Zobrazeno 1 - 10
of 513
pro vyhledávání: '"Rezagholizadeh A"'
The growth in prominence of large language models (LLMs) in everyday life can be largely attributed to their generative abilities, yet some of this is also owed to the risks and costs associated with their use. On one front is their tendency to \text
Externí odkaz:
http://arxiv.org/abs/2410.17477
We present a simple on the fly method for faster inference of large language models. Unlike other (self-)speculative decoding techniques, our method does not require fine-tuning or black-box optimization to generate a fixed draft model, relying inste
Externí odkaz:
http://arxiv.org/abs/2410.01028
Autor:
Rajabzadeh, Hossein, Jafari, Aref, Sharma, Aman, Jami, Benyamin, Kwon, Hyock Ju, Ghodsi, Ali, Chen, Boxing, Rezagholizadeh, Mehdi
Large Language Models (LLMs), with their increasing depth and number of parameters, have demonstrated outstanding performance across a variety of natural language processing tasks. However, this growth in scale leads to increased computational demand
Externí odkaz:
http://arxiv.org/abs/2409.14595
Despite their widespread adoption, large language models (LLMs) remain prohibitive to use under resource constraints, with their ever growing sizes only increasing the barrier for use. One noted issue is the high latency associated with auto-regressi
Externí odkaz:
http://arxiv.org/abs/2408.08470
Autor:
Kavehzadeh, Parsa, Pourreza, Mohammadreza, Valipour, Mojtaba, Zhu, Tinashu, Bai, Haoli, Ghodsi, Ali, Chen, Boxing, Rezagholizadeh, Mehdi
Deployment of autoregressive large language models (LLMs) is costly, and as these models increase in size, the associated costs will become even more considerable. Consequently, different methods have been proposed to accelerate the token generation
Externí odkaz:
http://arxiv.org/abs/2407.01955
Autor:
Dehghan, Mohammad, Alomrani, Mohammad Ali, Bagga, Sunyam, Alfonso-Hermelo, David, Bibi, Khalil, Ghaddar, Abbas, Zhang, Yingxue, Li, Xiaoguang, Hao, Jianye, Liu, Qun, Lin, Jimmy, Chen, Boxing, Parthasarathi, Prasanna, Biparva, Mahdi, Rezagholizadeh, Mehdi
The emerging citation-based QA systems are gaining more attention especially in generative AI search applications. The importance of extracted knowledge provided to these systems is vital from both accuracy (completeness of information) and efficienc
Externí odkaz:
http://arxiv.org/abs/2406.10393
Autor:
Mo, Fengran, Ghaddar, Abbas, Mao, Kelong, Rezagholizadeh, Mehdi, Chen, Boxing, Liu, Qun, Nie, Jian-Yun
In this paper, we study how open-source large language models (LLMs) can be effectively deployed for improving query rewriting in conversational search, especially for ambiguous queries. We introduce CHIQ, a two-step method that leverages the capabil
Externí odkaz:
http://arxiv.org/abs/2406.05013
Autor:
Huang, Chenyang, Ghaddar, Abbas, Kobyzev, Ivan, Rezagholizadeh, Mehdi, Zaiane, Osmar R., Chen, Boxing
Recently, there has been considerable attention on detecting hallucinations and omissions in Machine Translation (MT) systems. The two dominant approaches to tackle this task involve analyzing the MT system's internal states or relying on the output
Externí odkaz:
http://arxiv.org/abs/2406.01919
Autor:
Ghaddar, Abbas, Alfonso-Hermelo, David, Langlais, Philippe, Rezagholizadeh, Mehdi, Chen, Boxing, Parthasarathi, Prasanna
In this work, we dive deep into one of the popular knowledge-grounded dialogue benchmarks that focus on faithfulness, FaithDial. We show that a significant portion of the FaithDial data contains annotation artifacts, which may bias models towards com
Externí odkaz:
http://arxiv.org/abs/2405.15110
Large language models (LLMs) show an innate skill for solving language based tasks. But insights have suggested an inability to adjust for information or task-solving skills becoming outdated, as their knowledge, stored directly within their paramete
Externí odkaz:
http://arxiv.org/abs/2404.09339