Výsledky vyhledávání - "Petcu, Roxana"

Report

Leveraging Graph Structures to Detect Hallucinations in Large Language Models

Autor: Nonkes, Noa, Agaronian, Sergei, Kanoulas, Evangelos, Petcu, Roxana

Publikováno v: Proceedings of the TextGraphs-17 Workshop, ACL 2024

Large language models are extensively applied across a wide range of tasks, such as customer support, content creation, educational tutoring, and providing financial guidance. However, a well-known drawback is their predisposition to generate halluci

Externí odkaz: http://arxiv.org/abs/2407.04485

Zobrazit plný text záznamu

Report

A Survey on Recent Advances in Conversational Data Generation

Autor: Soudani, Heydar, Petcu, Roxana, Kanoulas, Evangelos, Hasibi, Faegheh

Recent advancements in conversational systems have significantly enhanced human-machine interactions across various domains. However, training these systems is challenging due to the scarcity of specialized dialogue data. Traditionally, conversationa

Externí odkaz: http://arxiv.org/abs/2405.13003

Zobrazit plný text záznamu

Report

Efficient data selection employing Semantic Similarity-based Graph Structures for model training

Autor: Petcu, Roxana, Maji, Subhadeep

Recent developments in natural language processing (NLP) have highlighted the need for substantial amounts of data for models to capture textual information accurately. This raises concerns regarding the computational resources and time required for

Externí odkaz: http://arxiv.org/abs/2402.14888

Zobrazit plný text záznamu

Report

Self-seeding and Multi-intent Self-instructing LLMs for Generating Intent-aware Information-Seeking dialogs

Autor: Askari, Arian, Petcu, Roxana, Meng, Chuan, Aliannejadi, Mohammad, Abolghasemi, Amin, Kanoulas, Evangelos, Verberne, Suzan

Identifying user intents in information-seeking dialogs is crucial for a system to meet user's information needs. Intent prediction (IP) is challenging and demands sufficient dialogs with human-labeled intents for training. However, manually annotati

Externí odkaz: http://arxiv.org/abs/2402.11633

Zobrazit plný text záznamu