Search results

Realistic image generation is an increasingly desired, but deceptively complicated computer vision task, especially when a specific object is required. Whether generating product advertisements or building novel datasets, object composition for realistic image generation depends on realistic object placements as well as believable object harmonization. To address this task, we introduce HopNet, the first

Developing Text Normalization (TN) systems for Text-to-Speech (TTS) on new languages is hard. We propose a novel architecture to facilitate it for multiple languages while using data less than 3% of the size of the data used by the state of the art results on English. We treat TN as a sequence classification problem and propose a granular tokenization mechanism that enables the system to learn majority

A popular approach to building agents using Language Models (LMs) involves iteratively prompting the LM, reflecting on its outputs, and updating the input prompts until the desired task is achieved. However, our analysis reveals two key shortcomings in the existing methods: (i) limited exploration of the decision space due to repetitive reflections, which result in redundant inputs, and (ii) an inability

A key challenge in reliable LLM deployment is recognizing when uncertainty reflects irreducible variability in the task rather than limitations in the model's knowledge. In language tasks, a central source of such aleatoric uncertainty is input ambiguity or underspecification, where multiple interpretations remain plausible. Existing decomposition methods estimate aleatoric uncertainty by generating multiple

Modern Automatic Speech Recognition (ASR) technology is typically fine-tuned for a targeted domain or application to obtain the best recognition results. This requires training and maintaining a dedicated ASR model for each domain, which increases the overall cost. Moreover, fine-tuned model might not be the most optimal way of sharing knowledge across domains. To address this, we propose a novel unified

Language model based methods are powerful techniques for text classification. However, the models have several shortcomings. (1) It is difficult to integrate human knowledge such as keywords. (2) It needs a lot of resources to train the models. (3) It relied on large text data to pretrain. In this paper, we propose Semi-Supervised vMF Neural Topic Modeling (S2vNTM) to overcome these difficulties. S2vNTM

Query Parsing aims to extract product attributes, such as color, brand, and product type, from search queries. These attributes play a crucial role in search engines for tasks such as matching, ranking, and recommendation. There are two types of attributes: explicit attributes that are mentioned explicitly in the search query, and implicit attributes that are mentioned implicitly. Existing works on query

E-commerce stores increasingly use Large Language Models (LLMs) to enhance catalog data quality through automated regeneration. A critical challenge is accurately predicting missing structured attribute values across multilingual product catalogs, where LLM performance varies significantly by language. While existing approaches leverage general knowledge through prompt engineering and external retrieval

We propose an approach for continuous prediction of turn-taking and backchanneling locations in spoken dialogue by fusing a neural acoustic model with a large language model (LLM). Experiments on the Switchboard human-human conversation dataset demonstrate that our approach consistently outperforms the baseline models with single modality. We also develop a novel multi-task instruction fine- tuning strategy

Information explosion has brought us a wide range of data formats and machine learning keeps in constant evolution to develop mechanisms to extract knowledge from them. Modern models in the Deep Learning space have proven to be very successful in multiple applications, yet in the tabular space they fail to provide consistent competitive performance. However, in this work we claim model selection can become

We revisit the problem of differentially private squared error linear regression. We observe that existing state- of-the-art methods are sensitive to the choice of hyperparameters — including the “clipping threshold” that cannot be set optimally in a data-independent way. We give a new algorithm for private linear regression based on gradient boosting. We show that our method consistently improves over

In complex transfer learning scenarios new tasks might not be tightly linked to previous tasks. Approaches that transfer information contained only in the final parameters of a source model will therefore struggle. Instead, transfer learning at a higher level of abstraction is needed. We propose Leap, a framework that achieves this by transferring knowledge across learning processes. We associate each task

Techniques for knowledge graph (KGs) enrichment have been increasingly crucial for commercial applications that rely on evolving product catalogues. However, because of the huge search space of potential enrichment, predictions from KG completion (KGC) methods suffer from low precision, making them unreliable for real-world catalogues. Moreover, candidate facts for enrichment have varied relevance to users

AI assistants are gradually becoming embedded in our lives, utilized for everyday tasks like shopping or music. In addition to the everyday utilization of AI assistants, many users engage them with playful shopping requests, gauging their ability to understand – or simply seeking amusement. However, these requests are often not being responded to in the same playful manner, causing dissatisfaction and even

In this work, we introduce Context-Aware MultiModal Learner (CaMML), for tuning large multimodal models (LMMs). CaMML, a lightweight module, is crafted to seamlessly integrate multimodal contextual samples into large models, thereby empowering the model to derive knowledge from analogous, domain-specific, up-to-date information and make grounded inferences. Importantly, CaMML is highly scalable and can

We introduce Neural Contextual Anomaly Detection (NCAD), a framework for anomaly detection on time series that scales seamlessly from the unsupervised to supervised setting, and is applicable to both univariate and multivariate time series. This is achieved by combining recent developments in representation learning for multivariate time series, with techniques for deep anomaly detection originally developed

Large Language Models (LLMs) tend to be unreliable in the factuality of their answers. To address this problem, NLP researchers have proposed a range of techniques to estimate LLM’s confidence over facts. However, due to the lack of a systematic comparison, it is not clear how the different methods compare to one another. To fill this gap, we present a survey and empirical comparison of estimators of factual

In clinical visits, clinical note writing is a timeconsuming and cost-prohibitive manual task for clinicians. Although virtual medical scribes have been proposed to generate clinical notes (semi-)automatically, the data sparsity issue is still a challenging problem in practice. Identifying the topic of clinical utterances in doctorpatient conversations is one of the key strategies for automation. In this

The rapid rise of generative AI (GenAI) has sparked the sustainability community to explore its potential applications, such as climate impact modeling and renewable energy optimization. However, deploying these GenAIpowered solutions in enterprise environments raises risk concerns. In particular, chatbots and similar GenAI applications face risks of misinformation and disinformation stemming from knowledge

Current abstractive summarization models often generate inconsistent content, i.e. texts that are not directly inferable from the source document, are not consistent with respect to world knowledge, or are self-contradictory. These inconsistencies motivate a new consistency taxonomy that we define as faithfulness, factuality, and self-supportiveness. However, most recent work on reducing inconsistency in