Zobrazeno 1 - 10
of 266
pro vyhledávání: '"Dumpala P"'
Autor:
Dumpala, Sri Harsha, Jaiswal, Aman, Sastry, Chandramouli, Milios, Evangelos, Oore, Sageev, Sajjad, Hassan
Despite the significant influx of prompt-tuning techniques for generative vision-language models (VLMs), it remains unclear how sensitive these models are to lexical and semantic alterations in prompts. In this paper, we evaluate the ability of gener
Externí odkaz:
http://arxiv.org/abs/2410.13030
Autor:
Sharma, Dushyant, Fosburgh, James, Dumpala, Sri Harsha, Sastri, Chandramouli Shama, Kruchinin, Stanislav Yu., Naylor, Patrick A.
We explore the recently proposed explainable acoustic neural embedding~(XANE) system that models the background acoustics of a speech signal in a non-intrusive manner. The XANE embeddings are used to estimate specific parameters related to the backgr
Externí odkaz:
http://arxiv.org/abs/2407.06342
Autor:
Dumpala, Sri Harsha, Dikaios, Katerina, Nunes, Abraham, Rudzicz, Frank, Uher, Rudolf, Oore, Sageev
Depression, a prevalent mental health disorder impacting millions globally, demands reliable assessment systems. Unlike previous studies that focus solely on either detecting depression or predicting its severity, our work identifies individual sympt
Externí odkaz:
http://arxiv.org/abs/2406.17229
Autor:
Rodriguez, Sebastian, Dumpala, Sri Harsha, Dikaios, Katerina, Rempel, Sheri, Uher, Rudolf, Oore, Sageev
Current automatic depression detection systems provide predictions directly without relying on the individual symptoms/items of depression as denoted in the clinical depression rating scales. In contrast, clinicians assess each item in the depression
Externí odkaz:
http://arxiv.org/abs/2406.16000
Autor:
Dumpala, Sri Harsha, Jaiswal, Aman, Sastry, Chandramouli, Milios, Evangelos, Oore, Sageev, Sajjad, Hassan
Despite their remarkable successes, state-of-the-art large language models (LLMs), including vision-and-language models (VLMs) and unimodal language models (ULMs), fail to understand precise semantics. For example, semantically equivalent sentences e
Externí odkaz:
http://arxiv.org/abs/2406.11171
Autor:
Dumpala, Sri Harsha, Sharma, Dushyant, Sastri, Chandramouli Shama, Kruchinin, Stanislav, Fosburgh, James, Naylor, Patrick A.
We present a novel method for extracting neural embeddings that model the background acoustics of a speech signal. The extracted embeddings are used to estimate specific parameters related to the background acoustic properties of the signal in a non-
Externí odkaz:
http://arxiv.org/abs/2406.05199
Autor:
Dumpala, Sri Harsha, Jaiswal, Aman, Sastry, Chandramouli, Milios, Evangelos, Oore, Sageev, Sajjad, Hassan
Despite their remarkable successes, state-of-the-art language models face challenges in grasping certain important semantic details. This paper introduces the VISLA (Variance and Invariance to Semantic and Lexical Alterations) benchmark, designed to
Externí odkaz:
http://arxiv.org/abs/2404.16365
Previous works on depression detection use datasets collected in similar environments to train and test the models. In practice, however, the train and test distributions cannot be guaranteed to be identical. Distribution shifts can be introduced due
Externí odkaz:
http://arxiv.org/abs/2404.05071
In this paper, we study the application of Test-Time Training (TTT) as a solution to handling distribution shifts in speech applications. In particular, we introduce distribution-shifts to the test datasets of standard speech-classification tasks --
Externí odkaz:
http://arxiv.org/abs/2309.10930
We introduce DiffAug, a simple and efficient diffusion-based augmentation technique to train image classifiers for the crucial yet challenging goal of improved classifier robustness. Applying DiffAug to a given example consists of one forward-diffusi
Externí odkaz:
http://arxiv.org/abs/2306.09192