Zobrazeno 1 - 10
of 105
pro vyhledávání: '"BANSAL, CHETAN"'
We introduce Simultaneous Weighted Preference Optimization (SWEPO), a novel extension of Direct Preference Optimization (DPO) designed to accommodate multiple dynamically chosen positive and negative responses for each query. SWEPO employs a weighted
Externí odkaz:
http://arxiv.org/abs/2412.04628
Autor:
Khan, Redwan Ibne Seraj, Jain, Kunal, Shen, Haiying, Mallick, Ankur, Parayil, Anjaly, Kulkarni, Anoop, Kofsky, Steve, Choudhary, Pankhuri, Amant, Renèe St., Wang, Rujia, Cheng, Yue, Butt, Ali R., Rühle, Victor, Bansal, Chetan, Rajmohan, Saravan
In a multi-tenant large language model (LLM) serving platform hosting diverse applications, some users may submit an excessive number of requests, causing the service to become unavailable to other users and creating unfairness. Existing fairness app
Externí odkaz:
http://arxiv.org/abs/2411.15997
Autor:
Soni, Aditya, Das, Mayukh, Parayil, Anjaly, Ghosh, Supriyo, Shandilya, Shivam, Cheng, Ching-An, Gopal, Vishak, Khairy, Sami, Mittag, Gabriel, Hosseinkashi, Yasaman, Bansal, Chetan
The difficulty of exploring and training online on real production systems limits the scope of real-time online data/feedback-driven decision making. The most feasible approach is to adopt offline reinforcement learning from limited trajectory sample
Externí odkaz:
http://arxiv.org/abs/2411.06815
Autor:
Gupta, Taneesh, Shandilya, Shivam, Zhang, Xuchao, Ghosh, Supriyo, Bansal, Chetan, Yao, Huaxiu, Rajmohan, Saravan
The use of large language models (LLMs) as evaluators has garnered significant attention due to their potential to rival human-level evaluations in long-form response assessments. However, current LLM evaluators rely heavily on static, human-defined
Externí odkaz:
http://arxiv.org/abs/2410.21545
Autor:
Wang, Zhaoyang, He, Weilei, Liang, Zhiyuan, Zhang, Xuchao, Bansal, Chetan, Wei, Ying, Zhang, Weitong, Yao, Huaxiu
Recent self-rewarding large language models (LLM) have successfully applied LLM-as-a-Judge to iteratively improve the alignment performance without the need of human annotations for preference data. These methods commonly utilize the same LLM to act
Externí odkaz:
http://arxiv.org/abs/2410.12735
Autor:
Jain, Kunal, Parayil, Anjaly, Mallick, Ankur, Choukse, Esha, Qin, Xiaoting, Zhang, Jue, Goiri, Íñigo, Wang, Rujia, Bansal, Chetan, Rühle, Victor, Kulkarni, Anoop, Kofsky, Steve, Rajmohan, Saravan
Large Language Model (LLM) workloads have distinct prefill and decode phases with different compute and memory requirements which should ideally be accounted for when scheduling input queries across different LLM instances in a cluster. However exist
Externí odkaz:
http://arxiv.org/abs/2408.13510
Autor:
Shetty, Manish, Chen, Yinfang, Somashekar, Gagan, Ma, Minghua, Simmhan, Yogesh, Zhang, Xuchao, Mace, Jonathan, Vandevoorde, Dax, Las-Casas, Pedro, Gupta, Shachee Mishra, Nath, Suman, Bansal, Chetan, Rajmohan, Saravan
The rapid growth in the use of Large Language Models (LLMs) and AI Agents as part of software development and deployment is revolutionizing the information technology landscape. While code generation receives significant attention, a higher-impact ap
Externí odkaz:
http://arxiv.org/abs/2407.12165
Autor:
Xia, Peng, Chen, Ze, Tian, Juanxi, Gong, Yangrui, Hou, Ruibo, Xu, Yue, Wu, Zhenbang, Fan, Zhiyuan, Zhou, Yiyang, Zhu, Kangyu, Zheng, Wenhao, Wang, Zhaoyang, Wang, Xiao, Zhang, Xuchao, Bansal, Chetan, Niethammer, Marc, Huang, Junzhou, Zhu, Hongtu, Li, Yun, Sun, Jimeng, Ge, Zongyuan, Li, Gang, Zou, James, Yao, Huaxiu
Artificial intelligence has significantly impacted medical applications, particularly with the advent of Medical Large Vision Language Models (Med-LVLMs), sparking optimism for the future of automated and personalized healthcare. However, the trustwo
Externí odkaz:
http://arxiv.org/abs/2406.06007
Autor:
Liu, Jun, Zhang, Chaoyun, Qian, Jiaxu, Ma, Minghua, Qin, Si, Bansal, Chetan, Lin, Qingwei, Rajmohan, Saravan, Zhang, Dongmei
Time series anomaly detection (TSAD) plays a crucial role in various industries by identifying atypical patterns that deviate from standard trends, thereby maintaining system integrity and enabling prompt response measures. Traditional TSAD models, w
Externí odkaz:
http://arxiv.org/abs/2405.15370
Autor:
Parayil, Anjaly, Zhang, Jue, Qin, Xiaoting, Goiri, Íñigo, Huang, Lexiang, Zhu, Timothy, Bansal, Chetan
Cloud providers introduce features (e.g., Spot VMs, Harvest VMs, and Burstable VMs) and optimizations (e.g., oversubscription, auto-scaling, power harvesting, and overclocking) to improve efficiency and reliability. To effectively utilize these featu
Externí odkaz:
http://arxiv.org/abs/2405.07250