import { Callout, FileTree } from 'nextra-theme-docs'
<Callout emoji="⚠️">
This section is under heavy development.
このセクションの内容は、鋭意開発進行中です。
</Callout>
This section consists of a collection and summary of notable and foundational LLMs. (Data adopted from [Papers with Code](https://paperswithcode.com/methods/category/language-models) and the recent work by [Zhao et al. (2023)](https://arxiv.org/pdf/2303.18223.pdf).
このセクションには、注目すべきLLMの基礎技術(モデル)の一覧とその概要をまとめています([Papers with Code](https://paperswithcode.com/methods/category/language-models)と[Zhao et al. (2023)](https://arxiv.org/pdf/2303.18223.pdf) による直近の研究成果を元に一覧を作成しています)。
## Models
| Model | Release Date | Description |
| モデル名 | 発表された年 | 概要説明 |
| --- | --- | --- |
| [BERT](https://arxiv.org/abs/1810.04805)| 2018 | Bidirectional Encoder Representations from Transformers |
| [GPT](https://s3-us-west-2.amazonaws.com/openai-assets/research-covers/language-unsupervised/language_understanding_paper.pdf) | 2018 | Improving Language Understanding by Generative Pre-Training |
| [HyperCLOVA](https://arxiv.org/abs/2109.04650) | 2021 | What Changes Can Large-scale Language Models Bring? Intensive Study on HyperCLOVA: Billions-scale Korean Generative Pretrained Transformers |
@ -37,8 +37,8 @@ This section consists of a collection and summary of notable and foundational LL
| [MT-NLG](https://arxiv.org/abs/2201.11990v3) | 2021 | Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model|
| [Yuan 1.0](https://arxiv.org/abs/2110.04725v2) | 2021| Yuan 1.0: Large-Scale Pre-trained Language Model in Zero-Shot and Few-Shot Learning |
| [WebGPT](https://arxiv.org/abs/2112.09332v3) | 2021 | WebGPT: Browser-assisted question-answering with human feedback |
| [Gopher](https://arxiv.org/abs/2112.11446v2) |2021 | Scaling Language Models: Methods, Analysis & Insights from Training Gopher |
| [ERNIE 3.0 Titan](https://arxiv.org/abs/2112.12731v1) |2021 | ERNIE 3.0 Titan: Exploring Larger-scale Knowledge Enhanced Pre-training for Language Understanding and Generation |
| [Gopher](https://arxiv.org/abs/2112.11446v2) |2021 | Scaling Language Models: Methods, Analysis & Insights from Training Gopher |
| [ERNIE 3.0 Titan](https://arxiv.org/abs/2112.12731v1) |2021 | ERNIE 3.0 Titan: Exploring Larger-scale Knowledge Enhanced Pre-training for Language Understanding and Generation |
| [GLaM](https://arxiv.org/abs/2112.06905) | 2021 | GLaM: Efficient Scaling of Language Models with Mixture-of-Experts |
| [InstructGPT](https://arxiv.org/abs/2203.02155v1) | 2022 | Training language models to follow instructions with human feedback |
| [GPT-NeoX-20B](https://arxiv.org/abs/2204.06745v1) | 2022 | GPT-NeoX-20B: An Open-Source Autoregressive Language Model |
@ -47,7 +47,7 @@ This section consists of a collection and summary of notable and foundational LL
| [Chinchilla](https://arxiv.org/abs/2203.15556) | 2022 | Shows that for a compute budget, the best performances are not achieved by the largest models but by smaller models trained on more data. |
| [Tk-Instruct](https://arxiv.org/abs/2204.07705v3) | 2022 | Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks |
| [PaLM](https://arxiv.org/abs/2204.02311v5) |2022| PaLM: Scaling Language Modeling with Pathways |
| [PaLM](https://arxiv.org/abs/2204.02311v5) |2022| PaLM: Scaling Language Modeling with Pathways |
| [OPT](https://arxiv.org/abs/2205.01068) | 2022 | OPT: Open Pre-trained Transformer Language Models |
| [BLOOM](https://arxiv.org/abs/2211.05100v3) | 2022 | BLOOM: A 176B-Parameter Open-Access Multilingual Language Model |
| [GLM-130B](https://arxiv.org/abs/2210.02414v1) | 2022 | GLM-130B: An Open Bilingual Pre-trained Model |
@ -59,6 +59,6 @@ This section consists of a collection and summary of notable and foundational LL
| [Galactica](https://arxiv.org/abs/2211.09085v1) | 2022 | Galactica: A Large Language Model for Science |
| [OPT-IML](https://arxiv.org/abs/2212.12017v3) | 2022 | OPT-IML: Scaling Language Model Instruction Meta Learning through the Lens of Generalization |
| [LLaMA](https://arxiv.org/abs/2302.13971v1) | 2023 | LLaMA: Open and Efficient Foundation Language Models |