Skip to content

Entry

Chain-of-Thought Prompting

Appears in 5 awesome lists

foundational result; intermediate reasoning steps improve performance.

Open arxiv.org

Found in these lists

Awesome AI Papers

Section: NLP

SlowScore 53

Awesome GPT Prompt Engineering

Section: Papers

SlowScore 62

Awesome LLM Reasoning

Section: 2022 · [blog]

SlowScore 72

awesome-nlp

Section: Reasoning and Test-Time Compute · foundational result; intermediate reasoning steps improve performance.

FreshScore 91

awesome-chatgpt

Section: Recent advances in Prompt engineering

StaleScore 46

ReAct

The foundational paper defining the Thought/Action/Observation loop structure that underlies virtually every agent harness. Required reading for understanding why the loop is structured the way it is and where each harness component maps onto the reasoning-acting cycle.

In 7 listsDetails

Self-Consistency Improves Chain of Thought Reasoning in Language Models

Samples diverse reasoning paths and selects their most consistent answer to improve inference-time reliability. (ICLR 2023)

In 6 listsDetails

Attention Is All You Need

(AIAYN) - Introducing multi-head self-attention neural networks with positional encoding to do sentence-level NLP without any RNN nor CNN - this paper is a must-read (also see this explanation and this visualization of the paper).

In 6 listsDetails

Training language models to follow instructions with human feedback

This paper presents an RLHF approach to using supervised learning to fine-tuning. It is also known as a paper that illustrates the kernel of ChatGPT's thinking. Presumably, ChatGPT is an extended version of InstructGPT that enables fine-tuning on larger datasets.

In 6 listsDetails

Large Language Models are Zero-Shot Reasoners

"Let's think step by step" — zero-shot CoT milestone

In 4 listsDetails

Constitutional AI

Uses written principles and model-generated critiques to scale supervision while retaining explicit behavioral constraints. (arXiv 2022)

In 4 lists

Language Models are Few-Shot Learners

by Tom B. Brown (OpenAI) et al. - "We train GPT-3, an autoregressive language model with 175 billion parameters :scream:, 10x more than any previous non-sparse language model, and test its performance in the few-shot setting."

In 4 lists

Training Compute-Optimal Large Language Models

by Hoffmann et al. at DeepMind. TLDR: introduces a new 70B LM called "Chinchilla" that outperforms much bigger LMs (GPT-3, Gopher). DeepMind has found the secret to cheaply scale large language models — to be compute-optimal, model size and training data must be scaled equally. It shows that most…

In 3 lists