Large language models (LLMs) like OpenAI's GPT series show promising progress in artificial intelligence, enabling models trained to predict the next word in a text to perform other tasks with intelligence.

Large Language Models

PaLM 540B, trained on 6144 TPU v4 chips using Pathways, achieves breakthrough performance in natural language understanding and generation tasks, outperforming finetuned state-of-the-art and average human performance on BIG-bench benchmarks.

PaLM: Scaling Language Modeling with Pathways

This paper provides a comprehensive overview of Large Language Models (LLMs) and their recent advances, highlighting background concepts and advanced topics for researchers and practitioners.

A Comprehensive Overview of Large Language Models

Large language models like GPT-3 can emergently find zero-shot solutions to a broad range of analogy problems, matching or surpassing human capabilities in most settings.

Emergent analogical reasoning in large language models

Large language models (LLMs) have a great deal to teach us about language, understanding, intelligence, sociality, and personhood, proving that statistics do amount to understanding and that complex sequence learning and social interaction may be sufficient for general intelligence.

Do Large Language Models Understand Us?

Large language models (LLMs) significantly improve performance and show special abilities in solving various NLP tasks, revolutionizing the AI community and advancing research on AI algorithms.

A Survey of Large Language Models

Large Language Models can enhance the Computational Social Science research pipeline by serving as zero-shot data annotators and bootstrapping challenging creative generation tasks.

Can Large Language Models Transform Computational Social Science?

Large language models, like ChatGPT and PaLM, show promise for advanced text comprehension and generation in applied mechanics, with potential for future applications and challenges.

Perspective: Large Language Models in Applied Mechanics

Large language models (LLMs) may serve as plausible models of human language understanding, despite their lack of symbolic structure and grounding.

Symbols and grounding in large language models

This paper identifies open problems and successful applications of Large Language Models, helping machine learning researchers better understand the field's current state and become more productive.

Challenges and Applications of Large Language Models

This paper reviews evaluation methods for large language models, focusing on what to evaluate, where to evaluate, and how to evaluate, to aid in the development of more proficient LLMs.

A Survey on Evaluation of Large Language Models

Large language models can effectively synthesize short Python programs from natural language descriptions, but their performance improves with fine-tuning and human feedback.

Program Synthesis with Large Language Models

Large Language Models (LLMs) improve with investment, learn and use representations of the outside world, and are not yet fully understood or steered by experts.

Eight Things to Know about Large Language Models

Scaling up language models can lead to emergent abilities, which are not present in smaller models but present in larger models, expanding their range of capabilities.

Emergent Abilities of Large Language Models

Finetuning large language models on psychological experiments can accurately represent human behavior and decision-making, potentially revolutionizing cognitive psychology and behavioral sciences.

Turning large language models into cognitive models

Baichuan 2 is a large-scale multilingual language model that outperforms other open-source models on public benchmarks and excels in vertical domains like medicine and law.

Baichuan 2: Open Large-scale Language Models

Large Language Models (LLMs) can enhance recommendation systems by providing high-quality representations of textual features and extensive coverage of external knowledge.

A Survey on Large Language Models for Recommendation

Large language models (LLMs) can serve as general sequence modelers, driven by in-context learning, potentially benefiting robotics and other fields.

Large Language Models as General Pattern Machines

Large language models (LLMs) have an intrinsic capacity for self-knowledge, but a significant gap remains between their capabilities and human proficiency in recognizing knowledge limits.

Do Large Language Models Know What They Don't Know?

Our structured pruning approach significantly speeds up large language models while maintaining performance, outperforming unstructured pruning methods at various sparsity levels.

Structured Pruning of Large Language Models

These studies suggest that large language models (LLMs) like GPT-3 and PaLM show significant advancements in natural language understanding and generation, outperforming state-of-the-art models and human performance in various tasks, and have broad applications in AI research and computational social science.