Skip to content
Review

Mutual Information-Guided Prompt Engineering for Synthetic Tourism Review Generation in Mexican Magical Towns

· 0 citations · 8 references

TL;DR

An information-theoretic framework that guides Large Language Models (LLMs) through discriminative lexical information automatically extracted from the original corpus is proposed and experimental evaluation demonstrates that incorporating information-theoretic lexical guidance improves lexical diversity, semantic consistency, and downstream sentiment classification performance compared with conventional prompting strategies.

View source

Similar papers

Review

Lexical-Only Synthetic Data Augmentation for Imbalanced Tourism Sentiment Analysis

Experimental results show that the proposed approach improves the Macro F1 score over the Majority Class baseline, demonstrating that isolated lexical information contributes to the learning process and indicates that lexical features alone are insufficient for accurately representing minority sentiment classes.

Juan David Jurado-Buch, L. Bustio-Martínez, Vitali Herrera-Semenets et al. · 0 citations
Review

Generating Useful Synthetic Text with Local LLMs and Data Valuation: A Study on Spanish Tourism Reviews

This work explores two complementary generation strategies: few-shot prompting with locally deployed instruction-tuned Qwen3.5 models and LoRA-based fine-tuning of meta-llama/Llama-3.2-3B on balanced review subsets, followed by a three-stage quality filtering pipeline that assesses label consistency, linguistic realism...

Carlos Minutti-Martinez, B. Escalante-Ramírez, Jimena Olveres · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.