More Like this

1. Retrieval as Attention: End-to-end Learning of Retrieval and Reading within a Single Transformer

   https://doi.org/10.18653/v1/2022.emnlp-main.149  

   Jiang, Zhengbao; Gao, Luyu; Wang, Zhiruo; Araki, Jun; Ding, Haibo; Callan, Jamie; Neubig, Graham (January 2022, Association for Computational Linguistics)

   Systems for knowledge-intensive tasks such as open-domain question answering (QA) usually consist of two stages: efficient retrieval of relevant documents from a large corpus and detailed reading of the selected documents. This is usually done through two separate models, a retriever that encodes the query and finds nearest neighbors, and a reader based on Transformers. These two components are usually modeled separately, which necessitates a cumbersome implementation and is awkward to optimize in an end-to-end fashion. In this paper, we revisit this design and eschew the separate architecture and training in favor of a single Transformer that performs retrieval as attention (RAA), and end-to-end training solely based on supervision from the end QA task. We demonstrate for the first time that an end-to-end trained single Transformer can achieve both competitive retrieval and QA performance on in-domain datasets, matching or even slightly outperforming state-of-the-art dense retrievers and readers. Moreover, end-to-end adaptation of our model significantly boosts its performance on out-of-domain datasets in both supervised and unsupervised settings, making our model a simple and adaptable end-to-end solution for knowledge-intensive tasks. 

   more » « less
2. Efficient Implicit Unsupervised Text Hashing using Adversarial Autoencoder

   https://doi.org/10.1145/3366423.3380150  

   Doan, Khoa; Reddy, Chandan K. (April 2020, Proceedings of The Web Conference (WWW))

   Searching for documents with semantically similar content is a fundamental problem in the information retrieval domain with various challenges, primarily, in terms of efficiency and effectiveness. Despite the promise of modeling structured dependencies in documents, several existing text hashing methods lack an efficient mechanism to incorporate such vital information. Additionally, the desired characteristics of an ideal hash function, such as robustness to noise, low quantization error and bit balance/uncorrelation, are not effectively learned with existing methods. This is because of the requirement to either tune additional hyper-parameters or optimize these heuristically and explicitly constructed cost functions. In this paper, we propose a Denoising Adversarial Binary Autoencoder (DABA) model which presents a novel representation learning framework that captures structured representation of text documents in the learned hash function. Also, adversarial training provides an alternative direction to implicitly learn a hash function that captures all the desired characteristics of an ideal hash function. Essentially, DABA adopts a novel single-optimization adversarial training procedure that minimizes the Wasserstein distance in its primal domain to regularize the encoder’s output of either a recurrent neural network or a convolutional autoencoder.We empirically demonstrate the effectiveness of our proposed method in capturing the intrinsic semantic manifold of the related documents. The proposed method outperforms the current state-of-the-art shallow and deep unsupervised hashing methods for the document retrieval task on several prominent document collections. 

   more » « less
3. Easy Regional Contrastive Learning of Expressive Fashion Representations

   Qi, Daiqing; Zhao, Handong; Li, Sheng (December 2024, Advances in Neural Information Processing Systems)

   When learning vision-language models (VLM) for the fashion domain, most existing works design new architectures from vanilla BERT with additional objectives, or perform dense multi-task learning with fashion-specific tasks. Though progress has been made, their architecture or objectives are often intricate and the extendibility is limited.By contrast, with simple architecture (comprising only two unimodal encoders) and just the contrastive objective, popular pre-trained VL models (e.g., CLIP) achieve superior performance in general domains, which are further easily extended to downstream tasks.However, inheriting such benefits of CLIP in the fashion domain is non-trivial in the presence of the notable domain gap. Empirically, we find that directly finetuning on fashion data leads CLIP to frequently ignore minor yet important details such as logos and composition, which are critical in fashion tasks such as retrieval and captioning.In this work, to maintain CLIP's simple architecture and objective while explicitly attending to fashion details, we propose E2 : Easy Regional Contrastive Learning of Expressive Fashion Representations. E2 introduces only a few selection tokens and fusion blocks (just 1.9\% additional parameters in total) with only contrastive losses. Despite lightweight, in our primary focus, cross-modal retrieval, E2 notably outperforms existing fashion VLMs with various fashion-specific objectives.Moreover, thanks to CLIP's widespread use in downstream tasks in general domains (e.g., zero-shot composed image retrieval and image captioning), our model can easily extend these models from general domain to the fashion domain with notable improvement.To conduct a comprehensive evaluation, we further collect data from Amazon Reviews to build a new dataset for cross-modal retrieval in the fashion domain. 

   more » « less
4. VDAM: VAE based domain adaptation for cloud property retrieval from multi-satellite data

   https://doi.org/10.1145/3557915.3561044  

   Huang, Xin; Wang, Chenxi; Purushotham, Sanjay; Wang, Jianwu (November 2022, Proceedings of the 30th International Conference on Advances in Geographic Information Systems (SIGSPATIAL 22))

   Domain adaptation techniques using deep neural networks have been mainly used to solve the distribution shift problem in homogeneous domains where data usually share similar feature spaces and have the same dimensionalities. Nevertheless, real world applications often deal with heterogeneous domains that come from completely different feature spaces with different dimensionalities. In our remote sensing application, two remote sensing datasets collected by an active sensor and a passive one are heterogeneous. In particular, CALIOP actively measures each atmospheric column. In this study, 25 measured variables/features that are sensitive to cloud phase are used and they are fully labeled. VIIRS is an imaging radiometer, which collects radiometric measurements of the surface and atmosphere in the visible and infrared bands. Recent studies have shown that passive sensors may have difficulties in prediction cloud/aerosol types in complicated atmospheres (e.g., overlapping cloud and aerosol layers, cloud over snow/ice surface, etc.). To overcome the challenge of the cloud property retrieval in passive sensor, we develop a novel VAE based approach to learn domain invariant representation that capture the spatial pattern from multiple satellite remote sensing data (VDAM), to build a domain invariant cloud property retrieval method to accurately classify different cloud types (labels) in the passive sensing dataset. We further exploit the weight based alignment method on the label space to learn a powerful domain adaptation technique that is pertinent to the remote sensing application. Experiments demonstrate our method outperforms other state-of-the-art machine learning methods and achieves higher accuracy in cloud property retrieval in the passive satellite dataset. 

   more » « less
5. WiCE: Real-World Entailment for Claims in Wikipedia

   https://doi.org/10.18653/v1/2023.emnlp-main.470  

   Kamoi, Ryo; Goyal, Tanya; Rodriguez, Juan; Durrett, Greg (January 2023, Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing)

   Textual entailment models are increasingly applied in settings like fact-checking, presupposition verification in question answering, or summary evaluation. However, these represent a significant domain shift from existing entailment datasets, and models underperform as a result. We propose WiCE, a new fine-grained textual entailment dataset built on natural claim and evidence pairs extracted from Wikipedia. In addition to standard claim-level entailment, WiCE provides entailment judgments over sub-sentence units of the claim, and a minimal subset of evidence sentences that support each subclaim. To support this, we propose an automatic claim decomposition strategy using GPT-3.5 which we show is also effective at improving entailment models’ performance on multiple datasets at test time. Finally, we show that real claims in our dataset involve challenging verification and retrieval problems that existing models fail to address. 

   more » « less