Customer-obsessed science
-
January 19, 2024Attention-based representation of multi-image inputs improves performance on downstream vision-language tasks.
-
January 17, 2024Representing facts using knowledge triplets rather than natural language enables finer-grained judgments.
-
December 26, 2023Theoretical analysis and experiments show that clipped stochastic gradient descent (SGD) enables robust online statistical estimation.
-
December 20, 2023Novel architectures and carefully prepared training data enable state-of-the-art performance.
January 17, 2024
Proposal topics include enhancements for dialogue, factual questioning and answering, text generation, document summarization, image generation, and more. The submission period is open now and closes on March 6.
-
ICASSP 20242024Modern Automatic Speech Recognition (ASR) systems are evaluated with respect to Word Error Rate (WER). While WER is a useful metric for training and evaluation of speech models, it does not fully reflect the difference in semantics between predicted and ground truth transcriptions. In conversational voice assistants, the ability to sufficiently understand the semantic meaning of the user request is often
-
ASPLOS 20242024Designing performant and noise-robust circuits for Quantum Machine Learning (QML) is challenging — the design space scales exponentially with circuit size, and there are few well-supported guiding principles for QML circuit design. Although recent Quantum Circuit Search (QCS) methods attempt to search for performant QML circuits that are also robust to hardware noise, they directly adopt designs from classical
-
ICSE 20242024Identifying correct and complete taint specifications is critical for detecting vulnerabilities in the ever-changing landscape of software security, and an automated scalable and practical solution remains elusive in the field. In this paper, we report our semi-automated scheme for inferring and maintaining taint specifications at industrial scale. Knowledge graph is adopted as the core engine to represent
-
January 25, 2024Amazon IIT–Bombay AI-ML Initiative seeks to advance artificial intelligence and machine learning research within speech, language, and multimodal-AI domains.
-
January 04, 2024Program empowers students from diverse backgrounds to become industry leaders through scholarship, research, and career opportunities.
-
January 03, 2024Researchers honored for their contributions to the scientific community in 2023.