• Liu et al 2021 - Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing
• Dong et al 2022 - A Survey on In-context Learning
• Qiao et al 2022 - Reasoning with Language Model Prompting: A Survey Very good
• Sahoo et al 2024 - A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications Not that great
• Tutorials, Courses, Slides and Guides
  • Guides
    • Prompt Engineering Guide This one is pretty good
  • Slides
    • UMass Amherst: Prompt-based learning
    • Stanford: Prompting, Instruction Finetuning, and RLHF
  • Blog: Lil'log Prompt Engineering
  • Github: BREX's Prompt Engineering Guide
  • Github: DAIR AI's Prompt Engineering Guide
  • Course: learnprompting.org

• Radford et al 2019 - Language Models Are Unsupervised Multitask Learners old link GPT-2
• Wei et al 2021 - Finetuned Language Models Are Zero-Shot Learners
• Lyu et al 2022 - Z-ICL: Zero-Shot In-Context Learning with Pseudo-Demonstrations

• Schick & Schütze 2020 - It's Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners
• Schick & Schütze 2020 - Few-Shot Text Generation with Natural Language Instructions GenPET, prompting for natural language generation
• Schick & Schütze 2021 - Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference Introduces PET, pre-dates GTP-3
• Brown et al 2021 - Language Models are Few-Shot Learners GPT-3
• Gao et al 2021 - Making Pre-trained Language Models Better Few-shot Learners

Prompting with a large context of many shots.

• Agarwal et al 2024 - Many-Shot In-Context Learning

• See Soft-prompting overview on p.3 of Zhao & Schütze 2021
• Shin et al 2020 - AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts
• Prefix-Tuning (aka P-Tuning): Liu et al 2021 - GPT understands, too Zhao 2021 finds this method to be the best.
• Qin & Eisner 2021 - Learning How to Ask: Querying LMs with Mixtures of Soft Prompts
• Prompt Tuning: Lester et al 2021 - The Power of Scale for Parameter-Efficient Prompt Tuning Can be seen as a “simplification of the recently proposed “prefix tuning” of Li and Liang (2021)”
• Zhao & Schütze 2021 - Discrete and Soft Prompting for Multilingual Models They find that soft prompting with an LSTM like Liu et al 2021 is best, both for English and cross-lingually.
• Liu et al 2021 - P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks
• Su et al 2021 - On Transferability of Prompt Tuning for Natural Language Processing
• Khashabi et al 2021 - Prompt Waywardness: The Curious Case of Discretized Interpretation of Continuous Prompts
• Tang et al 2022 - Context-Tuning: Learning Contextualized Prompts for Natural Language Generation
• Vu et al 2022 - SPoT: Better Frozen Model Adaptation through Soft Prompt Transfer - Multi-task, uses a library of learned soft prompts

Prompt tuning can be slower than fine-tuning. See the figure below.

Figure from Su et al 2022. See also figures 6-8 from Ding et al 2022.

See Prompt Engineering.

• Holtzman et al 2021 - Surface Form Competition: Why the Highest Probability Answer Isn’t Always Right
• Zhou et al 2023 - Batch Calibration: Rethinking Calibration for In-Context Learning and Prompt Engineering

• Wang et al 2022 - PromDA: Prompt-based Data Augmentation for Low-Resource NLU Tasks

See also Reasoning Chains.

• Overviews
  • Chu et al 2023 - A Survey of Chain of Thought Reasoning: Advances, Frontiers and Future
  • Besta et al 2024 - Topologies of Reasoning: Demystifying Chains, Trees, and Graphs of Thoughts
• Wei et al 2022 - Chain of Thought Prompting Elicits Reasoning in Large Language Models Introduced chain of thought prompting
• Kojima et al 2022 - Large Language Models are Zero-Shot Reasoners Introduced the prompt “Let's think step by step.”
• Wang et al 2022 - Self-Consistency Improves Chain of Thought Reasoning in Language Models Sample multiple chain of thought reasonings, and take the majority vote for the answer
• Wang et al 2022 - Iteratively Prompt Pre-trained Language Models for Chain of Thought
• Zelikman et al 2023 - STaR: Self-Taught Reasoner Bootstrapping Reasoning With Reasoning
• Dohan et al 2022 - Language Model Cascades
• Madaan & Yazdanbakhsh et al 2022 - Text and Patterns: For Effective Chain of Thought, It Takes Two to Tango
• Saparov & He 2022 - Language Models Are Greedy Reasoners: A Systematic Formal Analysis of Chain-of-Thought
• Yao et al 2022 - ReAct: Synergizing Reasoning and Acting in Language Models - The basis of LangChain
• Wang et 2023 - Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models
• Hao et al 2023 - Reasoning with Language Model is Planning with World Model
• Tree of Thought and Tree Search
  • Yao et al 2023 - Tree of Thoughts: Deliberate Problem Solving with Large Language Models
  • Long 2023 - Large Language Model Guided Tree-of-Thought
  • Feng et al 2023 - Alphazero-like Tree-Search can Guide Large Language Model Decoding and Training
  • Chi et al 2024 - THOUGHTSCULPT: Reasoning with Intermediate Revision and Search
• Li et al 2023 - Symbolic Chain-of-Thought Distillation: Small Models Can Also “Think” Step-by-Step
• Wang & Zhao 2023 - Metacognitive Prompting Improves Understanding in Large Language Models
• Yasunaga et al 2023 - Large Language Models as Analogical Reasoners Adds to the prompt “# Instruction: ## Recall relevant exemplars: ## Solve the initial problem:”, which helps more than “Let's think step by step.”
• Wang & Zhou et al 2024 - Chain-of-Thought Reasoning Without Prompting
• Chen et al 2024 - Masked Thought: Simply Masking Partial Reasoning Steps Can Improve Mathematical Reasoning Learning of Language Models Masks the CoT to get better results
• Zhang et al 2025 - LightThinker: Thinking Step-by-Step Compression

• Zhao & Schütze 2021 - Discrete and Soft Prompting for Multilingual Models

• Scao & Rush 2021 - How Many Data Points is a Prompt Worth? Prompts are very helpful in small data regimes, and are worth 100's of datapoints.
• Khashabi et al 2021 - Prompt Waywardness: The Curious Case of Discretized Interpretation of Continuous Prompts. See also Webson & Pavlick 2021
• Webson & Pavlick 2021 - Do Prompt-Based Models Really Understand the Meaning of Their Prompts?

For an overview see Tool Learning Papers

• Overviews
  • Qin et al 2023 - Tool Learning with Foundation Models
• Yao et al 2022 - ReAct: Synergizing Reasoning and Acting in Language Models. This kind of thing is implemented in LangChain
• Schick et al 2023 - Toolformer: Language Models Can Teach Themselves to Use Tools
• Qin et al 2023 - Tool Learning with Foundation Models
• Qin et al 2023 - ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs
  • Uses RapidAPI
• 2024 - InferCept: Efficient Intercept Support for Augmented Large Language Model Inference

• Mu et al 2024 - Learning to Compress Prompts with Gist Tokens

See Retrieval-Augmented Methods.

See also Membership Inference.

• Overviews
  • Ravaut et al 2024 - How Much are LLMs Contaminated? A Comprehensive Survey and the LLMSanitize Library
• Jacovi, et al 2023 - Stop Uploading Test Data in Plain Text: Practical Strategies for Mitigating Data Contamination by Evaluation Benchmarks
• Li & Flanigan 2023 - Task Contamination: Language Models May Not Be Few-Shot Anymore
• LLMSanitize: Ravaut et al 2024 - How Much are LLMs Contaminated? A Comprehensive Survey and the LLMSanitize Library
• Drinkall et al 2024 - Time Machine GPT
• GSM1k: Zhang et al 2024 - A Careful Examination of Large Language Model Performance on Grade School Arithmetic Re-evaluates GSM8K with a new dataset

• Scao & Rush 2021 - How Many Data Points is a Prompt Worth?
• Agarwal et al 2024 - Many-Shot In-Context Learning

• Mosbach et al 2023 - Few-shot Fine-tuning vs. In-context Learning: A Fair Comparison and Evaluation
• Balaguer et al 2024 - RAG vs Fine-tuning: Pipelines, Tradeoffs, and a Case Study on Agriculture

• Webson & Pavlick 2021 - Do Prompt-Based Models Really Understand the Meaning of Their Prompts?
• Min et al 2022 - Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?
• Garg et al 2022 - What Can Transformers Learn In-Context? A Case Study of Simple Function Classes
• Akyürek et al 2022 - What learning algorithm is in-context learning? Investigations with linear models
• Hendel et al 2023 - In-Context Learning Creates Task Vectors

• Datasets with Prompts for Evaluating Language Models
  • PromptSource: github Bach et al 2022 - PromptSource: An Integrated Development Environment and Repository for Natural Language Prompts 2,000 prompts for 170 datasets
  • BIG-Bench: github Srivastava et al 2022 - Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models Growing list of user-submitted tasks. Contains languages other than English
  • SuperNatural-Instructuctions: Wang et al 2022 - SUPER-NATURALINSTRUCTIONS: Generalization via Declarative Instructions on 1600+ NLP Tasks 1,600 instructions for 76 tasks across 55 languages
  • BIG-Bench-Hard
  • LM-Evaluation Harness: github

• LangChain Framework for building applications with prompting (chaining prompts, etc). This paper was the basis for it: Yao et al 2022 - ReAct: Synergizing Reasoning and Acting in Language Models

• Invited Talk @ NAACL 2021: Humans Learn From Task Descriptions and So Should Our Models - Hinrich Schütze

• Timo Schick

• Instruction-Tuning
• Few-Shot Learning
• Fine-Tuning
• Language Model
• Meta-Learning
• Pretraining
• Prompt Engineering
• Retrieval-Augmented Methods
• Natural Language Task Descriptions
• Zero-Shot Learning