Zobrazeno 1 - 10
of 228
pro vyhledávání: '"Xie, Yuxi"'
Software engineers operating in complex and dynamic environments must continuously adapt to evolving requirements, learn iteratively from experience, and reconsider their approaches based on new insights. However, current large language model (LLM)-b
Externí odkaz:
http://arxiv.org/abs/2410.20285
Autor:
Xie, Yuxi, Goyal, Anirudh, Wu, Xiaobao, Yin, Xunjian, Xu, Xiao, Kan, Min-Yen, Pan, Liangming, Wang, William Yang
Iterative refinement has emerged as an effective paradigm for enhancing the capabilities of large language models (LLMs) on complex tasks. However, existing approaches typically implement iterative refinement at the application or prompting level, re
Externí odkaz:
http://arxiv.org/abs/2410.09675
Humans perform visual perception at multiple levels, including low-level object recognition and high-level semantic interpretation such as behavior understanding. Subtle differences in low-level details can lead to substantial changes in high-level p
Externí odkaz:
http://arxiv.org/abs/2410.04345
Language Language Models (LLMs) face safety concerns due to potential misuse by malicious users. Recent red-teaming efforts have identified adversarial suffixes capable of jailbreaking LLMs using the gradient-based search algorithm Greedy Coordinate
Externí odkaz:
http://arxiv.org/abs/2408.14866
Autor:
Xie, Yuxi, Goyal, Anirudh, Zheng, Wenyue, Kan, Min-Yen, Lillicrap, Timothy P., Kawaguchi, Kenji, Shieh, Michael
We introduce an approach aimed at enhancing the reasoning capabilities of Large Language Models (LLMs) through an iterative preference learning process inspired by the successful strategy employed by AlphaZero. Our work leverages Monte Carlo Tree Sea
Externí odkaz:
http://arxiv.org/abs/2405.00451
Autor:
Do, Xuan Long, Zhao, Yiran, Brown, Hannah, Xie, Yuxi, Zhao, James Xu, Chen, Nancy F., Kawaguchi, Kenji, Shieh, Michael, He, Junxian
We propose a new method, Adversarial In-Context Learning (adv-ICL), to optimize prompt for in-context learning (ICL) by employing one LLM as a generator, another as a discriminator, and a third as a prompt modifier. As in traditional adversarial lear
Externí odkaz:
http://arxiv.org/abs/2312.02614
Autor:
Li, Kaixin, Hu, Qisheng, Zhao, Xu, Chen, Hui, Xie, Yuxi, Liu, Tiedong, Xie, Qizhe, He, Junxian
Code editing encompasses a variety of pragmatic tasks that developers deal with daily. Despite its relevance and practical usefulness, automatic code editing remains an underexplored area in the evolution of deep learning models, partly due to data s
Externí odkaz:
http://arxiv.org/abs/2310.20329
We introduce ECHo (Event Causality Inference via Human-Centric Reasoning), a diagnostic dataset of event causality inference grounded in visio-linguistic social scenarios. ECHo employs real-world human-centric deductive information building on a tele
Externí odkaz:
http://arxiv.org/abs/2305.14740
Chain-of-Thought (CoT) and Program-Aided Language Models (PAL) represent two distinct reasoning methods, each with its own strengths. CoT employs natural language, offering flexibility and interpretability, while PAL utilizes programming language, yi
Externí odkaz:
http://arxiv.org/abs/2305.14333
Breaking down a problem into intermediate steps has demonstrated impressive performance in Large Language Model (LLM) reasoning. However, the growth of the reasoning chain introduces uncertainty and error accumulation, making it challenging to elicit
Externí odkaz:
http://arxiv.org/abs/2305.00633