Efficient Adaptation of Large Language Models in Natural Language Processing

The rapid growth of Large Language Models (LLMs) has significantly improved performance across a wide range of Natural Language Processing (NLP) tasks, including Information Retrieval (IR). Despite their strong generalisation capabilities, LLMs still require domain- and task-specific fine-tuning to achieve competitive performance in highly specialised scenarios. LLMs exhibit some generalisation abilities, though these are not uniformly reliable across tasks and domains. In particular, adapting these models to downstream domains and tasks has become challenging as full fine-tuning is computationally expensive, memory-intensive and difficult to scale across multiple domains and tasks. Over the past few years, substantial research has been made into efficient fine-tuning of LLMs, resulting in various proposed approaches. Nevertheless, it remains an ongoing research area with several unresolved issues and challenges.

Grantor dc:publisher

Politecnico di Torino

Year dc:date

2026

Author dc:creator

• Braga, Marco

• Settore INFO-01/A - Informatica

Statement dc:rights

• info:eu-repo/semantics/openAccess
• license:Creative commons
• license uri:http://creativecommons.org/licenses/by-nc-nd/4.0/

Language dc:language

eng

Handle dc:identifier

https://hdl.handle.net/11583/3013101

OAI identifier oai:identifier

oai:iris.polito.it:11583/3013101