Zobrazeno 1 - 9
of 9
pro vyhledávání: '"Sinha, Sanchit"'
The rising need for explainable deep neural network architectures has utilized semantic concepts as explainable units. Several approaches utilizing disentangled representation learning estimate the generative factors and utilize them as concepts for
Externí odkaz:
http://arxiv.org/abs/2410.15491
Generalized additive models (GAMs) have long been a powerful white-box tool for the intelligible analysis of tabular data, revealing the influence of each feature on the model predictions. Despite the success of neural networks (NNs) in various domai
Externí odkaz:
http://arxiv.org/abs/2410.04723
Interpretability of Deep Neural Networks using concept-based models offers a promising way to explain model behavior through human-understandable concepts. A parallel line of research focuses on disentangling the data distribution into its underlying
Externí odkaz:
http://arxiv.org/abs/2407.19300
Adapting large language models (LLMs) to unseen tasks with in-context training samples without fine-tuning remains an important research problem. To learn a robust LLM that adapts well to unseen tasks, multiple meta-training approaches have been prop
Externí odkaz:
http://arxiv.org/abs/2405.11446
With the wide proliferation of Deep Neural Networks in high-stake applications, there is a growing demand for explainability behind their decision-making process. Concept learning models attempt to learn high-level 'concepts' - abstract entities that
Externí odkaz:
http://arxiv.org/abs/2405.00349
Autor:
Bhatia, Anshu, Sinha, Sanchit, Dingliwal, Saket, Gopalakrishnan, Karthik, Bodapati, Sravan, Kirchhoff, Katrin
Speech representations learned in a self-supervised fashion from massive unlabeled speech corpora have been adapted successfully toward several downstream tasks. However, such representations may be skewed toward canonical data characteristics of suc
Externí odkaz:
http://arxiv.org/abs/2307.00453
Deep neural networks are susceptible to human imperceptible adversarial perturbations. One of the strongest defense mechanisms is \emph{Adversarial Training} (AT). In this paper, we aim to address two predominant problems in AT. First, there is still
Externí odkaz:
http://arxiv.org/abs/2306.02618
Rising usage of deep neural networks to perform decision making in critical applications like medical diagnosis and financial analysis have raised concerns regarding their reliability and trustworthiness. As automated systems become more mainstream,
Externí odkaz:
http://arxiv.org/abs/2211.16080
Interpretability methods like Integrated Gradient and LIME are popular choices for explaining natural language model predictions with relative word importance scores. These interpretations need to be robust for trustworthy NLP applications in high-st
Externí odkaz:
http://arxiv.org/abs/2108.04990