本文提出了一種針對開源大型語言模型 (LLM) 的大規模提示探索方法 (PrExMe),用於評估機器翻譯和摘要任務,探討了不同提示策略對評估結果的影響,並發現了一些穩定和易變的模式。
본 논문에서는 다양한 프롬프트 전략을 사용하여 오픈 소스 대규모 언어 모델(LLM)을 머신 번역 및 요약 평가 지표로 활용하는 방법을 분석하고, 프롬프트 패턴의 안정성과 모델 성능에 미치는 영향을 평가합니다.
本稿では、機械翻訳と要約評価のためのオープンソース大規模言語モデル(LLM)ベースの評価指標について、720種類以上のプロンプトテンプレートを用いた大規模な分析を行い、その安定性と有効性を検証した。
Systematic exploration of prompt engineering reveals that open-source large language models can be effective for evaluating machine translation and summarization, but their performance is highly sensitive to even minor prompt variations, emphasizing the need for careful prompt design and selection.
This paper introduces a novel framework, ForPKG-1.0, for constructing a knowledge graph specifically for forestry policies, utilizing open-source large language models (LLMs) for information extraction and demonstrating its value in enhancing the performance of LLMs in retrieval-augmented generation tasks.
IPC, a novel computational method, leverages existing phonological knowledge of a learner's native language (L1) to construct composite sounds that approximate target phonemes in a second language (L2), leading to significant improvements in L2 pronunciation, particularly for Korean speakers learning English.
This research paper introduces SuDoSys, a novel structured dialogue system leveraging large language models (LLMs) and the World Health Organization's Problem Management Plus (PM+) guidelines to provide stage-aware psychological counseling, demonstrating promising results in generating coherent and effective counseling dialogues.
Med-Bot은 Llama-2 아키텍처와 AutoGPT-Q 양자화를 활용하여 의학 문헌에서 학습하고 사용자에게 정확하고 신뢰할 수 있는 의료 정보를 제공하는 AI 기반 챗봇입니다.
法律文書は複雑な構造や専門用語のため、一般の人々にとって理解が難しいことが多く、その結果、解釈のばらつきが生じる可能性があり、可読性を向上させるための標準的な指標の確立が課題となっている。
本文提出了一種名為「動態獎勵與提示優化」(DRPO)的新方法,無需微調或人工標註,即可實現大型語言模型的自我校準,並在多個基準測試中展現出超越微調模型的效能。