Discovering the intended items of user queries from a massive repository of items is one of the main goals of an e-commerce search system. Relevance prediction is essential to the search system since it helps improve performance. When online serving a relevance model, the model is required to perform fast and accurate inference. Currently, the widely used models such as Bi-encoder and Cross-encoder have their limitations in accuracy or inference speed respectively. In this work, we propose a novel model called the Entity-Based Relevance Model (EBRM). We identify the entities contained in an item and decompose the QI (query-item) relevance problem into multiple QE (query-entity) relevance problems; we then aggregate their results to form the QI prediction using a soft logic formulation. The decomposition allows us to use a Cross-encoder QE relevance module for high accuracy as well as cache QE predictions for fast online inference. Utilizing soft logic makes the prediction procedure interpretable and intervenable. We also show that pretraining the QE module with auto-generated QE data from user logs can further improve the overall performance. The proposed method is evaluated on labeled data from e-commerce websites. Empirical results show that it achieves promising improvements with computation efficiency.
Vision-language tasks, such as VQA, SNLI-VE, and VCR are challenging because they require the model's reasoning ability to understand the semantics of the visual world and natural language. Supervised methods working for vision-language tasks have been well-studied. However, solving these tasks in a zero-shot setting is less explored. Since Contrastive Language-Image Pre-training (CLIP) has shown remarkable zero-shot performance on image-text matching, previous works utilized its strong zero-shot ability by converting vision-language tasks into an image-text matching problem, and they mainly consider global-level matching (e.g., the whole image or sentence). However, we find visual and textual fine-grained information, e.g., keywords in the sentence and objects in the image, can be fairly informative for semantics understanding. Inspired by this, we propose a unified framework to take advantage of the fine-grained information for zero-shot vision-language learning, covering multiple tasks such as VQA, SNLI-VE, and VCR. Our experiments show that our framework outperforms former zero-shot methods on VQA and achieves substantial improvement on SNLI-VE and VCR. Furthermore, our ablation studies confirm the effectiveness and generalizability of our proposed method.
In this paper, we introduce a large Multi-Attribute and Language Search dataset for text-based person retrieval, called MALS, and explore the feasibility of performing pre-training on both attribute recognition and image-text matching tasks in one stone. In particular, MALS contains 1,510,330 image-text pairs, which is about 37.5 times larger than prevailing CUHK-PEDES, and all images are annotated with 27 attributes. Considering the privacy concerns and annotation costs, we leverage the off-the-shelf diffusion models to generate the dataset. To verify the feasibility of learning from the generated data, we develop a new joint Attribute Prompt Learning and Text Matching Learning (APTM) framework, considering the shared knowledge between attribute and text. As the name implies, APTM contains an attribute prompt learning stream and a text matching learning stream. (1) The attribute prompt learning leverages the attribute prompts for image-attribute alignment, which enhances the text matching learning. (2) The text matching learning facilitates the representation learning on fine-grained details, and in turn, boosts the attribute prompt learning. Extensive experiments validate the effectiveness of the pre-training on MALS, achieving state-of-the-art retrieval performance via APTM on three challenging real-world benchmarks. In particular, APTM achieves a consistent improvement of +6.96%, +7.68%, and +16.95% Recall@1 accuracy on CUHK-PEDES, ICFG-PEDES, and RSTPReid datasets by a clear margin, respectively.
Vision-language models (VLMs) are impactful in part because they can be applied to a variety of visual understanding tasks in a zero-shot fashion, without any fine-tuning. We study $\textit{generative VLMs}$ that are trained for next-word generation given an image. We explore their zero-shot performance on the illustrative task of image-text retrieval across 8 popular vision-language benchmarks. Our first observation is that they can be repurposed for discriminative tasks (such as image-text retrieval) by simply computing the match score of generating a particular text string given an image. We call this probabilistic score the $\textit{Visual Generative Pre-Training Score}$ (VisualGPTScore). While the VisualGPTScore produces near-perfect accuracy on some retrieval benchmarks, it yields poor accuracy on others. We analyze this behavior through a probabilistic lens, pointing out that some benchmarks inadvertently capture unnatural language distributions by creating adversarial but unlikely text captions. In fact, we demonstrate that even a "blind" language model that ignores any image evidence can sometimes outperform all prior art, reminiscent of similar challenges faced by the visual-question answering (VQA) community many years ago. We derive a probabilistic post-processing scheme that controls for the amount of linguistic bias in generative VLMs at test time without having to retrain or fine-tune the model. We show that the VisualGPTScore, when appropriately debiased, is a strong zero-shot baseline for vision-language understanding, oftentimes producing state-of-the-art accuracy.
Image-Text Matching (ITM) task, a fundamental vision-language (VL) task, suffers from the inherent ambiguity arising from multiplicity and imperfect annotations. Deterministic functions are not sufficiently powerful to capture ambiguity, prompting the exploration of probabilistic embeddings to tackle the challenge. However, the existing probabilistic ITM approach encounters two key shortcomings; the burden of heavy computations due to the Monte Carlo approximation, and the loss saturation issue in the face of abundant false negatives. To overcome the issues, this paper presents an improved Probabilistic Cross-Modal Embeddings (named PCME++) by introducing a new probabilistic distance with a closed-form solution. In addition, two optimization techniques are proposed to enhance PCME++ further: first, the incorporation of pseudo-positives to prevent the negative effect under massive false negatives; second, mixed sample data augmentation for probabilistic matching. Experimental results on MS-COCO Caption and two extended benchmarks, CxC and ECCV Caption, demonstrate the effectiveness of PCME++ compared to state-of-the-art ITM methods. The robustness of PCME++ is also evaluated under noisy image-text correspondences. In addition, the potential applicability of PCME++ in automatic prompt-filtering for zero-shot classification is shown. The code is available at https://github.com/naver-ai/pcmepp
Prior study has shown that pretrained language models (PLM) can boost the performance of text-based recommendation. In contrast to previous works that either use PLM to encode user history as a whole input text, or impose an additional aggregation network to fuse multi-turn history representations, we propose a unified local- and global-attention Transformer encoder to better model two-level contexts of user history. Moreover, conditioned on user history encoded by Transformer encoders, our framework leverages Transformer decoders to estimate the language perplexity of candidate text items, which can serve as a straightforward yet significant contrastive signal for user-item text matching. Based on this, our framework, UniTRec, unifies the contrastive objectives of discriminative matching scores and candidate text perplexity to jointly enhance text-based recommendation. Extensive evaluation shows that UniTRec delivers SOTA performance on three text-based recommendation tasks. Code is available at https://github.com/Veason-silverbullet/UniTRec.
Common document ranking pipelines in search systems are cascade systems that involve multiple ranking layers to integrate different information step-by-step. In this paper, we propose a novel re-ranker Fusion-in-T5 (FiT5), which integrates text matching information, ranking features, and global document information into one single unified model via templated-based input and global attention. Experiments on passage ranking benchmarks MS MARCO and TREC DL show that FiT5, as one single model, significantly improves ranking performance over complex cascade pipelines. Analysis finds that through attention fusion, FiT5 jointly utilizes various forms of ranking information via gradually attending to related documents and ranking features, and improves the detection of subtle nuances. Our code is open-sourced at https://github.com/OpenMatch/FiT5.
We present PESCO, a novel contrastive learning framework that substantially improves the performance of zero-shot text classification. We formulate text classification as a neural text matching problem where each document is treated as a query, and the system learns the mapping from each query to the relevant class labels by (1) adding prompts to enhance label matching, and (2) using retrieved labels to enrich the training set in a self-training loop of contrastive learning. PESCO achieves state-of-the-art performance on four benchmark text classification datasets. On DBpedia, we achieve 98.5\% accuracy without any labeled data, which is close to the fully-supervised result. Extensive experiments and analyses show all the components of PESCO are necessary for improving the performance of zero-shot text classification.
Image-to-recipe retrieval is a challenging vision-to-language task of significant practical value. The main challenge of the task lies in the ultra-high redundancy in the long recipe and the large variation reflected in both food item combination and food item appearance. A de-facto idea to address this task is to learn a shared feature embedding space in which a food image is aligned better to its paired recipe than other recipes. However, such supervised global matching is prone to supervision collapse, i.e., only partial information that is necessary for distinguishing training pairs can be identified, while other information that is potentially useful in generalization could be lost. To mitigate such a problem, we propose a mask-augmentation-based local matching network (MALM), where an image-text matching module and a masked self-distillation module benefit each other mutually to learn generalizable cross-modality representations. On one hand, we perform local matching between the tokenized representations of image and text to locate fine-grained cross-modality correspondence explicitly. We involve representations of masked image patches in this process to alleviate overfitting resulting from local matching especially when some food items are underrepresented. On the other hand, predicting the hidden representations of the masked patches through self-distillation helps to learn general-purpose image representations that are expected to generalize better. And the multi-task nature of the model enables the representations of masked patches to be text-aware and thus facilitates the lost information reconstruction. Experimental results on Recipe1M dataset show our method can clearly outperform state-of-the-art (SOTA) methods. Our code will be available at https://github.com/MyFoodChoice/MALM_Mask_Augmentation_based_Local_Matching-_for-_Food_Recipe_Retrieval
Diffusion models, such as Stable Diffusion, have shown incredible performance on text-to-image generation. Since text-to-image generation often requires models to generate visual concepts with fine-grained details and attributes specified in text prompts, can we leverage the powerful representations learned by pre-trained diffusion models for discriminative tasks such as image-text matching? To answer this question, we propose a novel approach, Discriminative Stable Diffusion (DSD), which turns pre-trained text-to-image diffusion models into few-shot discriminative learners. Our approach mainly uses the cross-attention score of a Stable Diffusion model to capture the mutual influence between visual and textual information and fine-tune the model via efficient attention-based prompt learning to perform image-text matching. By comparing DSD with state-of-the-art methods on several benchmark datasets, we demonstrate the potential of using pre-trained diffusion models for discriminative tasks with superior results on few-shot image-text matching.
In most Vision-Language models (VL), the understanding of the image structure is enabled by injecting the position information (PI) about objects in the image. In our case study of LXMERT, a state-of-the-art VL model, we probe the use of the PI in the representation and study its effect on Visual Question Answering. We show that the model is not capable of leveraging the PI for the image-text matching task on a challenge set where only position differs. Yet, our experiments with probing confirm that the PI is indeed present in the representation. We introduce two strategies to tackle this: (i) Positional Information Pre-training and (ii) Contrastive Learning on PI using Cross-Modality Matching. Doing so, the model can correctly classify if images with detailed PI statements match. Additionally to the 2D information from bounding boxes, we introduce the object's depth as new feature for a better object localization in the space. Even though we were able to improve the model properties as defined by our probes, it only has a negligible effect on the downstream performance. Our results thus highlight an important issue of multimodal modeling: the mere presence of information detectable by a probing classifier is not a guarantee that the information is available in a cross-modal setup.
Due to the lack of paired samples and the low signal-to-noise ratio of functional MRI (fMRI) signals, reconstructing perceived natural images or decoding their semantic contents from fMRI data are challenging tasks. In this work, we propose, for the first time, a task-agnostic fMRI-based brain decoding model, BrainCLIP, which leverages CLIP's cross-modal generalization ability to bridge the modality gap between brain activity, image, and text. Our experiments demonstrate that CLIP can act as a pivot for generic brain decoding tasks, including zero-shot visual categories decoding, fMRI-image/text matching, and fMRI-to-image generation. Specifically, BrainCLIP aims to train a mapping network that transforms fMRI patterns into a well-aligned CLIP embedding space by combining visual and textual supervision. Our experiments show that this combination can boost the decoding model's performance on certain tasks like fMRI-text matching and fMRI-to-image generation. On the zero-shot visual category decoding task, BrainCLIP achieves significantly better performance than BraVL, a recently proposed multi-modal method specifically designed for this task. BrainCLIP can also reconstruct visual stimuli with high semantic fidelity and establishes a new state-of-the-art for fMRI-based natural image reconstruction in terms of high-level semantic features.
Large-scale vision-language pre-training has achieved significant performance in multi-modal understanding and generation tasks. However, existing methods often perform poorly on image-text matching tasks that require structured representations, i.e., representations of objects, attributes, and relations. As illustrated in Fig.~reffig:case (a), the models cannot make a distinction between ``An astronaut rides a horse" and ``A horse rides an astronaut". This is because they fail to fully leverage structured knowledge when learning representations in multi-modal scenarios. In this paper, we present an end-to-end framework Structure-CLIP, which integrates Scene Graph Knowledge (SGK) to enhance multi-modal structured representations. Firstly, we use scene graphs to guide the construction of semantic negative examples, which results in an increased emphasis on learning structured representations. Moreover, a Knowledge-Enhance Encoder (KEE) is proposed to leverage SGK as input to further enhance structured representations. To verify the effectiveness of the proposed framework, we pre-train our model with the aforementioned approaches and conduct experiments on downstream tasks. Experimental results demonstrate that Structure-CLIP achieves state-of-the-art (SOTA) performance on VG-Attribution and VG-Relation datasets, with 12.5% and 4.1% ahead of the multi-modal SOTA model respectively. Meanwhile, the results on MSCOCO indicate that Structure-CLIP significantly enhances the structured representations while maintaining the ability of general representations. Our code is available at https://github.com/zjukg/Structure-CLIP.
Employing a dictionary can efficiently rectify the deviation between the visual prediction and the ground truth in scene text recognition methods. However, the independence of the dictionary on the visual features may lead to incorrect rectification of accurate visual predictions. In this paper, we propose a new dictionary language model leveraging the Scene Image-Text Matching(SITM) network, which avoids the drawbacks of the explicit dictionary language model: 1) the independence of the visual features; 2) noisy choice in candidates etc. The SITM network accomplishes this by using Image-Text Contrastive (ITC) Learning to match an image with its corresponding text among candidates in the inference stage. ITC is widely used in vision-language learning to pull the positive image-text pair closer in feature space. Inspired by ITC, the SITM network combines the visual features and the text features of all candidates to identify the candidate with the minimum distance in the feature space. Our lexicon method achieves better results(93.8\% accuracy) than the ordinary method results(92.1\% accuracy) on six mainstream benchmarks. Additionally, we integrate our method with ABINet and establish new state-of-the-art results on several benchmarks.
Visual Word Sense Disambiguation (VWSD) is a task to find the image that most accurately depicts the correct sense of the target word for the given context. Previously, image-text matching models often suffered from recognizing polysemous words. This paper introduces an unsupervised VWSD approach that uses gloss information of an external lexical knowledge-base, especially the sense definitions. Specifically, we suggest employing Bayesian inference to incorporate the sense definitions when sense information of the answer is not provided. In addition, to ameliorate the out-of-dictionary (OOD) issue, we propose a context-aware definition generation with GPT-3. Experimental results show that the VWSD performance significantly increased with our Bayesian inference-based approach. In addition, our context-aware definition generation achieved prominent performance improvement in OOD examples exhibiting better performance than the existing definition generation method.
With the extensive use of vision-language models in various downstream tasks, evaluating their robustness is crucial. In this paper, we propose a benchmark for assessing the robustness of vision-language models. We believe that a robust model should properly understand both linguistic and visual semantics and be resilient to explicit variations. In pursuit of this goal, we create new variants of texts and images in the MS-COCO test set and re-evaluate the state-of-the-art (SOTA) models with the new data. Specifically, we alter the meaning of text by replacing a word, and generate visually altered images that maintain some visual context while introducing noticeable pixel changes through image mixing techniques.Our evaluations on the proposed benchmark reveal substantial performance degradation in many SOTA models (e.g., Image-to-Text Recall@1: 81.9\% $\rightarrow$ 48.4\% in BLIP, 66.1\% $\rightarrow$ 37.6\% in VSE$\infty$), with the models often favoring the altered texts/images over the original ones. This indicates the current vision-language models struggle with subtle changes and often fail to understand the overall context of texts and images. Based on these findings, we propose semantic contrastive loss and visual contrastive loss to learn more robust embedding. Datasets and code are available at {\url{https://github.com/pseulki/rococo}}.
Embedding based retrieval has seen its usage in a variety of search applications like e-commerce, social networking search etc. While the approach has demonstrated its efficacy in tasks like semantic matching and contextual search, it is plagued by the problem of uncontrollable relevance. In this paper, we conduct an analysis of embedding-based retrieval launched in early 2021 on our social network search engine, and define two main categories of failures introduced by it, integrity and junkiness. The former refers to issues such as hate speech and offensive content that can severely harm user experience, while the latter includes irrelevant results like fuzzy text matching or language mismatches. Efficient methods during model inference are further proposed to resolve the issue, including indexing treatments and targeted user cohort treatments, etc. Though being simple, we show the methods have good offline NDCG and online A/B tests metrics gain in practice. We analyze the reasons for the improvements, pointing out that our methods are only preliminary attempts to this important but challenging problem. We put forward potential future directions to explore.
Understanding verbs is crucial to modelling how people and objects interact with each other and the environment through space and time. Recently, state-of-the-art video-language models based on CLIP have been shown to have limited verb understanding and to rely extensively on nouns, restricting their performance in real-world video applications that require action and temporal understanding. In this work, we improve verb understanding for CLIP-based video-language models by proposing a new Verb-Focused Contrastive (VFC) framework. This consists of two main components: (1) leveraging pretrained large language models (LLMs) to create hard negatives for cross-modal contrastive learning, together with a calibration strategy to balance the occurrence of concepts in positive and negative pairs; and (2) enforcing a fine-grained, verb phrase alignment loss. Our method achieves state-of-the-art results for zero-shot performance on three downstream tasks that focus on verb understanding: video-text matching, video question-answering and video classification. To the best of our knowledge, this is the first work which proposes a method to alleviate the verb understanding problem, and does not simply highlight it.
In recent years, short Text Matching tasks have been widely applied in the fields ofadvertising search and recommendation. The difficulty lies in the lack of semantic information and word ambiguity caused by the short length of the text. Previous works have introduced complement sentences or knowledge bases to provide additional feature information. However, these methods have not fully interacted between the original sentence and the complement sentence, and have not considered the noise issue that may arise from the introduction of external knowledge bases. Therefore, this paper proposes a short Text Matching model that combines contrastive learning and external knowledge. The model uses a generative model to generate corresponding complement sentences and uses the contrastive learning method to guide the model to obtain more semantically meaningful encoding of the original sentence. In addition, to avoid noise, we use keywords as the main semantics of the original sentence to retrieve corresponding knowledge words in the knowledge base, and construct a knowledge graph. The graph encoding model is used to integrate the knowledge base information into the model. Our designed model achieves state-of-the-art performance on two publicly available Chinese Text Matching datasets, demonstrating the effectiveness of our model.
We propose a visual-linguistic representation learning approach within a self-supervised learning framework by introducing a new operation, loss, and data augmentation strategy. First, we generate diverse features for the image-text matching (ITM) task via soft-masking the regions in an image, which are most relevant to a certain word in the corresponding caption, instead of completely removing them. Since our framework relies only on image-caption pairs with no fine-grained annotations, we identify the relevant regions to each word by computing the word-conditional visual attention using multi-modal encoder. Second, we encourage the model to focus more on hard but diverse examples by proposing a focal loss for the image-text contrastive learning (ITC) objective, which alleviates the inherent limitations of overfitting and bias issues. Last, we perform multi-modal data augmentations for self-supervised learning via mining various examples by masking texts and rendering distortions on images. We show that the combination of these three innovations is effective for learning a pretrained model, leading to outstanding performance on multiple vision-language downstream tasks.