MUSE: A comprehensive AI framework for evaluating machine unlearning in language models
Language models (LMs) face significant challenges related to privacy and copyright because they are trained on large amounts of text ...
Tag: language
Projected language models: a large model presegmented into smaller models
This paper has been accepted for the Foundation Models in the Wild workshop at ICML 2024. Large language models are ...
MMLongBench-Doc: A comprehensive benchmark for evaluating long-context document understanding on large language and vision models
Document understanding (DU) focuses on the automatic interpretation and processing of documents, which encompass complex layout structures and multimodal elements ...
Planetarium: A new benchmark for assessing Master's degrees in translating natural language planning problem descriptions into the planning domain definition language (PDDL)
Large language models (LLMs) have gained significant attention in solving planning problems, but current methodologies need to be revised. Direct ...
Branch-and-merge method: Improving language adaptation in AI models by mitigating catastrophic forgetting and ensuring retention of core language skills while learning new languages
Language model adaptation is a crucial area in artificial intelligence, which focuses on improving large pre-trained language models to perform ...
Stanford researchers present In-Context Vectors (ICV): a scalable and efficient AI approach for fine-tuning large language models
Large language models (LLMs) have been crucial in driving artificial intelligence and natural language processing to new heights. These models ...
The reasoning capabilities of large language models are often overestimated | MIT News
When it comes to artificial intelligence, appearances can be deceptive. The mystery surrounding the inner workings of large language models ...
Review-LLM: An end-to-end AI framework for generating personalized reviews using large language models and historical user data in recommender systems
Generating personalized reviews in recommendation systems is an area of growing interest, particularly in creating personalized reviews based on users’ ...
Researchers from Stanford and the University of Buffalo present innovative AI methods to improve recall quality in recurrent language models with JRT-Prompt and JRT-RNN
Language modeling has made significant progress in developing algorithms to understand, generate, and manipulate human language. These advances have led ...
TheoremLlama: A comprehensive framework for training a broad, general-purpose language model to become a Lean expert4
A major advance in mathematical reasoning is the use of computer-verifiable formal languages, such as Lean, to prove mathematical theorems. ...
Recommended.
