted with a question which it has to answer, captioning tasks, which evaluate the ability to describe a scene or an event, and close-ended tasks such as multiple-choice visual question-answering. For tasks lying anywhere on this spectrum, a single Flamingo model can achieve a new state of the art with few-shot learning, simply by prompting the model with task-specific examples. On numerous benchmarks, Flamingo outperforms models fine-tuned on thousands of times more task-specific data.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“vision-language models (VLMs) such as VilBERT 19 and Flamingo 2 combine a language model with an image encoder and are trained on a multi-modal corpus of text-image pairs.This enables them to predict how a given sequence of words will continue in the context of a given image.”
From Talking about Large Language Models (2024) · cites this paper as [2]
In the citing paper
“the latest generation of language models are multimodal, allowing interleaved image and text input (OpenAI, 2023a; Alayrac et al., 2022; Team et al., 2023; Li et al., 2023b)”
From Cognitive Architectures for Language Agents (2023) · cites this paper as Alayrac et al., 2022
In the citing paper
“The Switch Transformer [201] was one of the first MoE-based Transformers, activating experts selectively in each layer, reducing computational overhead while outperforming dense Transformers in NLP tasks.”
From Multimodal fusion and vision–language models: A survey for robot vision (2026) · cites this paper as [201]
In the citing paper
“The dominant architectural paradigm connects specialized, external multimodal encoders-such as CLIP for vision or CLAP for audio-to the LLM backbone via alignment modules like Q-Former or simple MLPs [19, 86, 615, 1139], a modular design that allows for independent encoder updates without retraining the entire model [624].”
From A Survey of Context Engineering for Large Language Models (2025) · cites this paper as [19]
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404 · cited 3×
“Language modelling has recently made substantial progress following the introduction of Transformers [115].”
BackgroundMethodInfluentialHochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 2×
“To ensure that at initialization, the conditioned model yields the same results as the original language model, we use a tanh-gating mechanism [41].”
MethodBERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325 · cited 4×
“In particular, BERT [23] inspired a large body of vision-language work [16, 28, 29, 38, 59, 61, 66, 101, 106, 107, 109, 118, 121, 142, 143, 151].”
BackgroundMethodInfluentialAn Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 47,472
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380 · cited 2×
“Im ag e ImageNet-1k [94] 3 Object classification Val Top-1 acc.”
BackgroundSwin Transformer: Hierarchical Vision Transformer using Shifted Windows
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., & Guo, B. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00986 · Cited by 35,534
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372 · cited 12×
“We evaluate the ability of our models to rapidly adapt to new tasks using in-context learning , analogously to GPT-3 [11], by interleaving support example pairs in the form of ( 𝑖𝑚𝑎𝑔𝑒, 𝑡𝑒𝑥𝑡 ) or ( 𝑣𝑖𝑑𝑒𝑜, 𝑡𝑒𝑥𝑡 ) , followed by the query visual input, to build a prompt (details in Method FT Shot OKVQA ( I )”
BackgroundMethodResultInfluentialEnd-to-End Object Detection with Transformers
Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., & Zagoruyko, S. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58452-8_13 · Cited by 18,285 · cited 2×
“[13] Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, and Sergey Zagoruyko.”
BackgroundMethodLearning Transferable Visual Models From Natural Language Supervision
Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., et al. (2021) · arXiv · doi:10.48550/arxiv.2103.00020 · Cited by 15,293 · cited 12×
“[85] we use “prompt-ensembling” in which we embed multiple texts using templates such as “A photo of a {class_name}” and average the resulting embedding.”
BackgroundMethodInfluentialExploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929 · cited 3×
“To measure the importance of text pretraining, we compare the performance of using a frozen decoder-only Transformer either pretrained on MassiveText (our main model) or pretrained on the C4 dataset [87] (row (iv) of Table 10).”
BackgroundMethod
Cited by
Sorted by citations
Items 101–200 of 860
Embodiment in multimodal large language models
Kadambi, A., Aziz‐Zadeh, L., Damásio, A. R., Iacoboni, M., & Narayanan, S. (2026) · Neuron · doi:10.1016/j.neuron.2026.03.004 · Cited by 2
“This limits their ability to dynamically integrate information across modalities to adapt to new contexts, unlike humans [1, 84].”
InfluentialGenerative Artificial Intelligence Models: A Survey
Almuammar, M., Alsultan, K., & Altukhaim, Y. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11546-1 · Cited by 8
“Finally, progress in multimodal integration and sustainable model design will shape future Transformer research (Alayrac et al. 2022 ).”
Generative adversarial fault-tolerance for robust multimodal sensor fusion in industrial systems
Altinses, D., & Schwung, A. (2026) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2026.114225 · Cited by 3
GroundCap: A visually grounded image captioning dataset with object and action identification
Oliveira, D. A. P., Teodoro, L., & de Matos, D. M. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.113703 · Cited by 7
Three-dimensional primitive prediction with building embedded Swin Transformer for urban turbulence
Park, J. S., Kim, H., Tsiotras, P., & Lee, S. (2026) · Physics of Fluids · doi:10.1063/5.0324405 · Cited by 2
Medical multimodal large language models: A survey
Xiao, H., Hui, N., Xu, Y., Li, Z., & Peng, J. (2026) · Information Fusion · doi:10.1016/j.inffus.2026.104386 · Cited by 2
Poudineh, M., Esmaeilzehi, A., & Ahmad, M. O. (2026) · IEEE Transactions on Image Processing · doi:10.1109/tip.2026.3685457 · Cited by 2
Gaze-Language Alignment for Zero-Shot Prediction of Visual Search Targets from Human Gaze Scanpaths
Mondal, S., Sendhilnathan, N., Zhang, T., Liu, Y., Proulx, M. J., Iuzzolino, M. L., Qin, C., & Jonker, T. R. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00263 · Cited by 3
MistSense: Versatile Online Detection of Procedural and Execution Mistakes
Patsch, C., Wu, Y., Zakour, M., Salihu, D., & Steinbach, E. G. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01348 · Cited by 5
“Additional work further expanded the reasoning capabilities of LLMs to additional modalities and in particu-14529 lar vision-based language models (VLMs) [1, 21, 40].”
Token-Efficient VLM: High-Resolution Image Understanding Via Dynamic Region Proposal
Jiang, Y., Gu, J., Xue, T., Cheung, K. C., Molchanov, P. A., Yin, H., & Liu, S. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02238 · Cited by 2
“Vision-Language Models (VLMs) [1, 36, 37, 80] have recently received significant attention due to their ability to integrate visual understanding into Large Language Models (LLMs) [50, 59, 77], enhancing performance in tasks like image captioning and visual question answering.”
Describe, Adapt and Combine: Empowering CLIP Encoders for Open-Set 3D Object Retrieval
Wang, Z., Zhou, Y., Liu, Z., Yu, R., Bai, S., Wang, Y., He, X., & Bai, X. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01954 · Cited by 3
“This indicates that DAC not only enhances intra-class similarity but also effectively diminishes inter-class similarity, thereby markedly improving the discriminative power of the retrieval features.”
VisualCloze: A Universal Image Generation Framework via Visual in-Context Learning
Li, Z. H., Du, R., Yan, J., Zhuo, L., Li, Z. H., Gao, P., Ma, Z., & Cheng, M.-M. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01763 · Cited by 31
“In recent years, leveraging inpainting model [3, 4, 72], masked image modeling [39, 57, 58], or vision-language model [1, 76], visual in-context learning is proposed to handle more tasks.”
Physics Context Builders: A Modular Framework for Physical Reasoning in Vision-Language Models
Balazadeh, V., Ataei, M., Cheong, H., Khasahmadi, A. H., & Krishnan, R. G. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00687 · Cited by 20
“However, phys-* This project was completed during an internship at Autodesk Research. ical reasoning remains a significant challenge for artifi-cial intelligence (AI) systems [10, 23, 35, 45], despite advances in computer vision and Vision-Language Models (VLMs) [2, 25, 26, 32, 36, 39, 46].”
4D-Bench: Benchmarking Multi-Modal Large Language Models for 4D Object Understanding
Zhu, W., Li, B., Zheng, C., Mai, J., Chen, J., Jiang, L., Hamdi, A., Martinez, S. R., Lin, C., Elhoseiny, M., et al. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01963 · Cited by 5
“To understand information across multiple modalities, MLLMs [5, 17, 22, 55, 66, 105, 129] extend the capabilities of LLMs to modalities such as 2D images, videos, and audio by introducing alignment modules and visual instruction tuning.”
Tikzero: Zero-Shot Text-Guided Graphics Program Synthesis
Belouadi, J., Ilg, E., Keuper, M., Tanaka, H., Utiyama, M., Dabre, R., Eger, S., & Ponzetto, S. P. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01653 · Cited by 15
“Although prior networks resemble our adapters and have been used with inverse graphics models [51], they target global image embed-dings containing only abstract information, which degrades performance when used with inverse graphics architectures that work best with patch-level details [52].”
From Few-Shot to Zero-Shot: Towards Generalist Graph Anomaly Detection
Liu, Y., Li, S., Zheng, Y., Chen, Q., Chengqi Zhang, B., Yu, P. S., & Pan, S. (2026) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2026.3691902 · Cited by 6
“It has shown effectiveness in natural language processing and vision, and is widely used in few-shot/zero-shot settings [41], [42].”
Size Matters: Reconstructing Real-Scale 3D Models from Monocular Images for Food Portion Estimation
Vinod, G., Coburn, B., Raghavan, S., et al. (2026) · IEEE Conference on Artificial Intelligence (CAI) · doi:10.1109/cai68641.2026.11536547 · Cited by 2
Ko, B. C. (2026) · Sensors · doi:10.3390/s26113541 · Cited by 3
“A representative example is Flamingo [92], which introduced the Perceiver Resampler and gated cross-attention to connect frozen vision and language backbones, enabling flexible multi-modal integration even when text and image, or video, inputs are arbitrarily interleaved.”
Shi, P., Chow, C., & Wong, W. R. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.116545 · Cited by 2
Bridging language and action: A survey of language-conditioned robot manipulation
Yao, X., Zhou, H., Mees, O., Meng, Y., Xiao, T. G., Bisk, Y., Oh, J., Johns, E., Shridhar, M., Shah, D., et al. (2026) · The International Journal of Robotics Research · doi:10.1177/02783649261468360 · Cited by 56
“In contrast, RoboFlamingo 98 is fine-tuned exclusively on language-conditioned manipulation datasets using IL, demonstrating that even a small amount of robotic demonstration can effectively adapt the model to downstream manipulation tasks.”
InfluentialQi, X. S., Wang, Y., Yang, X., Ren, L., Liu, W., Benedict, S., Xiao, Y., Xing, L., & Naqa, I. M. E. (2026) · Medical Physics · doi:10.1002/mp.70616 · Cited by 2
Leveraging Multi-Text Joint Prompts in SAM for Robust Medical Image Segmentation
Zhang, X., Zhao, H., Zhang, L., & Xiong, Y. (2025) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2025.3607023 · Cited by 3
“Besides CLIP-series, there are also flamingo [30], BLIP [31], [32].”
GenVC: Self-Supervised Zero-Shot Voice Conversion
Cai, Z., Xinyuan, H. L., Garg, A., Garcia, L. P., Duh, K., Khudanpur, S. P., Wiesner, M., & Andrews, N. (2025) · IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) · doi:10.1109/asru65441.2025.11434735 · Cited by 3
“This phase also incorporates a Perceiver encoder to extract fixed-length style representations from the prompt utterance [43].”
Toward Scalable ASL Education: Egocentric Stereo Sensing with LLM Feedback for Error-Aware Learning
Cai, Y., Li, Z., Lu, T., Zhu, Y., Wu, Y.-S., Zhang, Q., Xu, X., Jin, Z. K., Gowda, M. K., & Jin, Y. (2026) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3790774 · Cited by 2
A Design Space for Live Music Agents
Kim, Y., Brade, S., Wang, A., Zhou, D., Kim, H., Wang, B., Lee, S.-J., García, H. F., Huang, A., & Donahue, C. (2026) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3791291 · Cited by 4
“Shortly after, pioneering work such as George Lewis’s Voyager [177] and Oscar [23] offered a vision of live music agents that could not just follow along with but contribute directly to a musical performance alongside human musicians.”
MoMA: a mixture-of-multimodal-agents architecture for enhancing clinical prediction modelling
Gao, J., Rahman, M., Caskey, J., Oguss, M., O’Rourke, A., Brown, R., Stey, A., Mayampurath, A., Churpek, M. M., Chen, G., et al. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-02219-4 · Cited by 2
Sun, Y., Min, X., Zhang, Z., Gao, Y., Cao, Y., & Zhai, G. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3619558 · Cited by 2
“MLLMs exem-plified by models such as LLaVA and InstructBLIP have demonstrated their considerable potential in various domains [4], [5], including image captioning, visual question answering, image segmentation, medical image diagnosis, image quality assessment (IQA) and aesthetic evaluation [6],…”
Foundation Model for Medical Imaging: A Comprehensive Review
Jiao, L., Yang, J., Li, R., Liu, F., Liu, X., Chen, P., Guo, Y., Li, L., Shang, R., & Ma, W. (2025) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2025.3618796 · Cited by 4
“Subsequently, the model was continually pretrained on the MTB and PMC-OA [132] datasets based on the OpenFlamingo-9B model [133], utilizing 8 A100 GPUs.”
Constituency-Tree-Induced Vision–Language Alignment for Multimodal Large Language Models
Zhai, Y., Xu, N., Tian, H., Zheng, B., Yan, C. C., Cao, J., Kang, R., & Liu, A.-A. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3639574 · Cited by 1
“Flamingo [2] leverages cross-attention module equipped in LLMs layers to adapt the visual ability of language decoders.”
VLM-Driver: Human-Like Autonomous Driving Decision-Making via Vision Language Model
Zhao, R., Yuan, Q., Li, J., et al. (2025) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2025.3629119 · Cited by 1
“Typically, VLMs achieve efficient integration between the LLM backbone and the visual module in two ways: simple linearmapping,asinLLaVA[19]andPaLM-E[20],andscalable Q-formerarchitectures,asinFlamingo[17]andBLIP2[18].”
Question Understanding and Temporality Guiding for Video Question Answering
Lu, J., You, S., & Bao, B.-K. (2026) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2026.3651064 · Cited by 1
MAViS: A Multi-Agent Approach for Training-Free Referring Video Object Segmentation
Peng, T., & Chen, C.-S. (2026) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2025.3650288 · Cited by 1
“Early MLLMs, including Flamingo [28], BLIP-2 [29], align visual and textual modalities via cross-attention or adapter-based methods, enabling tasks such as image captioning and visual question answering.”
Adaptive in Adapter: Boosting Open-Vocabulary Semantic Segmentation With Adaptive Dropout Adapter
Wang, C., Xu, W., Xu, R., Zhang, Z., Xu, S., Zhang, J., Teng, X., Meng, W., & Zhang, X. (2026) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2026.3654453 · Cited by 1
Sparse-View 3-D Language Gaussian Splatting for Zero-Shot Robotic Grasping
Zhang, Y., Wang, Y., Chen, W., Xie, H., Xu, Z., Che, Z., Ren, P., Tang, J., & Sun, J. (2026) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2025.3645486 · Cited by 1
“Recent advances in vision-language models (VLMs) [14], [15], [16], [17], such as CLIP [18], have demonstrated remarkable capabilities in open-vocabulary recognition by learning rich semantic priors from large-scale image-text data.”
Vocabulary-Free Image Classification and Semantic Segmentation
Conti, A., Fini, E., Mancini, M., Rota, P., Wang, Y., & Ricci, E. (2026) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2026.3657989 · Cited by 4
“A separate stream of research has focused on enhancing vision-language pre-training for more intricate vision-language tasks, such as image captioning and visual question answering (VQA) [2], [3], [12], [26], [27].”
Long Video Understanding With Learnable Retrieval in Video-Language Models
Xu, J., Lan, C., Xie, W., Chen, X., & Lu, Y. (2026) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2026.3664996 · Cited by 29
“Flamingo [14] and BLIP-2 [15] are seminal works that leverage the pre-trained LLM and CLIP image encoder for zero-shot image-text abilities.”
Jing, Y., Zhang, X., Zhou, M., Ji, Y., Ding, Y., Yin, J., Jia, R., Xiong, Y., Zhao, K., Yang, J., et al. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3668085 · Cited by 2
“…{ I X , t } , the system optimizes the projector by minimizing the text generation loss L txt-gen arg min L Implementation architectures for Θ X → T include linear projector, multilayer perceptron (MLP), attention mechanism, perceiver resampler [92], [93], C-Abstractor [94], and D-Abstractor [94].”
3D-SceneQ: Empowering 3D LLM With Query-Guided Adaptive Pruning and Multi-Modal Feature Enhancement
Shan, Y., Zhang, L., Yu, J., Zhuge, Y., & Lu, H. (2026) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2026.3668502 · Cited by 2
Human centric general physical intelligence for agile manufacturing automation
Kanta, S., Tavassoli, M., Chirkuri, V. T., Kumar, V. A., Punati, S. B., Damacharla, P., & Katyara, S. (2025) · Advanced Engineering Informatics · doi:10.1016/j.aei.2026.104515 · Cited by 4
“…on real robots with up to 50% success rate on un-seen tasks after fine-tuning RoboFlamingo [134] Generalized Learning from Demonstrations Flamingo [135], ViT [126], CLIP [121], Flamingo-robot adapter Frozen backbone ensures stability while training adapter; shows strong generalization from…”
Chen, W., Huang, S., Zhang, H., Tu, Y., Lu, Y., Li, Q., & ZHANG, J. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11524-7 · Cited by 3
A Confidence-Aware Human–Machine Collaboration Framework for Harmful Meme Detection
Lin, X.-Y., Tseng, Y.-Y., Huang, S.-C., & Wang, M.-H. (2026) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2025.3649665 · Cited by 1
“With the recent rise of powerful generative AI models, LLMs have demonstrated strong capabilities in reasoning and multi-modal understanding [19].”
SPEX: A Vision–Language Model for Land Cover Extraction on Spectral Remote Sensing Images
Si, D., Wang, D., Gao, E., Qin, X., Zhao, L., Zhang, J., Xu, M., Zhan, J., Wang, J., Liu, L., et al. (2026) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2026.3670308 · Cited by 7
“For example, models such as Flamingo [24], LLaVA [16], Instruct-BLIP [25], and MiniGPT-4 [26] have achieved remarkable results in visual question answering and image–text multimodal dialog tasks, showcasing the potential of VLMs in cross-modal understanding.”
Deniz, O. F., Mao, R., Li, R., Tian, Y., & Khan, L. R. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11402394 · Cited by 1
“Vision tokens often outnumber text tokens by an order of magnitude, dominating memory usage, increasing latency, and straining the context length of the model [5], [6].”
MCDAN: Multi-Modal Cross-Domain Adaptation Network for Fake News Detection
Yang, H., Zhang, Q., & Zhang, Y. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11401755 · Cited by 1
“The most recent paradigm shift is towards end-to-end Vision-Language Models (VLMs) like Flamingo [19] and LLaVA [20].”
Embodied Referring Expression Comprehension in Human-Robot Interaction
Islam, M., Gladstone, A., Sarker, S. C., Nanduru, G., Fahim, M., Du, K., Chadha, A. R., & Iqbal, T. (2026) · ACM/IEEE International Conference on Human-Robot Interaction (HRI) · doi:10.1145/3757279.3785601 · Cited by 2
Physical AI goes to the operating room: are we ready for the Surgical Data Factory?
Oh, N., Jung, K.-H., & Choi, G. (2026) · Annals of Surgical Treatment and Research · doi:10.4174/astr.2026.110.3.135 · Cited by 2
Smart Vision-Language Reasoners
Roberts, D. A. O., & Roberts, L. (2025) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw69685.2025.00168 · Cited by 1
“…of GELU units with layer normalization boosts performance, as they do in many recent attention-based multimodal neural networks (Liu et al., 2023; Alayrac et al., 2022; Li et al., 2023b), by allowing for a smoother loss landscape than rectified linear units with batch normalization layers.”
Med-ViX-Ray: Enhancing explainable chest X-ray analysis with clinical knowledge graphs
Cieri, M., & Palomba, F. (2026) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2026.109313 · Cited by 1
Video-Based Traffic Anomaly Detection with Vision-Language Models: A Survey
Zheng, Y., Zhou, W., Nastić, S., & Wang, C. (2025) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc60802.2025.11423492 · Cited by 1
“For example, Flamingo [28] enables models to reason about and generate image content given textual contexts by modeling cross-modal sequences.”
Adversarial Disillusion Against PGD Attacks via Multimodal Generative Chain-of-Thought Reasoning
Chu, H., Ou, W., Li, X., & Ma, W. (2026) · International Conference on Artificial Intelligence and Social Network Systems (AISNS) · doi:10.1109/aisns67921.2026.11440364 · Cited by 1
Li, Z., Meng, X., He, X., Zhang, Y., & Yin, W. (2026) · Sensors · doi:10.3390/s26072022 · Cited by 6
Seg-LLaVA: Empowering pixel-level understanding with large vision language model
Yang, F., Zhu, Y., Zhan, Y., Zhao, H., Li, X., Wang, Y., Tang, M., Ning, X., & Wang, J. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.113560 · Cited by 2
Competitive Audio-Language Models with Data-Efficient Single-Stage Training on Public Data
Kumar, G. K., Saraf, R., Lepauloux, L., Muneer, A., Mokeddem, B., & Hacid, H. (2025) · IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) · doi:10.1109/asru65441.2025.11434596 · Cited by 2
“Models like Flamingo [10], LLaVA [11], ShareGPT4V [12], Qwen-Vision [13], and IDEFICS [14] demonstrate effective strategies for combining image and text inputs.”
Steerable Visual Representations
Ruthardt, J., Gaur, M., Ramanan, D., Tapaswi, M., & Asano, Y. M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37016-7_19 · Cited by 7
Bi, Z., Wu, X., Xu, X., et al. (2026) · Journal of Geophysical Research Machine Learning and Computation · doi:10.1029/2025jh000848 · Cited by 1
Detecting stress from videos via intra-subject and inter-subject learning
Ding, Y., Dai, Y. Q., Cao, L., Tian, J., Lin, Z., & Feng, L. (2026) · Health Information Science and Systems · doi:10.1007/s13755-026-00450-1 · Cited by 1
PeFoMed: Parameter efficient fine-tuning of multimodal large language models for medical CXR
Liu, G., Tang, X., He, J., Li, P., Chen, Z., & Zhong, S. (2026) · Scientific Reports · doi:10.1038/s41598-026-47871-2 · Cited by 81
Benchmarking Multimodal Large Language Models for Face Recognition
Shahreza, H. O., & Marcel, S. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11463782 · Cited by 1
CPJ: Explainable Agricultural Pest Diagnosis Via Caption–Prompt–Judge with LLM-Judged Refinement
Zhang, W., Fang, T., Lu, L., et al. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11464859 · Cited by 1
How large-scale foundation models benefit precision livestock farming: A survey
Tian, M., Guo, H., Long, C., Zhu, J., Zhao, W., Li, H., Ruchay, A. N., Pezzuolo, A., & Li, B. (2026) · Artificial Intelligence in Agriculture · doi:10.1016/j.aiia.2026.04.013 · Cited by 3
FDPT: Federated Discrete Prompt Tuning for Black-Box Visual-Language Models
Wu, J., Chen, S., Tang, J., Yang, Y., Chen, Y., Wang, L., Lin, S., Wang, Z., Chen, W., & Tian, Z. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00237 · Cited by 2
“Flamingo [1] leverages few-shot learning capabilities, allowing it to quickly adapt to new tasks and perform well across both VQA and classification.”
Training-Free Personalization via Retrieval and Reasoning on Fingerprints
Das, D., Talon, D., Wang, Y., Mancini, M., & Ricci, E. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00903 · Cited by 8
“” are questions easy to understand for humans, they are challenging for Vision and Language Models (VLMs) [1, 2, 5, 10, 23] due to their inherent lack of understanding for user-specific concepts.”
Ma, J., Peng, Q., Guo, X., Chen, C., Lu, H., & Yang, Z. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01554 · Cited by 7
LLaVA-SP: Enhancing Visual Representation with Visual Spatial Tokens for MLLMs
Lou, H., Fan, C., Liu, Z., Wu, Y., & Wang, X. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02044 · Cited by 1
TWIST & SCOUT: Grounding Multimodal LLM-Experts by Forget-Free Tuning
Bhowmik, A., Derakhshani, M. M., Koelma, D. C., Asano, Y. M., Oswald, M. R., & Snoek, C. G. M. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00134 · Cited by 2
“Multimodal LLMs. Large Language Models (LLMs), known for their instruction-following and generalization abilities, have been effectively integrated with vision encoders, achieving strong multimodal performance [1, 2, 4, 10, 11, 20, 22, 24, 33, 37, 41, 43, 47].”
Integrating Visual Interpretation and Linguistic Reasoning for Geometric Problem Solving
Guo, Z., Liu, M., Wang, Q., Ji, Z., Bai, J., Lei Zhang, A., & Zuo, W. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00380 · Cited by 1
“Current development of vision-language models typically employ a connector module, e.g. Linear layers [18], Q-Former [15], Perceiver-Resampler [1], etc., to integrate visual features with text features of LLMs.”
LIRA: Inferring Segmentation in Large Multi-Modal Models with Local Interleaved Region Assistance
Li, Z. L., Yang, B., Liu, Q., Zhang, X., Zx, M., Yin, L., Deng, L., Sun, Y., Liu, Y., & Bai, X. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02230 · Cited by 4
“In recent years, the development of LMMs has made remarkable strides, improving the visual comprehension capabilities of LLMs. Early works [1, 21, 29, 65, 73] explore various methods to integrate visual features into Large Language Models (LLMs).”
Scaling Inference-Time Search with Vision Value Model for Improved Visual Comprehension
Wang, X., Yang, Z., Li, L., Lu, H., Xu, Y., Lin, C.-C., Lin, K., Huang, F., & Wang, L. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00117 · Cited by 1
“Recently, modern vision language models [2, 3, 12, 33, 36, 45, 57, 67] further combine multimodal modeling with large language models to enable stronger capabilities, such as instruction following, in-context learning, and zero-shot generalization.”
VSP: Diagnosing the Dual Challenges of Perception and Reasoning in Spatial Planning Tasks for MLLMS
Wu, Q., Zhao, H., Saxon, M., Bui, T., Wang, W. Y., Zhang, Y., & Chang, S. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00219 · Cited by 1
“Recently, multimodal large language models (MLLMs) have advanced on LLMs through training on native image inputs, to achieve impressive performance generating text describing and relating to input images [1, 8, 10, Figure 1.”
RoboTron-Mani: All-in-One Multimodal Large Model for Robotic Manipulation
Yan, F., Liu, F., Huang, Y., Guan, Z., Zheng, L., Zhong, Y., Feng, C., & Ma, L. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01272 · Cited by 29
“It typically employs auto-regressive models such as LLaMA [62] or GPT [1], as well as cross-attention models like Flamingo [2] or LLaMA3.”
Acknowledging Focus Ambiguity in Visual Questions
Chen, C., Tseng, Y., Li, Z., Venkatesh, A., & Gurari, D. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00122 · Cited by 7
“This recent trend reflects that state-of-the-art performance regularly arises from foundation models in such settings; e.g., Frozen in 2021 [48], Flamingo in 2022 [1], ViTiS in 2023 [13], and LLaVA-v1.”
Cooperative Pseudo Labeling for Unsupervised Federated Classification
Guo, K., Sheng, L., Yu, Y., et al. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00318 · Cited by 1
“Vision-language models (VLMs) [1, 22, 39], particularly contrastive language-image pre-training (CLIP) [39], have recently garnered significant attention for their remarkable representation and generalization capabilities [7, 33, 50].”
Liu, H., Ruan, S., Huang, Y., Zhao, S., & Wei, X. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00976 · Cited by 2
Zero-Shot Compositional Video Learning with Coding Rate Reduction
Jung, H., Bak, J.-H., Jeong, Y., et al. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01907 · Cited by 1
“…and temporal categories, and the model is required to predict novel action labels that are unseen during training by recombining learned primitive concepts. across various downstream tasks but also for their ability to generate reasonable answers to unseen (novel) problems [1, 3, 20, 34, 47].”
ContextFace: Generating Facial Expressions from Emotional Contexts
Kim, M., Kim, M., & Baek, S. J. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01059 · Cited by 1
Igd: Instructional Graphic Design With Multimodal Layer Generatio
Qu, Y., Xie, H., Zhang, Y., et al. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01693 · Cited by 2
“By mapping the visual features extracted from the foundational vision model into the semantic space of LLMs, visual language models [2, 4, 7, 23, 49, 55] effectively integrate visual and language representations and achieve multimodal understanding.”
MarineEval: Assessing the Marine Intelligence of Vision-Language Models
Wong, Y. K., To, T.-A., Zhang, J., Zheng, Z., & Yeung, S.-K. (2026) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61042.2026.00148 · Cited by 2
“Flamingo [4] pioneered web-scale vision-language pretraining by bridging image and text models.”
Abdullah, S., Hossain, R., & Chowdhury, M. (2026) · Discover Applied Sciences · doi:10.1007/s42452-026-08793-w · Cited by 1
“By using a projection layer that maps visual information into the embedding space of the LLM, models like LLaVA [7] and Flamingo [8] bridge the gap between visual perception and textual reasoning.”
Cross-modal local and global alignment for Chinese Character Recognition
Yang, J., Zhu, A., & Cai, H. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.113930 · Cited by 1
Retrieval-augmented multimodal deep learning for interpretable lung nodule malignancy prediction
Hou, R., Chai, S., Jain, R. K., Li, Y., Liu, J., Teng, S., Shi, X., Lin, L., & Chen, Y. (2026) · Displays · doi:10.1016/j.displa.2026.103525 · Cited by 3
CNS-Obsidian: A Neurosurgical Vision-Language Model Built From Scientific Publications
Alyakin, A., Stryker, J., Alber, D. A., Lee, J. V., Sangwon, K. L., Duderstadt, B., Save, A., Kurland, D. B., Frome, S., Singh, S., et al. (2026) · Neurosurgery · doi:10.1227/neu.0000000000004070 · Cited by 2
“First, we focused on fully autoregressive models, excluding cross-attention-based architectures 28,29 or grounding strategies that better align text and image features.”
Wang, X., Feng, T., Liu, H., & Zhu, W. (2026) · Springer eBooks · doi:10.1007/978-981-95-7750-7_2 · Cited by 1
Saket, O., Aicha, A. B., & Fathallah, H. (2026) · IEEE Middle East and North Africa Communications Conference (MENACOMM) · doi:10.1109/menacomm69507.2026.11532827 · Cited by 1
“Recent years have witnessed the introduction of multi-modal deep learning models, termed Vision-Language Models (VLMs) [3], [4].”
From Prompts to Context: An Ontology-Driven Framework for Human-Generative AI Collaboration
Lê, N. L., Abel, M., & Laforge, B. (2026) · Information Systems Frontiers · doi:10.1007/s10796-026-10754-y · Cited by 2
Towards reasoning-empowered task-oriented communication for agent networks
Xie, S., Li, H. L., Wang, Z., Song, S., Zhang, J., & Letaief, K. B. (2026) · npj Wireless Technology · doi:10.1038/s44459-026-00028-z · Cited by 1
“The integration of complex, high-load models such as LLMs and world models demands a well-de fi ned architectural strategy to ensure system feasibility while effectively addressing the inherent trade-offs between computational overhead and ef fi ciency gains.”
InfluentialExploring the architectures of large language models and impact across multiple domains: a review
Waqas, M., Rajab, R., Laghari, A. A., Kryvinska, N., Hejaili, A. A., Khan, N., Khuhro, M. A., & Zhang, F. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11609-3 · Cited by 2
Beyond language: generative artificial intelligence as a general computing model for medicine
Sitek, A., & Bates, D. W. (2026) · The Lancet Digital Health · doi:10.1016/j.landig.2026.101011 · Cited by 1
Liu, T., Qi, X., Wang, L., Li, J., Lin, X., Zhu, M., Cui, Y., Liu, C. X., & Chen, Q. (2026) · Sensors · doi:10.3390/s26123698 · Cited by 1
“With the recent breakthrough of large language models (LLMs) [19,56] and large vision language models (LVLMs) [17,57–59], a promising avenue has emerged: leveraging their zero-shot reasoning capabilities to tackle VLN-CE without task-specific fine-tuning.”
Jiang, K., & Xiao, X. (2026) · Interactive Learning Environments · doi:10.1080/10494820.2026.2679730 · Cited by 1
A Unified Inference Optimization Framework for Qwen3-VL-2B-Instruct
Zhou, Q. (2026) · International Conference on Electronic Information Engineering, Big Data and Computer Technology (EIBDCT) · doi:10.1109/eibdct69742.2026.11566699 · Cited by 1
“Representative systems have demonstrated impressive ability in image grounded dialogue, reasoning, and instruction following, which has accelerated interest in deploying such models in practical settings [1].”
Accelerating Multimodal Large Language Models with Prior-Corrected Token Reduction
Chen, Z., Cai, Y., Guo, J., Cai, T., Yin, J., & Chen, Z. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-36839-3_13 · Cited by 1
Robust Onion: Peeling Open Vocab Object Detectors Under Noise
Pathak, P., Karuppasamy, M., Baranwal, A., Vyas, S., & Rawat, Y. S. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37095-2_20 · Cited by 1
SAHs: Unraveling Safety Attention Heads in Large Vision-Language Models
Zheng, Z., He, L., Zhao, J. Z., Yang, L., & Li, F. (2026) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2026.3710895 · Cited by 1
Instructta: instruction-tuned targeted attack for large vision-language models
Wang, X., Ma, P., Ji, Z., Li, Z., Wang, S., & Gu, W. (2026) · Cybersecurity · doi:10.1186/s42400-026-00612-4 · Cited by 37
“Benefiting from the strong comprehension of large language models (LLMs), recent LVLMs [12, 25, 27, 48] on top of LLMs have demonstrated a superior capability in solving complex tasks than that of small vision-language models.”
Oh, G., Kim, S., Park, S., & Kim, B.-H. (2026) · Journal of Medical Internet Research · doi:10.2196/90693 · Cited by 7
“More recently, models that integrate data from multiple modalities, particularly vision-language models (VLMs) [10-16], have garnered significant attention among multimodal approaches.”
From text to multimodality: A comprehensive survey of LLMs and MLLMs across diverse domains
Anbukkarasi, S., Hemalatha, S., Arunkumar, B., Shanmugavalli, V., Kokila, S., & A, Y. K. (2026) · ICT Express · doi:10.1016/j.icte.2026.07.023 · Cited by 1
Agentic Ai in Remote Sensing: Foundations, Taxonomy, and Emerging Systems
Talemi, N. A., Boone, J., & Afghah, F. (2026) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw68408.2026.00088 · Cited by 5
“Extending these capabilities, multimodal large language models (MLLMs) [5, 58, 130] couple visual encoders with large language models (LLMs) [9, 17] to support complex reasoning.”
Rocha, A., de Oliveira, P. C., Ferreira, J. P., Alves, M., & Sousa, A. J. (2026) · Applied Sciences · doi:10.3390/app16157745 · Cited by 1
Elsayed, S. (2026) · International Conference on Artificial Intelligence, Robotics and Control (AIRC) · doi:10.1109/airc69745.2026.11631388 · Cited by 1
“Subsequent works like Flamingo [5] and BLIP-2 [6] demonstrated strong performance on visual question answering and image captioning through efficient cross-modal fusion architectures.”
Adapting Vision-Language Models Without Labels: A Comprehensive Survey
Dong, H., Sheng, L., Liang, J., He, R., Chatzi, E., & Fink, O. (2026) · International Journal of Computer Vision · doi:10.1007/s11263-026-02987-w · Cited by 15
“V ISION-language models (VLMs), such as CLIP [1], ALIGN [2], Flamingo [3], and LLaVA [4] have attracted considerable attention from both academia and industry due to their powerful cross-modal reasoning capabilities.”
Metadata
- Authors
- Alayrac, J.-B., Donahue, J., Luc, P., Miech, A., Barr, I., Hasson, Y., Lenc, K., Mensch, A., Millican, K., Reynolds, M., Ring, R., Rutherford, E., Cabi, S., Han, T., Gong, Z., Samangooei, S., Monteiro, M., Menick, J. L., Borgeaud, S., Brock, A., Nematzadeh, A., Sharifzadeh, S., Binkowski, M., Barreira, R., Vinyals, O., Zisserman, A., & Simonyan, K.
- Venue
- Advances in Neural Information Processing Systems 35
- Published
- 2022
- Pages
- 23716-23736
- ISSN
- 1049-5258 (ISSN-L)
- OpenAlex
- W7133196460
- Cited by
- 966
- References
- 154
- Citations by year
- 2026: 847 · 2025: 107 · 2024: 3 · 2023: 7 · 2022: 2
- FWCI
- 15.48
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- neural information processing systems (publishedVersion)
- Oxford University Research Archive (ORA) (University of Oxford) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- vision-language models, few-shot learning, image-text models
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
