Use large language model to enhance reasoning of another large language model through reward updated GRPO [PDF]
Scientific ReportsRecent advancements in deep learning have significantly transformed natural language processing (NLP), enabling sophisticated reasoning and text generation.
Yiqiao Yin
doaj +2 more sources
NEJM Evidence, 2023 Large Language ModelsIn the latest edition of Stats, STAT!, Fralick and colleagues explain the statistics behind large language models - used in chat bots like ChatGPT and Bard. While these new tools may seem remarkably intelligent, at their core they just assemble sentences based on statistics from large amounts of text.
Michael, Fralick +6 more
+9 more sources
The Language Essence of the World: A Linguistic Interpretation of the Large Language Model
Computer Sciences & Mathematics Forum, 2023 The source of characters is hieroglyph. Hieroglyph is the imitation and reference of the phenomena and objects in the natural world. The objects and phenomena in the natural world are the original image.
Leiming Shi, Peng Wu
doaj +1 more source
Risk Analysis and Response Strategies of Large Language Models for Security Governance [PDF]
中国工程科学To address the challenges of fragmented understanding of Large Language Model (LLM) security risks and the inadequacy of LLM risk classification and grading frameworks, this study aims to construct a comprehensive framework that integrates risk mechanism
Kun Jia +4 more
doaj +1 more source
Joint morphological-lexical language modeling for processing morphologically rich languages with application to dialectal Arabic [PDF]
, 2007 Language modeling for an inflected language such as Arabic poses new challenges for speech recognition and machine translation due to its rich morphology.
Afify, Mohamed +6 more
core +2 more sources
A Unified Multilingual Handwriting Recognition System using multigrams sub-lexical units [PDF]
, 2018 We address the design of a unified multilingual system for handwriting recognition. Most of multi- lingual systems rests on specialized models that are trained on a single language and one of them is selected at test time.
Paquet, Thierry +2 more
core +2 more sources
Using Parsimonious Language Models on Web Data [PDF]
, 2008 In this paper we explore the use of parsimonious language models for web retrieval. These models are smaller thus more efficient than the standard language models and are therefore well suited for large-scale web retrieval.
Hiemstra, Djoerd +3 more
core +4 more sources
RNN Language Model with Word Clustering and Class-based Output Layer [PDF]
, 2013 The recurrent neural network language model (RNNLM) has shown significant promise for statistical language modeling. In this work, a new class-based output layer method is introduced to further improve the RNNLM. In this method, word class information is
Johnson, Michael T +3 more
core +2 more sources
Bilingual Sentiment Embeddings: Joint Projection of Sentiment Across Languages [PDF]
, 2018 Sentiment analysis in low-resource languages suffers from a lack of annotated corpora to estimate high-performing models. Machine translation and bilingual word embeddings provide some relief through cross-lingual sentiment approaches.
Barnes, Jeremy +2 more
core +3 more sources
Large Margin Neural Language Model
, 2018 We propose a large margin criterion for training neural language models. Conventionally, neural language models are trained by minimizing perplexity (PPL) on grammatical sentences.
Huang, Jiaji +3 more
core +1 more source