이진욱님의 블로그

https://arxiv.org/abs/2410.09008

https://arxiv.org/pdf/2410.13413

https://arxiv.org/abs/2411.17188

https://arxiv.org/pdf/2411.16594Assessment and evaluation have long been critical challenges in artificial intelligence (AI) and natural language processing (NLP). However, traditional methods, whether matching-based or embedding-based, often fall short of judging subtle attributes and delivering satisfactory results. Recent advancements in Large Language Models (LLMs) inspire the "LLM-as-ajudge..

https://github.com/saskia-rr/Evaluating-Constitutions GitHub - saskia-rr/Evaluating-ConstitutionsContribute to saskia-rr/Evaluating-Constitutions development by creating an account on GitHub.github.comhttps://arxiv.org/pdf/2411.10168 The growing capabilities of large language models (LLMs) have led to their use as substitutes for human feedback for training and assessing other LLMs. These method..

https://arxiv.org/pdf/2411.14794

https://arxiv.org/pdf/2410.04055

https://arxiv.org/pdf/2411.08147

https://arxiv.org/pdf/2410.05928

https://arxiv.org/pdf/2411.00855