NLP Lineage

5 April 2022 · Language modelling

PaLM

PaLM: Scaling Language Modeling with Pathways

A. Chowdhery, S. Narang, J. Devlin, et al. (67 authors)

Date
5 April 2022
What the date marks
arXiv submission
Kind
Model
Organisation
Google Research
Access
internal
Other dates
JMLR 2023
Threads
In the guided tour
Chapter 5. Scale becomes a science
Learning phases
2. How LLMs are trained

A milestone: 3 later milestones descend directly from it, and its descendants reach 8 of the 10 threads. 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

Reasoning gains from scale looked modest and uneven.

The problem

Would more scale actually help on tasks requiring multi-step inference?

What it introduced

540B dense parameters trained across TPU pods, reaching breakthrough results on reasoning benchmarks with chain-of-thought prompting.

Why it matters

It supplied the strongest evidence for the scale-unlocks-reasoning story, and the discontinuities that fed the emergence debate.

What modern models inherit

The pairing of large scale with chain-of-thought as the reasoning recipe of 2022.

What came next

Flan-PaLM instruction-tuned it; Gemini succeeded it.

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

PaLM has 22 documented ancestors across 6 chapters, reaching back to Recurrent networks in 1990. Showing its direct parents and the 12 turning points among them.

PaLM has 22 documented ancestors across 6 chapters, reaching back to Recurrent networks in 1990. Showing all of them.

PaLM has 31 recorded ancestors, including indirect ones, across 6 chapters, reaching back to Recurrent networks in 1990. Showing its direct parents and the 12 turning points among them.

PaLM has 31 recorded ancestors, including indirect ones, across 6 chapters, reaching back to Recurrent networks in 1990. Showing all of them.

  1. 2022PaLMThis milestone.
  2. 2022Chain of thought Turning pointPaLM was influenced by thisChapter 12. Thinking longer
  3. 2020GShard Turning pointPaLM was influenced by thisChapter 5. Scale becomes a science
  4. 2020GPT-3 Turning pointChain of thought was influenced by thisChapter 5. Scale becomes a science
  5. 2020Scaling laws Turning pointGPT-3 was influenced by thisChapter 5. Scale becomes a science
  6. 2019Sparse Transformer Turning pointGPT-3 built directly on thisChapter 4. Pretrain once, use everywhere
  7. 2019GPT-2 Turning pointGPT-3 built directly on thisChapter 4. Pretrain once, use everywhere
  8. 2019WebTextGPT-2 was influenced by thisChapter 4. Pretrain once, use everywhere
  9. 2018GPT-1GPT-2 built directly on thisChapter 4. Pretrain once, use everywhere
  10. 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
  11. 2017Sparse MoE Turning pointGShard built directly on thisChapter 3. Attention, then only attention
  12. 2016Layer normGPT-2 built directly on thisChapter 3. Attention, then only attention
  13. 2015Residual connectionsTransformer was influenced by thisChapter 3. Attention, then only attention
  14. 2015LM pretraining Turning pointGPT-1 was influenced by thisChapter 2. Words become vectors
  15. 2015BPE tokenizationGPT-2 built directly on thisChapter 2. Words become vectors
  16. 2015Luong attentionTransformer was influenced by thisChapter 3. Attention, then only attention
  17. 2014AdamTransformer was influenced by thisChapter 2. Words become vectors
  18. 2014Seq2seq Turning pointTransformer built directly on thisChapter 2. Words become vectors
  19. 2014Attention Turning pointTransformer built directly on thisChapter 3. Attention, then only attention
  20. 2014GRU / encoder-decoderAttention built directly on thisChapter 2. Words become vectors
  21. 1997LSTM Turning pointSparse MoE built directly on thisChapter 1. Language as probability
  22. 1991Mixture of expertsSparse MoE built directly on thisChapter 1. Language as probability
  23. 1990Recurrent networksLSTM was influenced by thisChapter 1. Language as probability
  1. 2022PaLMThis milestone.
  2. 2022Chain of thought Turning pointPaLM was influenced by thisChapter 12. Thinking longer
  3. 2020GShard Turning pointPaLM was influenced by thisChapter 5. Scale becomes a science
  4. 2020GPT-3 Turning pointChain of thought was influenced by thisChapter 5. Scale becomes a science
  5. 2020Scaling laws Turning pointGPT-3 was influenced by thisChapter 5. Scale becomes a science
  6. 2019Sparse Transformer Turning pointGPT-3 built directly on thisChapter 4. Pretrain once, use everywhere
  7. 2019GPT-2 Turning pointGPT-3 built directly on thisChapter 4. Pretrain once, use everywhere
  8. 2019WebTextGPT-2 was influenced by thisChapter 4. Pretrain once, use everywhere
  9. 2019Transformer-XLSparse Transformer was influenced by this (indirect influence)Chapter 4. Pretrain once, use everywhere
  10. 2018GPT-1GPT-2 built directly on thisChapter 4. Pretrain once, use everywhere
  11. 2018ELMoGPT-1 was influenced by this (indirect influence)Chapter 4. Pretrain once, use everywhere
  12. 2018ULMFiTGPT-1 was influenced by this (indirect influence)Chapter 4. Pretrain once, use everywhere
  13. 2017CoVeELMo was influenced by thisChapter 3. Attention, then only attention
  14. 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
  15. 2017ConvS2STransformer was influenced by this (indirect influence)Chapter 3. Attention, then only attention
  16. 2017Sparse MoE Turning pointGShard built directly on thisChapter 3. Attention, then only attention
  17. 2016Layer normGPT-2 built directly on thisChapter 3. Attention, then only attention
  18. 2015Residual connectionsTransformer was influenced by thisChapter 3. Attention, then only attention
  19. 2015LM pretraining Turning pointGPT-1 was influenced by thisChapter 2. Words become vectors
  20. 2015BPE tokenizationGPT-2 built directly on thisChapter 2. Words become vectors
  21. 2015Luong attentionTransformer was influenced by thisChapter 3. Attention, then only attention
  22. 2014AdamTransformer was influenced by thisChapter 2. Words become vectors
  23. 2014GloVeELMo built directly on thisChapter 2. Words become vectors
  24. 2014Seq2seq Turning pointTransformer built directly on thisChapter 2. Words become vectors
  25. 2014Attention Turning pointTransformer built directly on thisChapter 3. Attention, then only attention
  26. 2014GRU / encoder-decoderAttention built directly on thisChapter 2. Words become vectors
  27. 2013word2vecELMo was influenced by thisChapter 2. Words become vectors
  28. 2011NLP from scratchELMo was influenced by thisChapter 1. Language as probability
  29. 2003Neural LMNLP from scratch was influenced by this
    word2vec was influenced by this    Chapter 1. Language as probability
  30. 1997LSTM Turning pointSparse MoE built directly on thisChapter 1. Language as probability
  31. 1991Mixture of expertsSparse MoE built directly on thisChapter 1. Language as probability
  32. 1990Recurrent networksLSTM was influenced by thisChapter 1. Language as probability

After: what it led to

PaLM has 13 documented descendants across 7 chapters, reaching forward to Gemini 3.1 in 2026. Showing its direct descendants.

PaLM has 13 documented descendants across 7 chapters, reaching forward to Gemini 3.1 in 2026. Showing all of them.

PaLM has 13 recorded descendants, including indirect ones, across 7 chapters, reaching forward to Gemini 3.1 in 2026. Showing its direct descendants.

PaLM has 13 recorded descendants, including indirect ones, across 7 chapters, reaching forward to Gemini 3.1 in 2026. Showing all of them.

  1. 2022PaLMThis milestone.
  2. 2022Emergent abilitieswas influenced by PaLMChapter 5. Scale becomes a science
  3. 2022Flan-T5 / Flan-PaLMbuilt directly on PaLMChapter 6. Teaching models what people want
  4. 2023Emergence miragesucceeded Emergent abilitiesChapter 5. Scale becomes a science
  5. 2023Gemini 1.0succeeded PaLMChapter 11. Seeing and hearing
  6. 2024Gemini 1.5built directly on Gemini 1.0Chapter 10. Reading more than you remember
  7. 2024Gemmabuilt directly on Gemini 1.0Chapter 8. The weights get out
  8. 2024Gemini 2.0built directly on Gemini 1.5Chapter 14. Where the threads meet now
  9. 2025Gemini 2.5was influenced by Gemini 1.5Chapter 12. Thinking longer
  10. 2025IMO goldbuilt directly on Gemini 2.5Chapter 12. Thinking longer
  11. 2025IMO-ProofBenchwas influenced by IMO goldChapter 12. Thinking longer
  12. 2025Gemini 3was influenced by Gemini 1.5Chapter 14. Where the threads meet now
  13. 2026Gemini 3 Deep Thinkbuilt directly on Gemini 3Chapter 12. Thinking longer
  14. 2026Gemini 3.1built directly on Gemini 3Chapter 14. Where the threads meet now
  1. 2022PaLMThis milestone.
  2. 2022Emergent abilitieswas influenced by PaLMChapter 5. Scale becomes a science
  3. 2022Flan-T5 / Flan-PaLMbuilt directly on PaLMChapter 6. Teaching models what people want
  4. 2023Emergence miragesucceeded Emergent abilitiesChapter 5. Scale becomes a science
  5. 2023Gemini 1.0succeeded PaLMChapter 11. Seeing and hearing
  6. 2024Gemini 1.5built directly on Gemini 1.0Chapter 10. Reading more than you remember
  7. 2024Gemmabuilt directly on Gemini 1.0Chapter 8. The weights get out
  8. 2024Gemini 2.0built directly on Gemini 1.5Chapter 14. Where the threads meet now
  9. 2025Gemini 2.5was influenced by Gemini 1.5Chapter 12. Thinking longer
  10. 2025IMO goldbuilt directly on Gemini 2.5Chapter 12. Thinking longer
  11. 2025IMO-ProofBenchwas influenced by IMO goldChapter 12. Thinking longer
  12. 2025Gemini 3was influenced by Gemini 1.5Chapter 14. Where the threads meet now
  13. 2026Gemini 3 Deep Thinkbuilt directly on Gemini 3Chapter 12. Thinking longer
  14. 2026Gemini 3.1built directly on Gemini 3Chapter 14. Where the threads meet now

Direct links and their sources

Where it came from

Influenced by

Developed alongside

What it led to

Built on directly by

Influenced

Replaced by

Terms