Papers by Radu Soricut
TeaForN: Teacher-Forcing with N-grams (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to address exposure bias and lack of differentiability in sequence generation models with teacherforcing have failed to address these issues. |
| Approach: | They propose a method that uses a stack of N decoders to decode along a secondary time axis and allows model-parameter updates based on N prediction steps. |
| Outcome: | Empirically, teaForN boosts generation quality on one Machine Translation benchmark, WMT 2014 English-French, and two News Summarization benchmarks, CNN/Dailymail and Gigaword. |
MaXM: Towards Multilingual Visual Question Answering (2023.findings-emnlp)
Copied to clipboard
Soravit Changpinyo, Linting Xue, Michal Yarom, Ashish Thapliyal, Idan Szpektor, Julien Amelot, Xi Chen, Radu Soricut
| Challenge: | Visual Question Answering (VQA) has been studied in the English language, but in other languages it would require a considerable amount of resources. |
| Approach: | They propose scalable solutions to multilingual visual question answering using an English language framework and an annotation protocol. |
| Outcome: | The proposed framework reduces human annotation efforts and creates a test-only VQA benchmark in 7 languages. |
Beyond Instructional Videos: Probing for More Diverse Visual-Textual Grounding on YouTube (2020.emnlp-main)
Copied to clipboard
| Challenge: | a representative pretraining model is fit to a diverse YouTube8M dataset . a priori, this domain is relatively easy for instructional videos . |
| Approach: | They fit a representative pretraining model to a YouTube8M dataset and examine its success and failure cases. |
| Outcome: | The proposed model can be trained on more diverse video corpora and achieve high performance on many video understanding tasks. |
End-to-end Dense Video Captioning as Sequence Generation (2022.coling-1)
Copied to clipboard
| Challenge: | Existing methods for dense video captioning use a two-stage generative process . but, more complex tasks are not able to fully utilize this powerful paradigm . |
| Approach: | They propose to model two subtasks of dense video captioning as one sequence generation task and predict the events and the corresponding descriptions. |
| Outcome: | Experiments on YouCook2 and ViTT show that the proposed model can be used on any video platform. |
SHAPED: Shared-Private Encoder-Decoder for Text Style Adaptation (N18-1)
Copied to clipboard
| Challenge: | Experimentally, we find that the proposed models consistently outperform models that encapsulate single-style or average-style language generation capabilities. |
| Approach: | They propose a family of model architectures capable of capturing both generic language characteristics via shared model parameters, as well as particular style characteristics via private model parameters. |
| Outcome: | The proposed models outperform models that encapsulate single-style or average-style language generation capabilities. |
Conceptual Captions: A Cleaned, Hypernymed, Image Alt-text Dataset For Automatic Image Captioning (P18-1)
Copied to clipboard
| Challenge: | Practical applications of automatic image description systems include leveraging descriptions for image indexing or retrieval, and helping those with visual impairments by transforming visual signals into information that can be communicated via text-to-speech technology. |
| Approach: | They propose to extract and filter image caption annotations from billions of webpages and use them to train models. |
| Outcome: | The proposed model architectures perform better when trained on the Conceptual Captions dataset. |
COSMic: A Coherence-Aware Generation Metric for Image Descriptions (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to evaluate captions have limited learning of their output . previous methods focused on n-gram measures of similarity to reference output based on a ngram of similarities to the output metric. |
| Approach: | They propose a first discourse-aware learned generation metric for evaluating image descriptions. |
| Outcome: | The proposed metric predicts human ratings of captions on out-of-domain images. |
All You May Need for VQA are Image Captions (2022.naacl-main)
Copied to clipboard
| Challenge: | Visual Question Answering (VQA) has benefited from increasingly sophisticated models, but has not enjoyed the same level of engagement in terms of data creation. |
| Approach: | They propose a method that automatically derives VQA examples at volume by leveraging existing image-caption annotations combined with neural models for textual question generation. |
| Outcome: | The proposed method improves state-of-the-art zero-shot accuracy by double digits and achieves robustness that lacks in the same model trained on human-annotated VQA data. |
Informative Image Captioning with External Sources of Information (P19-1)
Copied to clipboard
| Challenge: | Current captioning models are trained to generate captions that only contain common object names, thus falling short on an important “informativeness” dimension. |
| Approach: | They propose a mechanism for integrating image information and fine-grained labels into a caption that describes the image in a fluent and informative manner. |
| Outcome: | The proposed model integrates image information with fine-grained labels to produce fluent captions . it can control the appearance of these labels in the output, resulting in fluent and informative captions. |
Multimodal Pretraining for Dense Video Captioning (2020.aacl-main)
Copied to clipboard
| Challenge: | a billion hours of videos are being watched on YouTube every day . videos are difficult to skim through, making it harder to quickly target the relevant part(s) of a video. |
| Approach: | They propose to use a video timeline tag dataset to generate time-stamped annotations for videos . they propose to pretrain and finetune captioning models using YouCook2 and ViTT . |
| Outcome: | The proposed model generalizes well and is robust over a wide variety of instructional videos. |
H-Transformer-1D: Fast One-Dimensional Hierarchical Attention for Sequences (2021.acl-long)
Copied to clipboard
| Challenge: | Existing hierarchical methods to compute attention are superior to sub-quadratic ones . a particular type of attention, called multi-head scaled dot-product attention, is one of the main components of the Transformer architecture . |
| Approach: | They propose a hierarchical method to compute attention in the Transformer architecture . they perform extensive experiments to show that it captures hierarchic structure in sequences . |
| Outcome: | The proposed method outperforms sub-quadratic models on the Long Range Arena benchmark by over +6 points on average. |
Denoising Large-Scale Image Captioning from Alt-text Data Using Content Selection Models (2022.coling-1)
Copied to clipboard
| Challenge: | Recent approaches to training large-scale image captioning (IC) models often fall short in terms of performance in leveraging noisy datasets in favor of clean annotations. |
| Approach: | They propose a technique that breaks down the task into two smaller, more controllable tasks - skeleton prediction and skelet-based caption generation. |
| Outcome: | The proposed method can generate better and denoised captions when using noisy datasets. |
Quality Estimation for Image Captions Based on Large-scale Human Evaluations (2021.naacl-main)
Copied to clipboard
| Challenge: | a problem with automatic image captioning is that it produces low quality captions when used in the wild. |
| Approach: | They propose to model caption quality from a human perspective and *without* access to ground-truth references. |
| Outcome: | The proposed model can detect and filter out low-quality captions on previously unseen images. |
Cross-modal Coherence Modeling for Caption Generation (2020.acl-main)
Copied to clipboard
| Challenge: | Existing methods for image captioning do not guarantee consistent image-text relations . current models do not provide enough data for training robust captioning models . |
| Approach: | They use an annotation protocol specifically devised for capturing image–caption coherence relations to study image captioning. |
| Outcome: | The proposed protocol improves image captioning models with coherence relations . the dataset is large enough to alleviate content hallucinations, the authors show . |
CrossVQA: Scalably Generating Benchmarks for Systematically Testing VQA Generalization (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing visual question answering models leverage spurious biases and take shortcuts to improve performance. |
| Approach: | They propose a semi-automatic framework for generating disentangled shifts by introducing a controllable visual question-answer generation module that generates highly-relevant question-announcer pairs with the desired dataset style. |
| Outcome: | The proposed framework generates highly-relevant and diverse question-answer pairs with the desired dataset style. |
Crossmodal-3600: A Massively Multilingual Multimodal Evaluation Dataset (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies on multilingual image captioning have been hampered by a lack of high-quality evaluation datasets. |
| Approach: | They present a dataset of 3600 images annotated with human-generated captions in 36 languages. |
| Outcome: | The proposed dataset shows that it is feasible to build multilingual image captioning models trained on machine-translated data. |
Decoupled Box Proposal and Featurization with Ultrafine-Grained Semantic Labels Improve Image Captioning and Visual Question Answering (D19-1)
Copied to clipboard
| Challenge: | Object detection is used in vision and language tasks but is expensive to learn . popular models rely on annotating ground-truths for bounding boxes and semantic labels . empirically, object detection leads to effective transfer learning and improved captioning and visual question answering models . |
| Approach: | They examine the effect of decoupling box proposal and featurization on down-stream tasks . they propose a family of "two-stage" object detectors that propose category-agnostic bounding boxes . |
| Outcome: | The proposed method improves image captioning and visual question answering models by leveraging large amounts of labeled annotations. |
Cross-modal Language Generation using Pivot Stabilization for Web-scale Language Coverage (2020.acl-main)
Copied to clipboard
| Challenge: | Existing datasets have extensive labeled data for En-glish, but labeles are extremely scarce in other languages. |
| Approach: | They propose a method that leverages existing annotations with machine translation capabilities to create cross-modal language generation systems at web-scale. |
| Outcome: | The proposed model outperforms other candidates in evaluations performed over 5 target languages. |
ImageInWords: Unlocking Hyper-Detailed Image Descriptions (2024.emnlp-main)
Copied to clipboard
Roopal Garg, Andrea Burns, Burcu Karagol Ayan, Yonatan Bitton, Ceslee Montgomery, Yasumasa Onoe, Andrew Bunner, Ranjay Krishna, Jason Baldridge, Radu Soricut
| Challenge: | generating accurate hyper-detailed image descriptions is challenging for vision-language models trained on web-scraped image-text. |
| Approach: | They propose a data-centric framework for generating hyper-detailed image descriptions using web-scraped image-text. |
| Outcome: | The proposed framework improves on human evaluations on the data, even with only 9k samples. |