Zobrazeno 1 - 10
of 5 565
pro vyhledávání: '"Badrinath, A"'
Autor:
Krishna, Varun Badrinath
Retrieval-augmented generation (RAG) on specialized domain datasets has shown improved performance when large language models (LLMs) are fine-tuned for generating responses to user queries. In this study, we develop a cybersecurity question-answering
Externí odkaz:
http://arxiv.org/abs/2411.01073
Large Language Models (LLMs) for public use require continuous pre-training to remain up-to-date with the latest data. The models also need to be fine-tuned with specific instructions to maintain their ability to follow instructions accurately. Typic
Externí odkaz:
http://arxiv.org/abs/2410.10739
Do different generative image models secretly learn similar underlying representations? We investigate this by measuring the latent space similarity of four different models: VAEs, GANs, Normalizing Flows (NFs), and Diffusion Models (DMs). Our method
Externí odkaz:
http://arxiv.org/abs/2407.13449
For aligning large language models (LLMs), prior work has leveraged reinforcement learning via human feedback (RLHF) or variations of direct preference optimization (DPO). While DPO offers a simpler framework based on maximum likelihood estimation, i
Externí odkaz:
http://arxiv.org/abs/2405.17956
Autor:
Nie, Allen, Chandak, Yash, Yuan, Christina J., Badrinath, Anirudhan, Flet-Berliac, Yannis, Brunskil, Emma
Offline policy evaluation (OPE) allows us to evaluate and estimate a new sequential decision-making policy's performance by leveraging historical interaction data collected from other policies. Evaluating a new policy online without a confident estim
Externí odkaz:
http://arxiv.org/abs/2405.17708
Autor:
Guo, Lin, Badrinath, Padmanabhan, Mookherjee, Jessica, Ghosh, Anjan, McCallum, Edyta, Dissanayake, Nirosha, George, Abraham
Publikováno v:
International Journal of Prison Health, 2024, Vol. 20, Issue 4, pp. 410-421.
Externí odkaz:
http://www.emeraldinsight.com/doi/10.1108/IJOPH-08-2023-0049
A common way to explore text corpora is through low-dimensional projections of the documents, where one hopes that thematically similar documents will be clustered together in the projected space. However, popular algorithms for dimensionality reduct
Externí odkaz:
http://arxiv.org/abs/2308.01420
Chain-of-thought (CoT) prompting has been shown to empirically improve the accuracy of large language models (LLMs) on various question answering tasks. While understanding why CoT prompting is effective is crucial to ensuring that this phenomenon is
Externí odkaz:
http://arxiv.org/abs/2307.13339
Autor:
Ganesh, Aditya Nalgunda, Badrinath, Dhruval Pobbathi, Kumar, Harshith Mohan, SS, Priya, Narayan, Surabhi
Modern approaches for vision-centric environment perception for autonomous navigation make extensive use of self-supervised monocular depth estimation algorithms that output disparity maps. However, when this disparity map is projected onto 3D space,
Externí odkaz:
http://arxiv.org/abs/2307.10934
Despite the recent advancements in offline reinforcement learning via supervised learning (RvS) and the success of the decision transformer (DT) architecture in various domains, DTs have fallen short in several challenging benchmarks. The root cause
Externí odkaz:
http://arxiv.org/abs/2306.14069