Time awareness is a fundamental ability of omni large language models, especially for understanding long videos and answering complex questions. Previous approa...
This chapter explores the application of Large Language Models in the legal domain, showcasing their potential to optimise and augment traditional legal tasks b...
This paper presents OnCoCo 1.0, a new public dataset for fine-grained message classification in online counseling. It is based on a new, integrative system of c...
Culture is a core component of human-to-human interaction and plays a vital role in how we perceive and interact with others. Advancements in the effectiveness ...
Role-playing agents (RPAs) must simultaneously master many conflicting skills -- following multi-turn instructions, exhibiting domain knowledge, and adopting a ...
Constructing a Pareto set is pivotal for navigating the capability-efficiency trade-offs in Large Language Models (LLMs); however, existing merging techniques r...
Constructing a Pareto set is pivotal for navigating the capability-efficiency trade-offs in Large Language Models (LLMs); however, existing merging techniques r...
LLMs are useful because they generalize so well. But can you have too much of a good thing? We show that a small amount of finetuning in narrow contexts can dra...
!Cover image for I Built a Brazilian Portuguese LLM from Scratch - Here's What I Learnedhttps://media2.dev.to/dynamic/image/width=1000,height=420,fit=cover,grav...
While scaling laws for Large Language Models (LLMs) traditionally focus on proxy metrics like pretraining loss, predicting downstream task performance has been ...
Retrieval-Augmented Generation (RAG) improves the factuality of large language models (LLMs) by grounding outputs in retrieved evidence, but faithfulness failur...
Gradually growing the depth of Transformers during training can not only reduce training cost but also lead to improved reasoning performance, as shown by MIDAS...
