Zobrazeno 1 - 10
of 3 192
pro vyhledávání: '"P Natesan"'
Autor:
Wei, Dennis, Padhi, Inkit, Ghosh, Soumya, Dhurandhar, Amit, Ramamurthy, Karthikeyan Natesan, Chang, Maria
Training data attribution (TDA) is the task of attributing model behavior to elements in the training data. This paper draws attention to the common setting where one has access only to the final trained model, and not the training algorithm or inter
Externí odkaz:
http://arxiv.org/abs/2412.03906
Autor:
Miehling, Erik, Desmond, Michael, Ramamurthy, Karthikeyan Natesan, Daly, Elizabeth M., Dognin, Pierre, Rios, Jesus, Bouneffouf, Djallel, Liu, Miao
Building pluralistic AI requires designing models that are able to be shaped to represent a wide range of value systems and cultures. Achieving this requires first being able to evaluate the degree to which a given model is capable of reflecting vari
Externí odkaz:
http://arxiv.org/abs/2411.12405
Mechanistic interpretability aims to provide human-understandable insights into the inner workings of neural network models by examining their internals. Existing approaches typically require significant manual effort and prior knowledge, with strate
Externí odkaz:
http://arxiv.org/abs/2410.16484
We consider the Weak Galerkin finite element approximation of the Singularly Perturbed Biharmonic elliptic problem on a unit square domain with clamped boundary conditions. Shishkin mesh is used for domain discretization as the solution exhibits boun
Externí odkaz:
http://arxiv.org/abs/2409.07217
Autor:
Lee, Bruce W., Padhi, Inkit, Ramamurthy, Karthikeyan Natesan, Miehling, Erik, Dognin, Pierre, Nagireddy, Manish, Dhurandhar, Amit
LLMs have shown remarkable capabilities, but precisely controlling their response behavior remains challenging. Existing activation steering methods alter LLM behavior indiscriminately, limiting their practical applicability in settings where selecti
Externí odkaz:
http://arxiv.org/abs/2409.05907
Autor:
Padhi, Inkit, Ramamurthy, Karthikeyan Natesan, Sattigeri, Prasanna, Nagireddy, Manish, Dognin, Pierre, Varshney, Kush R.
Aligning large language models (LLMs) to value systems has emerged as a significant area of research within the fields of AI and NLP. Currently, this alignment process relies on the availability of high-quality supervised and preference data, which c
Externí odkaz:
http://arxiv.org/abs/2408.10392
The ability to summarize and organize knowledge into abstract concepts is key to learning and reasoning. Many industrial applications rely on the consistent and systematic use of concepts, especially when dealing with decision-critical knowledge. How
Externí odkaz:
http://arxiv.org/abs/2405.20163
Autor:
Natesan, Sanjay, Beigi, Homayoon
Publikováno v:
Recognition Technologies, Inc. Technical Report (2024), RTI-20240524-01
Large scale machine learning-based Raga identification continues to be a nontrivial issue in the computational aspects behind Carnatic music. Each raga consists of many unique and intrinsic melodic patterns that can be used to easily identify them fr
Externí odkaz:
http://arxiv.org/abs/2405.16000
Autor:
Paes, Lucas Monteiro, Wei, Dennis, Do, Hyo Jin, Strobelt, Hendrik, Luss, Ronny, Dhurandhar, Amit, Nagireddy, Manish, Ramamurthy, Karthikeyan Natesan, Sattigeri, Prasanna, Geyer, Werner, Ghosh, Soumya
Perturbation-based explanation methods such as LIME and SHAP are commonly applied to text classification. This work focuses on their extension to generative language models. To address the challenges of text as output and long text inputs, we propose
Externí odkaz:
http://arxiv.org/abs/2403.14459
Autor:
Achintalwar, Swapnaja, Baldini, Ioana, Bouneffouf, Djallel, Byamugisha, Joan, Chang, Maria, Dognin, Pierre, Farchi, Eitan, Makondo, Ndivhuwo, Mojsilovic, Aleksandra, Nagireddy, Manish, Ramamurthy, Karthikeyan Natesan, Padhi, Inkit, Raz, Orna, Rios, Jesus, Sattigeri, Prasanna, Singh, Moninder, Thwala, Siphiwe, Uceda-Sosa, Rosario A., Varshney, Kush R.
The alignment of large language models is usually done by model providers to add or control behaviors that are common or universally understood across use cases and contexts. In contrast, in this article, we present an approach and architecture that
Externí odkaz:
http://arxiv.org/abs/2403.09704