IP Library Granted Patent US 12,619,654
Granted Patent B2
US 12,619,654 · App. 18/141,337 · Granted May 5, 2026

Language model for processing a multi-mode query input

Inventors: Jean-Baptiste Alayrac (London, GB); Jeffrey Donahue (London, GB); Karel Lenc (London, GB); Karen Simonyan (London, GB); Malcolm Kevin Campbell Reynolds (London, GB); Pauline Luc (London, GB); Arthur Mensch (Paris, FR); Iain Barr (Cambridge, GB); Antoine Miech (London, GB); Yana Elizabeth Hasson (Paris, FR); Katherine Elizabeth Millican (London, GB); Roman Ring (London, GB)
G06F16/432G06F16/438G06F40/284
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,619,654
App. No.
18/141,337
Granted
May 5, 2026
Kind
B2
Abstract

A query processing system is described which receives a query input comprising an input token string and also at least one data item having a second, different modality, and generates a corresponding output token string.

Claims (56)

1 . A computer-implemented method of training a query processing system, the query processing system being for generating an output token string based on a query input comprising an input token string and one or more data items, the input token string and output token string being strings of tokens selected from a token vocabulary, and the data items being of a modality other than tokens selected from the token vocabulary,

the method employing a token processing model comprising a stack of token processing layers, the stack of token processing layer being configured to receive input token strings and to generate corresponding output token strings, and a database of training examples, each training example comprising at least one data item and at least one token string;

the method comprising:

forming a data-item-token processing model by interleaving token processing layers from a token processing model with gated cross-attention layers, the data-item-token processing model being configured to generate an output token string upon receiving a prompt input which is a token string, the token processing model comprising a stack of the token processing layers, the stack of token processing layers being configured to receive input token strings and to generate corresponding output token strings, and a database of training examples, each training example comprising at least one data item and at least one token string;

forming the query processing system, the query processing system comprising:

(a) a modality network configured to receive the data items of the query input, to generate one or more compressed representations of each data item; and

(b) the data-item-token processing model, the data-item-token processing model being configured to receive a prompt input comprising the input token string of the query input, and each gated cross-attention layer being arranged to receive at least one of the compressed representations; and

using the training database, training:

the modality network, and

the plurality of gated cross-attention layers.

2 . The computer-implemented method of claim 1 in which the training trains the query processing system, upon an encoder of the modality network receiving the at least one data item of any of the training examples, and the data-item-token processing model receiving a prompt input comprising a first portion of the token string of the training example, to generate an output of the query processing system which is positively statistically correlated with a subsequent portion of the token string of the training example.

3 . The computer-implemented method of claim 1 in which the modality network comprises:

an encoder configured to encode a data item received by the encoder to generate an encoded data item, and

a compressed representation generation system arranged to receive the encoded data item and generate an output, the output of the modality network being based on the output of the compressed representation generation system.

4 . The computer-implemented method of claim 3 , in which the encoder has been trained to encode a data item received by the encoder to generate an encoded data item, and the training of the modality network and the plurality of gated cross-attention layers comprises training the compressed representation generation system without further training the encoder.

5 . The computer-implemented method of claim 3 , in which the compressed representation generation system comprises a stack of one or more resampler layers,

each resampler layer being adapted to perform an attention operation which employs a key vector, a value vector and a query vector, a subset of the key vector, value vector and query vector being based on the encoded data item, and the remainder of the key vector, value vector and query vector being based on either an output of the preceding one of the resampler layers or, in the case of the first resampler layer of the stack, a set of input latent values,

the output of the modality network being based on an output of the last resampler layer of the stack of resampler layers.

6 . The computer-implemented method of claim 5 in which the key vector and value vector of each resampler layer are based on the encoded data item and a latent input which is either the output of the preceding one of the resampler layers or, in the case of the first resampler layer of the stack, the set of input latent values, and the query vector is based on the latent input.

7 . The computer-implemented method of claim 5 in which each resampler layer further comprises a perceptron arranged to receive the output of the attention operation, and to generate an output, the output of the modality network being based on the output of the perceptron of the last resampler layer of the stack.

8 . The computer-implemented method of claim 1 , in which the prompt input further comprises one or more corresponding marker items for each data item in the query input, the one or more marker items being indicative of the presence of the data item in the query input.

9 . The computer-implemented method of claim 8 in which a position of each marker item in the prompt input is indicative of a position of the corresponding data item in the query input.

10 . The computer-implemented method of claim 1 in which each gated cross-attention layer generates its output as a component-wise sum of:

a first input which is the output of the preceding processing layer in the stack of processing layers or, in the case that the gated cross-attention layer is the first processing layer of the stack of processing layers, the prompt input, and

an interaction term based on the output of the compressed representation generation system received by the gated cross-attention layer, and at least part of the first input to the gated cross-attention layer.

11 . The computer-implemented method of claim 10 , in which the interaction term has a magnitude which depends positively upon the value of a gating parameter, the training comprising incrementally increasing the learning parameter.

12 . The computer-implemented method of claim 10 which includes, in the case of a query input comprising a plurality of portions, each portion comprising one of the data items, for each portion:

the modality network generating at least one respective compressed representation of the corresponding data item, and

at least one of the gated cross-attention layers generating the interaction term based only on the compressed representation of the corresponding data item and without employing data generated based on tokens of the input token string other than within the portion.

13 . The computer-implemented method of claim 1 in which each gated cross-attention layer comprises a cross-attention layer, which employs a key vector, a value vector and a query vector, a subset of the key vector, value vector and query vector being based on the at least one compressed representation received by the gated cross-attention layer, and the remainder of the key vector, value vector and query vector being based on the output of the preceding processing layer in the stack of processing layers or, in the case that the gated cross-attention layer is the first processing layer of the stack of processing layers, based on the prompt input.

14 . The computer-implemented method of claim 13 in which the key vector and value vector of each gated cross-attention layer are obtained based on the at least one compressed representation received by the gated cross-attention layer, and the query vector of each gated cross-attention layer is based on the output of the preceding processing layer in the stack of processing layers or, in the case that the gated cross-attention layer is the first processing layer of the stack of processing layers, based on the prompt input.

15 . The computer-implemented method of claim 13 in which the gated cross-attention layer further comprises a perceptron which receives the output of the cross-attention layer, the output of the gated cross-attention layer being based on an output of the perceptron.

16 . A computer-implemented method of generating an output token string based on a query input comprising an input token string and one or more data items, the input token string and output token string being strings of tokens selected from a token vocabulary, and the data items being of a modality other than tokens selected from the token vocabulary, the method comprising:

(a) generating one or more compressed representations of each data item by processing the data item using a modality network which comprises:

an encoder configured to encode the data item to generate an encoded data item, and

a compressed representation generation system arranged to receive the encoded data item and generate an output,

wherein the compressed representation generation system comprises a stack of one or more resampler layers,

each resampler layer being configured to perform an attention operation which employs a key vector, a value vector and a query vector, the key vector, value vector and query vector each being based on at least one of the encoded data item and a latent input which is either an output of the preceding one of the resampler layers or, in the case of the first resampler layer of the stack, a set of input latent values, at least one of the key vector, value vector and query vector being based on both the encoded data item and the latent input,

the output of the modality network being based on an output of the last resampler layer of the stack of resampler layers;

(b) generating a prompt input comprising the input token string of the query input; and

(c) inputting the prompt input and the compressed representation of each data item to a data-item-token processing model configured to generate the output token string based on the prompt input and the compressed representation of each data item.

17 . The computer-implemented method of claim 16 in which the output token string is the response to a query about the content of a subject data item which is one of the data items in the query input, the query being defined based on the input token string.

18 . The computer-implemented method of claim 17 in which the query input comprises, in addition to the subject data item, one or more task example portions which each include a respective data item and a respective section of the input token string, and for each task example portion the respective section of the input token string is the response to the query when the query is about the content of the respective data item.

19 . The computer-implemented method of claim 17 , wherein the query input is a question and the response to the query is an answer to the question.

20 . The computer-implemented method of claim 16 in which the token vocabulary comprises the symbols of a natural language writing system.

21 . A system comprising one or more processors and one or more storage devices storing instructions that when executed by the one or more processors cause the one or more processors to perform operations for generating an output token string based on a query input comprising an input token string and one or more data items, the input token string and output token string being strings of tokens selected from a token vocabulary, and the data items being of a modality other than tokens selected from the token vocabulary, the operations comprising:

(a) generating one or more compressed representations of each data item by processing the data item using a modality network which comprises:

an encoder configured to encode the data item to generate an encoded data item, and

a compressed representation generation system arranged to receive the encoded data item and generate an output,

wherein the compressed representation generation system comprises a stack of one or more resampler layers,

each resampler layer being configured to perform an attention operation which employs a key vector, a value vector and a query vector, the key vector, value vector and query vector each being based on at least one of the encoded data item and a latent input which is either an output of the preceding one of the resampler layers or, in the case of the first resampler layer of the stack, a set of input latent values, at least one of the key vector, value vector and query vector being based on both the encoded data item and the latent input,

the output of the modality network being based on an output of the last resampler layer of the stack of resampler layers;

(b) generating a prompt input comprising the input token string of the query input; and

(c) inputting the prompt input and the compressed representation of each data item to a data-item-token processing model configured to generate the output token string based on the prompt input and the compressed representation of each data item.

22 . The system of claim 21 in which the output token string is the response to a query about the content of a subject data item which is one of the data items in the query input, the query being defined based on the input token string.

23 . The system of claim 21 in which the query input comprises, in addition to the subject data item, one or more task example portions which each include a respective data item and a respective section of the input token string, and for each task example portion the respective section of the input token string is the response to the query when the query is about the content of the respective data item.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 6, 2025
From: DEEPMIND TECHNOLOGIES LIMITED
To: GDM HOLDING LLC
Reel/Frame 071498/0210 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 28, 2023
From: ALAYRAC, JEAN-BAPTISTE; DONAHUE, JEFFREY; LENC, KAREL; SIMONYAN, KAREN; REYNOLDS, MALCOLM KEVIN; LUC, PAULINE; MENSCH, ARTHUR; BARR, IAIN; MIECH, ANTOINE; HASSON, YANA ELIZABETH; MILLICAN, KATHERINE ELIZABETH; RING, ROMAN
To: DEEPMIND TECHNOLOGIES LIMITED
Reel/Frame 064416/0056 →
Continuity (2)
Provisional Application 63336192 · Apr 28, 2022
Related Publication 20230350936A1 · Nov 2, 2023
References Cited (168)
US 8875110B2 · Lee · 2014 [cited by examiner]
US 11615084B1 · Echeverria · 2023 [cited by examiner]
US 20210232773A1 · Wang et al. · 2021 [cited by applicant]
US 20210365633A1 · Wagner · 2021 [cited by examiner]
US 20220237380A1 · Chen et al. · 2022 [cited by applicant]
US 20220245156A1 · Kulkarni · 2022 [cited by examiner]
WO WO2021010203 · 2021 [cited by applicant]
International Preliminary Report on Patentability in International Appln. No. PCT/EP2023/061330, mailed on Nov. 7, 2024, 14 pages. [cited by applicant]
Aghajanyan et al., “CM3: A Causal Masked Multimodal Model of the Internet,” CoRR, Submitted on Jan. 19, 2022, arXiv:2201.07520v1, 20 pages. [cited by applicant]
Alayrac et al., “Self-Supervised MultiModal Versatile Networks,” 34th Conference on Neural Information Processing Systems (NeurIPS 2020), 2020, 13 pages. [cited by applicant]
Antol et al., “VQA: Visual Question Answering,” 2015 IEEE International Conference on Computer Vision (ICCV), 2015, pp. 2425-2433. [cited by applicant]
Bachlechner et al., “ReZero is All You Need: Fast Convergence a Large Depth,” Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence (UAI 2021), 2021, PMLR 161, pp. 1352-1361. [cited by applicant]
Bain et al,. “Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval,” 2021 IEEE/CVF International Conference on Computer Vision (ICCV), 2021, pp. 1728-1738. [cited by applicant]
Bao et al., “VLMo: Unified Vision-Language Pre-Training with Mixture-of-Modality-Experts,” CoRR, Submitted on Nov. 3, 2021, arXiv:2111.02358v1, 15 pages. [cited by applicant]
Bertinetto et al., “Learning feed-forward one-shot learners,” 30th Conference on Neural Information Processing Systems (NIPS 2016), 2016, 9 pages. [cited by applicant]
Bertinetto et al., “Meta-learning with differentiable closed-form solvers,” CoRR, Submitted on May 21, 2018, arXiv:1805.08136v1, 11 pages. [cited by applicant]
Bridle, “Probabilistic Interpretation of Feedforward Classification Network Outputs, with Relationships to Statistical Pattern Recognition,” Neurocomputing, 1990. [cited by applicant]
Brock et al., “High-Performance Large-Scale Image Recognition Without Normalization,” CoRR, Submitted on Feb. 11, 2021, arXiv:2102.06171v1, 22 pages. [cited by applicant]
Brown et al., “Language Models are Few-Shot Learners,” 34th Conference on Neural Information Processing Systems (NeurIPS 2020), 2020, 25 pages. [cited by applicant]
Buolamwini et al., “Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification,” Proceedings of the Ist Conference on Fairness, Accountability, and Transparency, 2018, 15 pages. [cited by applicant]
Carion et al., “End-to-End Object Detection with Transformers,” European Conference on Computer Vision, 2020, 17 pages. [cited by applicant]
Changpinyo et al., “Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training to Recognize Long-Tail Visual Concepts,” Changpinyo, Soravit et al. “Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize L… [cited by applicant]
Chen et al., “Microsoft COCO Captions: Data Collection and Evaluation Server,” CoRR, Submitted on Apr. 1, 2015, arXiv:1504.00325v1, 7 pages. [cited by applicant]
Chen et al., “UNITER: Universal Image-TExt Representation Learning,” European Conference on Computer Vision, 2020, 17 pages. [cited by applicant]
Cho et al., “Unifying Vision-and-Language Tasks via Text Generation,” Proceedings of the 38th International Conference on Machine Learning, PMLR 139, 2021, 12 pages. [cited by applicant]
Chowdhery et al., “PaLM: Scaling Language Modeling with Pathways,” CoRR, Submitted on Apr. 5, 2022, arXiv:2204.02311v1, 83 pages. [cited by applicant]
Dai et al., “Enabling Multimodal Generation on CLIP via Vision-Language Knowledge Distillation,” Findings of the Association for Computational Linguistics: ACL 2022, May 2022, pp. 2383-2395. [cited by applicant]
Das et al., “Visual Dialog.,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 326-335. [cited by applicant]
Desai et al., “VirTex: Learning Visual Representations from Textual Annotations,” 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 11162-11173. [cited by applicant]
Devlin et al., “BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding,” CoRR, Submitted on Oct. 11, 2018, arXiv:1810.04805v1, 14 pages. [cited by applicant]
DeVries et al., “Does Object Recognition Work for Everyone?,” CoRR, Submitted on Jun. 6, 2019, arXiv:1906.02659v1, 8 pages. [cited by applicant]
Doersch et al., “CrossTransformers: spatially-aware few-shot transfer,” 3rth Conference on Neural Information Processing Systems (NeurIPS 2020), 2020, 13 pages. [cited by applicant]
Donahue et al., “Long-Term Recurrent Convolutional Long-Term Recurrent Convolutional Networks for Visual Recognition and Description,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, pp. 2625-2634. [cited by applicant]
Eichenberg et al. ,“MAGMA—Multimodal Augmentation of Generative Models through Adapter-based Finetuning,” CoRR, Submitted on Dec. 9, 2021, arXiv:2112.05253v1, 11 pages. [cited by applicant]
Finn et al., “Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks,” Proceedings of the 34th International Conference on Machine Learning, PMRL 70, 2017, 10 pages. [cited by applicant]
Fu et al., “VIOLET : End-to-End Video-Language Transformers with Masked Visual-token Modeling,” CoRR, Submitted on Nov. 24, 2021, arXiv:2111.12681v1, 19 pages. [cited by applicant]
Gan et al., “Large-Scale Adversarial Training for Vision-and-Language Representation Learning,” 34th Conference on Neural Information Processing Systems (NeurIPS 2020), 2020, 13 pages. [cited by applicant]
Gebru et al., “Datasheets for Datasets,” Communications of the ACM, 2021, 64(12):86-92. [cited by applicant]
GitHub.com [online], “google / jax,” available on or before Nov. 22, 2018 via Internet Archive: Wayback Machine URL<https://web.archive.org/web/20181122121702/https://github.com/google/jax>, retrieved on Jul. 26, 2024, … [cited by applicant]
GitHub.com [online], “google-deepmind / dm-haiku,” available on or before Mar. 5, 2020, via Archive: Wayback Machine URL<https://web.archive.org/web/20200305221206/http://github.com/deepmind/dm-haiku>, retrieved on Jul.… [cited by applicant]
Gordon et al., “Meta-Learning Probabilistic Inference For Prediction,” CoRR, Submitted on Nov. 28, 2018, arXiv:1805.09921v3, 21 pages. [cited by applicant]
Graves, “Generating Sequences With Recurrent Neural Networks,” CoRR, Submitted on Aug. 4, 2013, arXiv:1308.0850v1, 43 pages. [cited by applicant]
Griffiths et al., “Doing more with less: meta-reasoning and meta-learning in humans and machines,” Current Opinion in Behavioral Sciences, 2019, 29:24-30. [cited by applicant]
Gui et al., “KAT: A Knowledge Augmented Transformer for Vision-and-Language,” CoRR, Submitted on Dec. 16, 2021, arXiv:2112.08614v1, 11 pages. [cited by applicant]
Gurari et al., “VizWiz Grand Challenge: Answering Visual Questions from Blind People,” 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, pp. 3608-3617. [cited by applicant]
Haviv et al., “Transformer Language Models without Positional Encodings Still Learn Positional Information,” CoRR, Submitted on Mar. 30, 2022, arXiv:2203.16634v1, 7 pages. [cited by applicant]
Hendricks et al., “Decoupling the Role of Data, Attention, and Losses in Multimodal Transformers,” Transactions of the Association for Computational Linguistics, 2021, vol. 9, pp. 570-585. [cited by applicant]
Hendricks et al., “Women Also Snowboard: Overcoming Bias in Captioning Models,” European Conference on Computer Vision (ECCV 2018), 2018, 17 pages. [cited by applicant]
Hendrycks et al., “Gaussian Error Linear Units (GELUs),” CoRR, Submitted on Nov. 11, 2018, arXiv:1606.08415v3, 9 pages. [cited by applicant]
Hochreiter et al., “Long Short-Term Memory,” Neural Computation, 1997, 9(8):1735-1780. [cited by applicant]
Hoffmann et al., “Training Compute-Optimal Large Language Models,” CoRR, Submitted on Mar. 29, 2022, arXiv:2203.15556v1, 36 pages. [cited by applicant]
Houlsby et al., “Parameter-Efficient Transfer Learning for NLP,” Proceedings of the 36th International Conference on Machine Learning, PMLR 97, 2019, 10 pages. [cited by applicant]
Howard et al., “Universal Language Model Fine-tuning for Text Classification,” CoRR, Submitted on May 14, 2018, arXiv:1801.06146v2, 12 pages. [cited by applicant]
Hu et al., “Scaling Up Vision-Language Pre-training for Image Captioning,” CoRR, Submitted on Nov. 24, 2021, arXiv:2111.12233v1, 14 pages. [cited by applicant]
Huang et al., “Attention on Attention for Image Captioning,” 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 4634-4643. [cited by applicant]
International Search Report and Written Opinion in International Appln. No. PCT/EP2023/061330, mailed on Aug. 4, 2023, 21 pages. [cited by applicant]
Islam et al., “Global Pooling, More than Meets the Eye: Position Information is Encoded Channel-Wise in CNNs,” 2021 IEEE/CVF International Conference on Computer Vision (ICCV), 2021, pp. 793-801. [cited by applicant]
Jaegle et al., “Perceiver: General Perception with Iterative Attention,” CoRR, Submitted on Mar. 4, 2021, arXiv:2103.03206v1, 16 pages. [cited by applicant]
Jain et al., “MURAL: Multimodal, Multitask Retrieval Across Languages,” CoRR, Submitted on Sep. 10, 2021, arXiv:2109.05125v1, 15 pages. [cited by applicant]
Jia et al., “Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision,” CoRR, Submitted on Feb. 11, 2021, arXiv:2102.05918v1, 11 pages. [cited by applicant]
Jozefowicz et al., “Exploring the Limits of Language Modeling,” CoRR, Submitted on Feb. 7, 2016, arXiv:1602.02410v1, 11 pages. [cited by applicant]
Kaplan et al., “Scaling Laws for Neural Language Models,” CoRR, Submitted on Jan. 23, 2020, arXiv:2001.08361v1, 30 pages. [cited by applicant]
Kiela et al., “The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes,” 34th Conference on Neural Information Processing Systems (NeurIPS 2020), 2020, 14 pages. [cited by applicant]
Kumar et al., “Affective Feedback Synthesis Towards Multimodal Text and Image Data,” CoRR, Submitted on Mar. 23, 2022, arXiv:2203.12692v1, 23 pages. [cited by applicant]
Lester et al., “The Power of Scale for Parameter-Efficient Prompt Tuning,” CoRR, Submitted on Apr. 18, 2021, arXiv:2104.08691v1, 13 pages. [cited by applicant]
Li et al., “Align before Fuse: Vision and Language Representation Learning with Momentum Distillation,” 35h Conference on Neural Information Processing Systems (NeurIPS 2021), 12 pages. [cited by applicant]
Li et al., “BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation,” CoRR, Submitted on Jan. 28, 2022, arXiv:2201.12086v1, 12 pages. [cited by applicant]
Li et al., “HERO: Hierarchical Encoder for Video+Language Omni-representation Pre-training,” CoRR, Submitted on May 1, 2020, arXiv:2005.00200v1, 14 pages. [cited by applicant]
Li et al., “Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks,” European Conference on Computer Vision, 2020, 16 pages. [cited by applicant]
Li et al., “Prefix-Tuning: Optimizing Continuous Prompts for Generation,” CoRR, Submitted on Jan. 1, 2021, arXiv:2101.00190v1, 15 pages. [cited by applicant]
Lippe et al., “A Multimodal Framework for the Detection of Hateful Memes,” CoRR, Submitted on Dec. 23, 2020, arXiv:2012.12871v1, 14 pages. [cited by applicant]
Liu et al., “Optimization of image description metrics using policy gradient methods,” CoRR, Submitted on Dec. 1, 2016, arXiv:1612.00370v1, 10 pages. [cited by applicant]
Liu et al., “What Makes Good In-Context Examples for GPT-3?,” CoRR, Submitted on Jan. 17, 2021, arXiv:2101.06804v1, 12 pages. [cited by applicant]
Lu et al., “ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks,” 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), 2019, 11 pages. [cited by applicant]
Luo et al., “UniViLM: A Unified Video and Language Pre-Training Model for Multimodal Understanding and Generation,” CoRR, Submitted on Feb. 15, 2020, arXiv:2002.06353v1, 11 pages. [cited by applicant]
Luo et al., “VC-GPT: Visual Conditioned GPT for End-to-End Generative Vision-and-Language Pre-training,” CoRR, Submitted on Jan. 30, 2022, arXiv:2201v1, 10 pages. [cited by applicant]
Marino et al., “OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledge,” 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 3195-3204. [cited by applicant]
McCloskey et al., “Catastrophic Interference in Connectionist Networks: The Sequential Learning Problem,” Psychology of Learning and Motivation, 1989, vol. 24, pp. 109-165. [cited by applicant]
Menick et al., “Teaching language models to support answers with verified quotes,” CoRR, Submitted on Mar. 21, 2022, arXiv:2203.11147v1, 40 pages. [cited by applicant]
Miech et al., “End-to-End Learning of Visual Representations from Uncurated Instructional Videos,” 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, pp. 9879-9889. [cited by applicant]
Miech et al., “RareAct: A video dataset of unusual interactions,” CoRR, Submitted on Aug. 3, 2020, arXiv:2008.01018v1, 6 pages. [cited by applicant]
Mikolov et al., “Recurrent neural network-based language model,” Interspeech, Sep. 20910, pp. 1045-1048. [cited by applicant]
Min et al., “Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?,” CoRR, Submitted on Feb. 25, 2022, arXiv:2202.12837v1, 14 pages. [cited by applicant]
Mitchell et al., “Model Cards for Model Reporting,” Proceedings of the Conference on Fairness, Accountability, and Transparency, Jan. 2019, pp. 220-229. [cited by applicant]
Mokady et al., “ClipCap: CLIP Prefix for Image Captioning,” CoRR, Submitted on Nov. 18, 2021, arXiv:2111.09734v1, 10 pages. [cited by applicant]
Murahari et al., “Large-scale Pretraining for Visual Dialog: A Simple State-of-the-Art Baseline,” European Conference on Computer Vision, 2020, 18 pages. [cited by applicant]
Perez et al., “Red Teaming Language Models with Language Models,” CoRR, Submitted on Feb. 7, 2022, arXiv:2202.03286v1, 31 pages. [cited by applicant]
Perez et al., “True Few-Shot Learning with Language Models,” 35th Conference on Neural Information Processing Systems (NeurIPS 2021), 2021, 17 pages. [cited by applicant]
Pham et al., “Combined Scaling for Zero-shot Transfer Learning,” CoRR, Submitted on Nov. 19, 2021, arXiv:2111.10050v1, 18 pages. [cited by applicant]
Press et al., “Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation,” International Conference on Learning Representations (ICLR 2022), 2022, 25 pages. [cited by applicant]
Qiao et al., “Winner Team Mia at TextVQA Challenge 2021: Vision-and-Language Representation Learning with Pre-trained Sequence-to-Sequence Model,” CoRR, Submitted on Jun. 24, 2021, arXiv:2106.15332v1, 2 pages. [cited by applicant]
Radford et al., “Leaming Transferable Visual Models From Natural Language Supervision,” CoRR, Submitted on Feb. 26, 2021, arXiv:2103.00020v1, 48 pages. [cited by applicant]
Rae et al., “Scaling Language Models: Methods, Analysis & Insights from Training Gopher,” CoRR, Submitted on Dec. 8, 2021, arXiv:2112.11446v1, 118 pages. [cited by applicant]
Raffel et al., “Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer,” CoRR, Submitted on Oct. 23, 2019, arXiv:1910.10683v1, 52 pages. [cited by applicant]
Rajbhandari et al., “ZeRO: Memory optimizations Toward Training Trillion Parameter Models,” SC20: International Conference for High Performance Computing, Networking, Storage and Analysis, 2020, 16 pages. [cited by applicant]
Ramesh et al., “Hierarchical Text-Conditional Image Generation with CLIP Latents,” CoRR, Submitted on Apr. 13, 2022, arXiv:2204.06125v1, 27 pages. [cited by applicant]
Rennie et al., “Self-critical Sequence Training for Image Captioning,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, Pages pp. 7008-7024. [cited by applicant]
Requeima et al., “Fast and Flexible Multi-Task Classification Using Conditional Neural Adaptive Processes,” 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), 2019, 12 pages. [cited by applicant]
Reynolds et al., “Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm,” CoRR, Submitted on Feb. 15, 2021, arXiv:2102.07350v1, 10 pages. [cited by applicant]
Rudinger et al., “Gender Bias in Coreference Resolution,” CoRR, Submitted on Apr. 25, 2018, arXiv:1804.09301v1, 7 pages. [cited by applicant]
Russakovsky et al., “ImageNet Large Scale Visual Recognition Challenge,” CoRR, Submitted on Jan. 30, 2015, arXiv:1409.0575v3, 43 pages. [cited by applicant]
Sanh et al. “Multitask Prompted Training Enables Zero-Shot Task Generalization,” CoRR, Submitted on Mar. 17, 2022, arXiv:2110.08207v3, 216 pages. [cited by applicant]
Schuhmann et al., “LAION-400M: Open Dataset of CLIP-Filtered 400 million Image-Text Pairs,” CoRR, Submitted on Nov. 3, 2021, arXiv:2111.02114v1, 5 pages. [cited by applicant]
Schwemmer et al., “Diagnosing Gender Bias in Image Recognition Systems,” Socius, Jan.-Dec. 2020, vol. 6, pp. 1-17. [cited by applicant]
Sharma et al., “Conceptual Captions: A Cleaned, Hypernymed, Image Alt-text Dataset For Automatic Image Captioning,” Proceedings of the 6th Annual Meeting of the Association for Computational Linguistics, vol. 1: Long Pa… [cited by applicant]
Shoeybi et al., “Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism,” CoRR, Submitted on Sep. 17, 2019, arXiv:2104.08691v1, 15 pages. [cited by applicant]
Singh et al., “FLAVA: A Foundational Language And Vision Alignment Model,” CoRR, Submitted on Dec. 8, 2021, arXiv:2112.04482v1, 17 pages. [cited by applicant]
Singh et al., “Towards VQA Models That Can Read,” 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 8317-8326. [cited by applicant]
slideslive.com [online], “Few-shot Classification by Recycling Deep Learning,” Invited Talk at the S3D-OLAD Workshop, ICLR, Jul. 3, 2021, retrieved on Jul. 12, 2024, retrieved from URL<https://slideslive.com/38955350/fe… [cited by applicant]
Smaira et al., “A Short Note on the Kinetics-700-2020 Human Action Dataset,” CoRR, Submitted on Oct. 21, 2020, arXiv:2010.10864v1, 5 pages. [cited by applicant]
Snell et al., , “Prototypical Networks for Few-shot Learning,” 31st Conference on Neural Information Processing Systems (NIPS 2017), 2017, 11 pages. [cited by applicant]
So et al., “Primer: Searching for Efficient Transformers for Language Modeling,” CoRR, Submitted on Sep. 17, 2021, arXiv:2109.08668v1, 35 pages. [cited by applicant]
Strubell et al., “Energy and Policy Considerations for Deep Learning in NLP,” CoRR, Submitted on Jun. 5, 2019, arXiv:1906.02243v1, 6 pages. [cited by applicant]
Su et al., “VL-BERT: Pre-training of Generic Visual-Linguistic Representations,” CoRR, Submitted on Aug. 22, 2019, arXiv:1908.08530v1, 13 pages. [cited by applicant]
Sun et al., “VideoBERT: A Joint Model for Video and Language Representation Learning,” 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 7464-7473. [cited by applicant]
Sutskever et al., “Generating Text with Recurrent Neural Networks,” Proceedings of the 28th International Conference on Machine Learning, 2011, 8 pages. [cited by applicant]
Tan et al., “LXMERT: Learning Cross-Modality Encoder Representations from Transformers,” Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference … [cited by applicant]
Thomee et al., “YFCC100M: the new data in multimedia research,” Communications of the ACM, Feb. 2016, 59(2):64-73. [cited by applicant]
Thoppilan et al., “LaMDA: Language Models for Dialog Applications,” CoRR, Submitted on Jan. 20, 2022, arXiv:2201.08239v1, 47 pages. [cited by applicant]
Tian et al., “Rethinking Few-shot Image Classification: A Good Embedding is All You Need?,” European Conference on Computer Vision, 2020, pp. 1-17. [cited by applicant]
Touvron et al., “Fixing the train-test resolution discrepancy,” 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), 2019, 11 pages. [cited by applicant]
Tsimpoukelli et al., “Multimodal Few-Shot Learning with Frozen Language Models,” 35th Conference on Neural Information Processing Systems (NeurIPS 2021), 2021, 13 pages. [cited by applicant]
Vaswani et al., “Attention Is All You Need,” 31st Conference on Neural Information Processing Systems (NIPS 2017), 2017, 11 pages. [cited by applicant]
Vinyals et al., “Matching Networks for One Shot Learning,” 30th Conference on Neural Information Processing Systems (NIPS 2016), 2016, 9 pages. [cited by applicant]
Vinyals et al., “Show and Tell: A Neural Image Caption Generator,” International Conference on Computer Vision, 2015, pp. 3156-3164. [cited by applicant]
Wang et al., “All in One: Exploring Unified Video-Language Pre-training,” CoRR, Submitted on Mar. 14, 2022, arXiv:2203.07303v1, 18 pages. [cited by applicant]
Wang et al., “SimVLM: Simple Visual Language Model Pretraining with Weak Supervision,” CoRR, Submitted on Aug. 24, 2021, arXiv:2108.10904v1, 16 pages. [cited by applicant]
Wang et al., “UFO: A UniFied Transformer for Vision-Language Representation Learning,” CoRR, Submitted on Nov. 19, 2021, arXiv:2111.10023v1, 14 pages. [cited by applicant]
Wang et al., “Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework,” CoRR, Submitted on Feb. 7, 2022, arXiv:2202.03052v1, 23 pages. [cited by applicant]
Wang et al., “VATEX: A Large-Scale, High-Quality Multilingual Dataset for Video-and-Language Research,” International Conference on Computer Vision, 2019, pp. 4581-4591. [cited by applicant]
Wang et al., “VD-BERT: A Unified Vision and Dialog Transformer with BERT,” Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), Nov. 2020, pp. 3325-3338. [cited by applicant]
Wang et al., “What Language Model Architecture and Pretraining Objective Work Best for Zero-Shot Generalization?,” CoRR, Submitted on Apr. 12, 2022, arXiv:2204.05832v1, 26 pages. [cited by applicant]
Wei et al., “Finetuned Language Models Are Zero-Shot Learners,” CoRR, Submitted on Sep. 3, 2021, arXiv:2109.01652v1, 46 pages. [cited by applicant]
Weidinger et al., “Ethical and social risks of harm from Language Models,” CoRR, Submitted on Dec. 8, 2021, arXiv:2112.04359v1, 64 pages. [cited by applicant]
Wortsman et al., “Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time,” CoRR, Submitted on Mar. 10, 2022, arXiv:2203.05482v1, 35 pages. [cited by applicant]
Wu et al., “STAR: A Benchmark for Situated Reasoning in Real-World Videos,” 35th Conference on Neural Information Processing Systems (NeurIPS 2021), 2021, 13 pages. [cited by applicant]
Xiao et al., “NExT-QA: Next Phase of Question-Answering to Explaining Temporal Actions,” IEEE Computer Vision and Pattern Recognition, 2021, pp. 9777-9786. [cited by applicant]
Xu et al., “Video Question Answering via Gradually Refined Attention over Appearance and Motion,” MM '17: Proceedings of the 25th ACM international conference on Multimedia, Oct. 2017, 9 pages. [cited by applicant]
Xu et al., “VLM: Task-agnostic Video-Language Model Pre-training for Video Understanding,” CoRR, Submitted on May 20, 2021, arXiv:2105.09996v1, 13 pages. [cited by applicant]
Xu et al., “ZeroPrompt: Scaling Prompt-Based Pretraining to 1,000 Tasks Improves Zero-Shot Generalization,” CoRR, Submitted on Jan. 18, 2022, arXiv:2201.06910v1, 23 pages. [cited by applicant]
Yan et al., “Achieving Human Parity on Visual Question Answering,” CoRR, Submitted on Nov. 17, 2021, arXiv:2111.08896v1, 27 pages. [cited by applicant]
Yan et al., “Multiview Transformers for Video Recognition,” CoRR, Submitted on Jan. 12, 2022, arXiv:2201.04288v1, 15 pages. [cited by applicant]
Yang et al., “An Empirical Study of GPT-3 for Few-Shot Knowledge-Based VQA,” The Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI-22), Jun. 2022, pp. 3081-3089. [cited by applicant]
Yang et al., “Just Ask: Learning to Answer Questions from Millions of Narrated Videos,” International Conference on Computer Vision, 2021, pp. 1686-1697. [cited by applicant]
Yang et al., “TAP: Text-Aware Pre-training for Text-VQA and Text-Caption,” IEEE Computer Vision and Pattern Recognition, 2021, pp. 8751-8761. [cited by applicant]
Yao et al., “FILIP: Fine-grained Interactive Language-Image Pre-Training,” CoRR, Submitted on Nov. 9, 2021, arXiv:2111.07783v1, 18 pages. [cited by applicant]
Young et al., “From image descriptions to visual denotations: New similarity metrics for semantic inference over event descriptions,” Transactions of the Association for Computational Linguistics, vol. 2, 2014, pp. 67-7… [cited by applicant]
Yuan et al., “Florence: A New Foundation Model for Computer Vision,” CoRR, Submitted on Nov. 22, 2021, arXiv:2111.11432v1, 17 pages. [cited by applicant]
Zaken et al., “BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models,” CoRR, Submitted on Jun. 18, 2021, arXiv:2106.10199v1, 8 pages. [cited by applicant]
Zellers et al., “Merlot Reserve: Neural Script Knowledge through Vision and Language and Sound,” IEEE Computer Vision and Pattern Recognition, 2022, pp. 16375-16387. [cited by applicant]
Zellers et al., “Merlot: Multimodal Neural Script Knowledge Models,” 35th Conference on Neural Information Processing Systems (NeurIPS 2021), 18 pages. [cited by applicant]
Zeng et al., “Socratic Models: Composing Zero-Shot Multimodal Reasoning with Language,” CoRR, Submitted on Apr. 1, 2022, arXiv:2204.00598v1, 20 pages. [cited by applicant]
Zhai et al., “LiT: Zero-Shot Transfer with Locked-image Text Tuning,” CoRR, Submitted on Nov. 15, 2021, arXiv:2111.07991v1, 19 pages. [cited by applicant]
Zhai et al., “Scaling Vision Transformers,” CoRR, Submitted on Jun. 8, 2021, arXiv:2106.04560v1, 31 pages. [cited by applicant]
Zhao et al., “Calibrate Before Use: Improving Few-Shot Performance of Language Models,” Proceedings of the 38th International Conference on Machine Learning, PMLR 139, 2021, 10 pages. [cited by applicant]
Zhao et al., “Understanding and Evaluating Racial Biases in Image Captioning,” IEEE Computer Vision and Pattern Recognition, 2021, pp. 14830-14840. [cited by applicant]
Zhou et al., “Towards Automatic Learning of Procedures from Web Instructional Videos,” The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18), 2018, pp. 7590-7598. [cited by applicant]
Zhou et al., “Unified Vision-Language Pre-Training for Image Captioning and VQA,” The Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI-20), 2020, pp. 13041-13049. [cited by applicant]
Zhu et al., “ActBERT: Learning Global-Local Video-Text Representations,” IEEE Computer Vision and Pattern Recognition, 2020, pp. 8746-8755. [cited by applicant]
Zhu et al., “Enhance Multimodal Transformer With External Label And In-Domain Pretrain: Hateful Meme Challenge Winning Solution,” CoRR, Submitted on Dec. 15, 2020, arXiv:2012.08290v1, 10 pages. [cited by applicant]
Zhu et al., “Uni-Perceiver: Pre-training Unified Architecture for Generic Perception for Zero-shot and Few-shot Tasks,” CoRR, Submitted on Dec. 2, 2021, arXiv:2112.01522v1, 15 pages. [cited by applicant]
Zhu et al., “Vatex Video Captioning Challenge 2020: Multi-View Features and Hybrid Reward Strategies for Video Captioning,” CoRR, Submitted on Jun. 9, 2020, arXiv:1910.11102v3, 4 pages. [cited by applicant]
Zintgraf et al., “Fast Context Adaptation via Meta-Learning,” Proceedings of the 36th International Conference on Machine Learning, PMLR 97, 2019, 10 pages. [cited by applicant]
Zohourianshahzadi et al., “Neural attention for image captioning: review of outstanding methods,” Artificial Intelligence Review, 2022, available online Nov. 29, 2021, 2022(55):3833-3862. [cited by applicant]
Office Action in Canadian Appln. No. 3,248,013, Nov. 26, 2025, 4 pages. [cited by applicant]
Office Action in Australian Appln. No. 2023258672, mailed on Jan. 15, 2026, 4 pages. [cited by applicant]
Jaegle et al., “Perceiver IO: A General Architecture for Structured Inputs & Outputs,” CoRR, submitted on Mar. 15, 2022, arXiv:2107.14795v3, 30 pages. [cited by applicant]
Office Action in Japanese Appln. No. 2024-563398, mailed on Feb. 17, 2026, 6 pages (with English translation). [cited by applicant]