Results 281 to 290 of about 2,331,313 (323)
Some of the next articles are maybe not open access.
An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-Tuning
IEEE Transactions on Audio, Speech, and Language Processing, 2023Catastrophic forgetting (CF) is a phenomenon that occurs in machine learning when a model forgets previously learned information while acquiring new knowledge for achieving satisfactory performance in downstream tasks.
Yun Luo +5 more
semanticscholar +1 more source
LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models
Annual Meeting of the Association for Computational LinguisticsEfficient fine-tuning is vital for adapting large language models (LLMs) to downstream tasks. However, it requires non-trivial efforts to implement these methods on different models. We present LlamaFactory, a unified framework that integrates a suite of
Yaowei Zheng +5 more
semanticscholar +1 more source
Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success
RoboticsRecent vision-language-action models (VLAs) build upon pretrained vision-language models and leverage diverse robot datasets to demonstrate strong task execution, language following ability, and semantic generalization.
Moo Jin Kim, Chelsea Finn, Percy Liang
semanticscholar +1 more source
Visual-RFT: Visual Reinforcement Fine-Tuning
arXiv.orgReinforcement Fine-Tuning (RFT) in Large Reasoning Models like OpenAI o1 learns from feedback on its answers, which is especially useful in applications when fine-tuning data is scarce.
Ziyu Liu +7 more
semanticscholar +1 more source
Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey
Trans. Mach. Learn. Res.Large models represent a groundbreaking advancement in multiple application fields, enabling remarkable achievements across various tasks. However, their unprecedented scale comes with significant computational costs.
Zeyu Han +4 more
semanticscholar +1 more source
LESS: Selecting Influential Data for Targeted Instruction Tuning
International Conference on Machine LearningInstruction tuning has unlocked powerful capabilities in large language models (LLMs), effectively using combined datasets to develop generalpurpose chatbots. However, real-world applications often require a specialized suite of skills (e.g., reasoning).
Mengzhou Xia +4 more
semanticscholar +1 more source
Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models
International Conference on Machine LearningHarnessing the power of human-annotated data through Supervised Fine-Tuning (SFT) is pivotal for advancing Large Language Models (LLMs). In this paper, we delve into the prospect of growing a strong LLM out of a weak one without the need for acquiring ...
Zixiang Chen +4 more
semanticscholar +1 more source
Generative Representational Instruction Tuning
arXiv.orgAll text-based language problems can be reduced to either generation or embedding. Current models only perform well at one or the other. We introduce generative representational instruction tuning (GRIT) whereby a large language model is trained to ...
Niklas Muennighoff +7 more
semanticscholar +1 more source
Early Years Educator, 2014
Interactions between babies and parents provide the foundations for communication and language. We explore what happens to language when adults and children fail to ‘tune into’ each other early in life.
openaire +1 more source
Interactions between babies and parents provide the foundations for communication and language. We explore what happens to language when adults and children fail to ‘tune into’ each other early in life.
openaire +1 more source
Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?
Conference on Empirical Methods in Natural Language ProcessingWhen large language models are aligned via supervised fine-tuning, they may encounter new factual information that was not acquired through pre-training.
Zorik Gekhman +6 more
semanticscholar +1 more source