BriefGPT.xyz
Mar, 2024
MALTO参与SemEval-2024任务6: 运用合成数据进行LLM幻觉检测
MALTO at SemEval-2024 Task 6: Leveraging Synthetic Data for LLM Hallucination Detection
HTML
PDF
Federico Borra, Claudio Savelli, Giacomo Rosso, Alkis Koudounas, Flavio Giobergia
TL;DR
自然语言生成面临若干挑战，我们通过引入数据增强管道和投票集成来解决生成流畅但不准确以及过度依赖流畅度评测指标的问题。
Abstract
In
natural language generation
(NLG), contemporary
large language models
(LLMs) face several challenges, such as generating fluent yet inaccurate outputs and reliance on fluency-centric metrics. This often leads
→