paper-with-me

홈 › Papers

Advanced Natural-based interaction for the ITAlian language: LLaMAntino-3-ANITA

2024-05-11 · Marco Polignano, Pierpaolo Basile, Giovanni Semeraro

In the pursuit of advancing natural language processing for the Italian language, we introduce a state-of-the-art Large Language Model (LLM) based on the novel Meta LLaMA-3 model: LLaMAntino-3-ANITA-8B-Inst-DPO-ITA. We fine-tuned the original 8B parameters instruction tuned model using the Supervised Fine-tuning (SFT) technique on the English and Italian language datasets in order to improve the original performance. Consequently, a Dynamic Preference Optimization (DPO) process has been used to align preferences, avoid dangerous and inappropriate answers, and limit biases and prejudices. Our model leverages the efficiency of QLoRA to fine-tune the model on a smaller portion of the original model weights and then adapt the model specifically for the Italian linguistic structure, achieving significant improvements in both performance and computational efficiency. Concurrently, DPO is employed to refine the model's output, ensuring that generated content aligns with quality answers. The synergy between SFT, QLoRA's parameter efficiency and DPO's user-centric optimization results in a robust LLM that excels in a variety of tasks, including but not limited to text completion, zero-shot classification, and contextual understanding. The model has been extensively evaluated over standard benchmarks for the Italian and English languages, showing outstanding results. The model is freely available over the HuggingFace hub and, examples of use can be found in our GitHub repository. https://huggingface.co/swap-uniba/LLaMAntino-3-ANITA-8B-Inst-DPO-ITA

📄 PDF Abstract BibTeX arXiv:2405.07101

Code (0)

등록된 구현이 없습니다.

Tasks

Computational EfficiencyLanguage ModellingLarge Language Modelzero-shot-classificationZero-Shot Learning

Methods 이 논문이 사용한 방법론

DPO 설명 없음
ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…
SFT Shrink and Fine-Tune, or SFT, is a type of distillation that avoids explicit distillation by copying parameters to a student student model and then fine-tuning.…

Similar Papers 제목 키워드 기반

LLaMAntino: LLaMA 2 Models for Effective Text Generation in Italian Language

2023-12-15 · Pierpaolo Basile, Elio Musacchio, Marco Polignano, Lucia Siciliani 외

Large Language Models represent state-of-the-art linguistic models designed to equip computers with the ability to comprehend natural language. With its exceptional capacity to capture complex contextual relationships, t…

Language ModellingLarge Language ModelNatural Language UnderstandingPhilosophy+1

Benchmarking EngGPT2-16B-A3B against Comparable Italian and International Open-source LLMs

2026-05-08 · Andrea Sassella, Andrea Chizzola, Tommaso Bianchi, Luca Alessandrelli 외 arxiv

This report benchmarks the performance of ENGINEERING Ingegneria Informatica S.p.A.'s EngGPT2MoE-16B-A3B LLM, a 16B parameter Mixture of Experts (MoE) model with 3B active parameters. Performance is investigated across a…

Igea: a Decoder-Only Language Model for Biomedical Text Generation in Italian

2024-07-08 · Tommaso Mario Buonocore, Simone Rancati, Enea Parimbelli

The development of domain-specific language models has significantly advanced natural language processing applications in various specialized fields, particularly in biomedicine. However, the focus has largely been on En…

Computational EfficiencyDecoderGeneral KnowledgeLanguage Modeling+2

Presenting a system of human-machine interaction for performing map tasks.

2014-05-01 · LREC 2014 5 · Gabriele Pallotti, Francesca Frontini, Fabio Aff{\`e}, Monica Monachini 외

A system for human machine interaction is presented, that offers second language learners of Italian the possibility of assessing their competence by performing a map task, namely by guiding the a virtual follower throug…

Language Acquisition

Evalita-LLM: Benchmarking Large Language Models on Italian

2025-02-04 · Bernardo Magnini, Roberto Zanoli, Michele Resta, Martin Cimmino 외

We describe Evalita-LLM, a new benchmark designed to evaluate Large Language Models (LLMs) on Italian tasks. The distinguishing and innovative features of Evalita-LLM are the following: (i) all tasks are native Italian, …

BenchmarkingMultiple-choice