Zobrazeno 1 - 10
of 35
pro vyhledávání: '"Kalyan, Ashwin"'
Autor:
Dogra, Atharvan, Deshpande, Ameet, Nay, John, Rajpurohit, Tanmay, Kalyan, Ashwin, Ravindran, Balaraman
Recent developments in large language models (LLMs), while offering a powerful foundation for developing natural language agents, raise safety concerns about them and the autonomous agents built upon them. Deception is one potential capability of AI
Externí odkaz:
http://arxiv.org/abs/2405.04325
Autor:
Chaudhari, Shreyas, Aggarwal, Pranjal, Murahari, Vishvak, Rajpurohit, Tanmay, Kalyan, Ashwin, Narasimhan, Karthik, Deshpande, Ameet, da Silva, Bruno Castro
State-of-the-art large language models (LLMs) have become indispensable tools for various tasks. However, training LLMs to serve as effective assistants for humans requires careful consideration. A promising approach is reinforcement learning from hu
Externí odkaz:
http://arxiv.org/abs/2404.08555
Autor:
Aggarwal, Pranjal, Murahari, Vishvak, Rajpurohit, Tanmay, Kalyan, Ashwin, Narasimhan, Karthik, Deshpande, Ameet
The advent of large language models (LLMs) has ushered in a new paradigm of search engines that use generative models to gather and summarize information to answer user queries. This emerging technology, which we formalize under the unified framework
Externí odkaz:
http://arxiv.org/abs/2311.09735
Autor:
Gupta, Shashank, Shrivastava, Vaishnavi, Deshpande, Ameet, Kalyan, Ashwin, Clark, Peter, Sabharwal, Ashish, Khot, Tushar
Recent works have showcased the ability of LLMs to embody diverse personas in their responses, exemplified by prompts like 'You are Yoda. Explain the Theory of Relativity.' While this ability allows personalization of LLMs and enables human behavior
Externí odkaz:
http://arxiv.org/abs/2311.04892
Autor:
Murahari, Vishvak, Deshpande, Ameet, Clark, Peter, Rajpurohit, Tanmay, Sabharwal, Ashish, Narasimhan, Karthik, Kalyan, Ashwin
Quantitative evaluation metrics have traditionally been pivotal in gauging the advancements of artificial intelligence systems, including large language models (LLMs). However, these metrics have inherent limitations. Given the intricate nature of re
Externí odkaz:
http://arxiv.org/abs/2311.02807
Despite recent successes in language models, their ability to represent numbers is insufficient. Humans conceptualize numbers based on their magnitudes, effectively projecting them on a number line; whereas subword tokenization fails to explicitly ca
Externí odkaz:
http://arxiv.org/abs/2310.06204
The generation of effective latent representations and their subsequent refinement to incorporate precise information is an essential prerequisite for Vision-Language Understanding (VLU) tasks such as Video Question Answering (VQA). However, most exi
Externí odkaz:
http://arxiv.org/abs/2309.00133
Offline reinforcement learning (RL) methods strike a balance between exploration and exploitation by conservative value estimation -- penalizing values of unseen states and actions. Model-free methods penalize values at all unseen actions, while mode
Externí odkaz:
http://arxiv.org/abs/2308.03882
Anthropomorphization is the tendency to attribute human-like traits to non-human entities. It is prevalent in many social contexts -- children anthropomorphize toys, adults do so with brands, and it is a literary device. It is also a versatile tool i
Externí odkaz:
http://arxiv.org/abs/2305.14784
Autor:
Deshpande, Ameet, Jimenez, Carlos E., Chen, Howard, Murahari, Vishvak, Graf, Victoria, Rajpurohit, Tanmay, Kalyan, Ashwin, Chen, Danqi, Narasimhan, Karthik
Semantic textual similarity (STS), a cornerstone task in NLP, measures the degree of similarity between a pair of sentences, and has broad application in fields such as information retrieval and natural language understanding. However, sentence simil
Externí odkaz:
http://arxiv.org/abs/2305.15093