NLP Lineage

24 May 2022 · Reasoning

Zero-shot CoT

Large Language Models are Zero-Shot Reasoners

T. Kojima, S. Gu, M. Reid, Y. Matsuo, Y. Iwasawa

Date
24 May 2022
What the date marks
arXiv submission
Kind
Technique
Organisation
University of Tokyo / Google
Access
Not applicable
Other dates
NeurIPS 2022
Threads
In the guided tour
Chapter 12. Thinking longer
Learning phases
3. API usage and prompting

Notable: no later milestone in the data is documented as descending from it. How levels are decided

Read the primary source (arxiv.org)

Full summary in progress. Until then, this page shows the milestone's record from the research data.

What it changed

Before it

Chain of thought needed hand-written worked examples in the prompt.

The problem

Writing exemplars for every task is laborious and task-specific.

What it introduced

A single phrase asking the model to think step by step, with no examples, unlocks most of the same gain.

Why it matters

It showed the reasoning behaviour was already in the model, waiting for a cue. Cheap, universal, and much copied.

What modern models inherit

The instruction to reason before answering.

What came next

Absorbed into system prompts everywhere, then trained in directly by reasoning models.

Before and after

The milestones it descends from, and the milestones that descend from it. Work developed alongside it is not counted. Weak links are left out unless you include them.

Before: where it came from

Zero-shot CoT has 19 documented ancestors across 6 chapters, reaching back to Recurrent networks in 1990. Showing its direct parents and the 10 turning points among them.

Zero-shot CoT has 19 documented ancestors across 6 chapters, reaching back to Recurrent networks in 1990. Showing all of them.

Zero-shot CoT has 28 recorded ancestors, including indirect ones, across 6 chapters, reaching back to Recurrent networks in 1990. Showing its direct parents and the 10 turning points among them.

Zero-shot CoT has 28 recorded ancestors, including indirect ones, across 6 chapters, reaching back to Recurrent networks in 1990. Showing all of them.

  1. 2022Zero-shot CoTThis milestone.
  2. 2022Chain of thought Turning pointZero-shot CoT built directly on thisChapter 12. Thinking longer
  3. 2020GPT-3 Turning pointZero-shot CoT was influenced by thisChapter 5. Scale becomes a science
  4. 2020Scaling laws Turning pointGPT-3 was influenced by thisChapter 5. Scale becomes a science
  5. 2019Sparse Transformer Turning pointGPT-3 built directly on thisChapter 4. Pretrain once, use everywhere
  6. 2019GPT-2 Turning pointGPT-3 built directly on thisChapter 4. Pretrain once, use everywhere
  7. 2019WebTextGPT-2 was influenced by thisChapter 4. Pretrain once, use everywhere
  8. 2018GPT-1GPT-2 built directly on thisChapter 4. Pretrain once, use everywhere
  9. 2017Transformer Turning pointGPT-2 built directly on this
    Sparse Transformer built directly on this
    Scaling laws built directly on this    Chapter 3. Attention, then only attention
  10. 2016Layer normGPT-2 built directly on thisChapter 3. Attention, then only attention
  11. 2015Residual connectionsTransformer was influenced by thisChapter 3. Attention, then only attention
  12. 2015LM pretraining Turning pointGPT-1 was influenced by thisChapter 2. Words become vectors
  13. 2015BPE tokenizationGPT-2 built directly on thisChapter 2. Words become vectors
  14. 2015Luong attentionTransformer was influenced by thisChapter 3. Attention, then only attention
  15. 2014AdamTransformer was influenced by thisChapter 2. Words become vectors
  16. 2014Seq2seq Turning pointTransformer built directly on thisChapter 2. Words become vectors
  17. 2014Attention Turning pointTransformer built directly on thisChapter 3. Attention, then only attention
  18. 2014GRU / encoder-decoderAttention built directly on thisChapter 2. Words become vectors
  19. 1997LSTM Turning pointSeq2seq built directly on thisChapter 1. Language as probability
  20. 1990Recurrent networksLSTM was influenced by thisChapter 1. Language as probability
  1. 2022Zero-shot CoTThis milestone.
  2. 2022Chain of thought Turning pointZero-shot CoT built directly on thisChapter 12. Thinking longer
  3. 2020GPT-3 Turning pointZero-shot CoT was influenced by thisChapter 5. Scale becomes a science
  4. 2020Scaling laws Turning pointGPT-3 was influenced by thisChapter 5. Scale becomes a science
  5. 2019Sparse Transformer Turning pointGPT-3 built directly on thisChapter 4. Pretrain once, use everywhere
  6. 2019GPT-2 Turning pointGPT-3 built directly on thisChapter 4. Pretrain once, use everywhere
  7. 2019WebTextGPT-2 was influenced by thisChapter 4. Pretrain once, use everywhere
  8. 2019Transformer-XLSparse Transformer was influenced by this (indirect influence)Chapter 4. Pretrain once, use everywhere
  9. 2018GPT-1GPT-2 built directly on thisChapter 4. Pretrain once, use everywhere
  10. 2018ELMoGPT-1 was influenced by this (indirect influence)Chapter 4. Pretrain once, use everywhere
  11. 2018ULMFiTGPT-1 was influenced by this (indirect influence)Chapter 4. Pretrain once, use everywhere
  12. 2017CoVeELMo was influenced by thisChapter 3. Attention, then only attention
  13. 2017Transformer Turning pointGPT-2 built directly on this
    Sparse Transformer built directly on this
    Scaling laws built directly on this    Chapter 3. Attention, then only attention
  14. 2017ConvS2STransformer was influenced by this (indirect influence)Chapter 3. Attention, then only attention
  15. 2016Layer normGPT-2 built directly on thisChapter 3. Attention, then only attention
  16. 2015Residual connectionsTransformer was influenced by thisChapter 3. Attention, then only attention
  17. 2015LM pretraining Turning pointGPT-1 was influenced by thisChapter 2. Words become vectors
  18. 2015BPE tokenizationGPT-2 built directly on thisChapter 2. Words become vectors
  19. 2015Luong attentionTransformer was influenced by thisChapter 3. Attention, then only attention
  20. 2014AdamTransformer was influenced by thisChapter 2. Words become vectors
  21. 2014GloVeELMo built directly on thisChapter 2. Words become vectors
  22. 2014Seq2seq Turning pointTransformer built directly on thisChapter 2. Words become vectors
  23. 2014Attention Turning pointTransformer built directly on thisChapter 3. Attention, then only attention
  24. 2014GRU / encoder-decoderAttention built directly on thisChapter 2. Words become vectors
  25. 2013word2vecELMo was influenced by thisChapter 2. Words become vectors
  26. 2011NLP from scratchELMo was influenced by thisChapter 1. Language as probability
  27. 2003Neural LMNLP from scratch was influenced by this
    word2vec was influenced by this    Chapter 1. Language as probability
  28. 1997LSTM Turning pointSeq2seq built directly on this
    ELMo built directly on this    Chapter 1. Language as probability
  29. 1990Recurrent networksLSTM was influenced by thisChapter 1. Language as probability

After: what it led to

No documented descendants are recorded for Zero-shot CoT without weak links.

No documented descendants are recorded for Zero-shot CoT without weak links.

No recorded descendants are recorded for Zero-shot CoT.

No recorded descendants are recorded for Zero-shot CoT.

Direct links and their sources

Where it came from

Built directly on

Influenced by

What it led to

None recorded in the sources used.

Terms