Zobrazeno 1 - 10
of 3 874
pro vyhledávání: '"Chidambaram, P."'
Autor:
Lau, Allison, Choi, Younwoo, Balazadeh, Vahid, Chidambaram, Keertana, Syrgkanis, Vasilis, Krishnan, Rahul G.
Reinforcement Learning from Human Feedback (RLHF) is widely used to align Language Models (LMs) with human preferences. However, existing approaches often neglect individual user preferences, leading to suboptimal personalization. We present the Pref
Externí odkaz:
http://arxiv.org/abs/2410.14001
Autor:
Fasciati, Simone D., Shteynas, Boris, Campanaro, Giulio, Bakr, Mustafa, Cao, Shuxiang, Chidambaram, Vivek, Wills, James, Leek, Peter J.
We realize a single-Josephson-junction transmon qubit shunted by a simple geometric inductor. We couple it capacitively to a conventional transmon and show that the ZZ interaction between the two qubits is completely suppressed when they are flux-bia
Externí odkaz:
http://arxiv.org/abs/2410.10416
The use of guidance in diffusion models was originally motivated by the premise that the guidance-modified score is that of the data distribution tilted by a conditional likelihood raised to some power. In this work we clarify this misconception by r
Externí odkaz:
http://arxiv.org/abs/2409.13074
This work introduces a new approach to building crash-safe file systems for persistent memory. We exploit the fact that Rust's typestate pattern allows compile-time enforcement of a specific order of operations. We introduce a novel crash-consistency
Externí odkaz:
http://arxiv.org/abs/2406.09649
Autor:
Chidambaram, Muthu, Ge, Rong
A machine learning model is calibrated if its predicted probability for an outcome matches the observed frequency for that outcome conditional on the model prediction. This property has become increasingly important as the impact of machine learning
Externí odkaz:
http://arxiv.org/abs/2406.04068
RLHF has emerged as a pivotal step in aligning language models with human objectives and values. It typically involves learning a reward model from human preference data and then using reinforcement learning to update the generative model accordingly
Externí odkaz:
http://arxiv.org/abs/2405.15065
Autor:
Rathinasamy, Kamalkumar, Nettar, Jayarama, Kumar, Amit, Manchanda, Vishal, Vijayakumar, Arun, Kataria, Ayush, Manjunath, Venkateshprasanna, GS, Chidambaram, Sodhi, Jaskirat Singh, Shaikh, Shoeb, Khan, Wasim Akhtar, Singh, Prashant, Ige, Tanishq Dattatray, Tiwari, Vipin, Mondal, Rajab Ali, K, Harshini, Reka, S, Amancharla, Chetana, Rahman, Faiz ur, A, Harikrishnan P, Saha, Indraneel, Tiwary, Bhavya, Patel, Navin Shankar, S, Pradeep T, J, Balaji A, Priyapravas, Tarafdar, Mohammed Rafee
Enterprises grapple with the significant challenge of managing proprietary unstructured data, hindering efficient information retrieval. This has led to the emergence of AI-driven information retrieval solutions, designed to adeptly extract relevant
Externí odkaz:
http://arxiv.org/abs/2406.00010
Autor:
Balazadeh, Vahid, Chidambaram, Keertana, Nguyen, Viet, Krishnan, Rahul G., Syrgkanis, Vasilis
We study the problem of online sequential decision-making given auxiliary demonstrations from experts who made their decisions based on unobserved contextual information. These demonstrations can be viewed as solving related but slightly different ta
Externí odkaz:
http://arxiv.org/abs/2404.07266
We upgrade the results of Borot--Bouchard--Chidambaram--Creutzig to show that the Gaiotto vector in $\mathcal{N} = 2$ pure supersymmetric gauge theory admits an analytic continuation with respect to the energy scale (which can therefore be taken to b
Externí odkaz:
http://arxiv.org/abs/2403.16938
Informally, a model is calibrated if its predictions are correct with a probability that matches the confidence of the prediction. By far the most common method in the literature for measuring calibration is the expected calibration error (ECE). Rece
Externí odkaz:
http://arxiv.org/abs/2402.10046