@misc{kojima_large_2023,
	title = {Large {Language} {Models} are {Zero}-{Shot} {Reasoners}},
	url = {http://arxiv.org/abs/2205.11916},
	abstract = {Pretrained large language models (LLMs) are widely used in many sub-fields of natural language processing (NLP) and generally known as excellent few-shot learners with task-specific exemplars. Notably, chain of thought (CoT) prompting, a recent technique for eliciting complex multi-step reasoning through step-by-step answer examples, achieved the state-of-the-art performances in arithmetics and symbolic reasoning, difficult system-2 tasks that do not follow the standard scaling laws for LLMs. While these successes are often attributed to LLMs' ability for few-shot learning, we show that LLMs are decent zero-shot reasoners by simply adding "Let's think step by step" before each answer. Experimental results demonstrate that our Zero-shot-CoT, using the same single prompt template, significantly outperforms zero-shot LLM performances on diverse benchmark reasoning tasks including arithmetics (MultiArith, GSM8K, AQUA-RAT, SVAMP), symbolic reasoning (Last Letter, Coin Flip), and other logical reasoning tasks (Date Understanding, Tracking Shuffled Objects), without any hand-crafted few-shot examples, e.g. increasing the accuracy on MultiArith from 17.7\% to 78.7\% and GSM8K from 10.4\% to 40.7\% with large InstructGPT model (text-davinci-002), as well as similar magnitudes of improvements with another off-the-shelf large model, 540B parameter PaLM. The versatility of this single prompt across very diverse reasoning tasks hints at untapped and understudied fundamental zero-shot capabilities of LLMs, suggesting high-level, multi-task broad cognitive capabilities may be extracted by simple prompting. We hope our work not only serves as the minimal strongest zero-shot baseline for the challenging reasoning benchmarks, but also highlights the importance of carefully exploring and analyzing the enormous zero-shot knowledge hidden inside LLMs before crafting finetuning datasets or few-shot exemplars.},
	urldate = {2023-06-07},
	publisher = {arXiv},
	author = {Kojima, Takeshi and Gu, Shixiang Shane and Reid, Machel and Matsuo, Yutaka and Iwasawa, Yusuke},
	month = jan,
	year = {2023},
	note = {arXiv:2205.11916 [cs]},
	keywords = {Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Machine Learning},
	file = {arXiv.org Snapshot:/home/rok/Zotero/storage/BT2HY266/2205.html:text/html;Full Text PDF:/home/rok/Zotero/storage/292GVMWC/Kojima 等 - 2023 - Large Language Models are Zero-Shot Reasoners.pdf:application/pdf},
}

@misc{hoffmann_training_2022,
	title = {Training {Compute}-{Optimal} {Large} {Language} {Models}},
	url = {http://arxiv.org/abs/2203.15556},
	doi = {10.48550/arXiv.2203.15556},
	abstract = {We investigate the optimal model size and number of tokens for training a transformer language model under a given compute budget. We find that current large language models are significantly undertrained, a consequence of the recent focus on scaling language models whilst keeping the amount of training data constant. By training over 400 language models ranging from 70 million to over 16 billion parameters on 5 to 500 billion tokens, we find that for compute-optimal training, the model size and the number of training tokens should be scaled equally: for every doubling of model size the number of training tokens should also be doubled. We test this hypothesis by training a predicted compute-optimal model, Chinchilla, that uses the same compute budget as Gopher but with 70B parameters and 4\${\textbackslash}times\$ more more data. Chinchilla uniformly and significantly outperforms Gopher (280B), GPT-3 (175B), Jurassic-1 (178B), and Megatron-Turing NLG (530B) on a large range of downstream evaluation tasks. This also means that Chinchilla uses substantially less compute for fine-tuning and inference, greatly facilitating downstream usage. As a highlight, Chinchilla reaches a state-of-the-art average accuracy of 67.5\% on the MMLU benchmark, greater than a 7\% improvement over Gopher.},
	urldate = {2023-06-08},
	publisher = {arXiv},
	author = {Hoffmann, Jordan and Borgeaud, Sebastian and Mensch, Arthur and Buchatskaya, Elena and Cai, Trevor and Rutherford, Eliza and Casas, Diego de Las and Hendricks, Lisa Anne and Welbl, Johannes and Clark, Aidan and Hennigan, Tom and Noland, Eric and Millican, Katie and Driessche, George van den and Damoc, Bogdan and Guy, Aurelia and Osindero, Simon and Simonyan, Karen and Elsen, Erich and Rae, Jack W. and Vinyals, Oriol and Sifre, Laurent},
	month = mar,
	year = {2022},
	note = {arXiv:2203.15556 [cs]},
	keywords = {Computer Science - Machine Learning, Computer Science - Computation and Language},
	file = {arXiv Fulltext PDF:/home/rok/Zotero/storage/DXSZCEVL/Hoffmann 等 - 2022 - Training Compute-Optimal Large Language Models.pdf:application/pdf;arXiv.org Snapshot:/home/rok/Zotero/storage/FC3TJ3H2/2203.html:text/html},
}