A Survey on Context Learning

By Guangxu Xun, Xiaowei Jia, Vishrawas Gopalakrishnan, Aidong Zhang. IEEE Transactions on Knowledge and Data Engineering, 2017

1,654 citations recorded by the source corpus.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/0c0a778e6fdf7e36b1750c533dcc916f86608607

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Jianfeng Gao
2. Mohit Bansal
3. Dacheng Tao
4. Quoc V. Le
5. Neil Houlsby
6. Yi Tay
7. Percy Liang
8. Mostafa Dehghani

Citations in the corpus

• A Comprehensive Overview of Large Language Models (2023, neutral): “LLMs can be prompted in various prompt setups, where they can be adapted to the instructions without fine-tuning and in other cases with fine-tuning on data containing different prompt styles [25], [83], [84].”
• Can We Edit Factual Knowledge by In-Context Learning? (2023, methodology): “Demonstration Organization Previous studies (Liu et al., 2022; Rubin et al., 2022; Lu et al., 2022) suggest that demonstration organization including Demonstration Selection and Demonstration Ordering (Dong et al., 2023) is also crucial for ICL.”
• Text-to-SQL Empowered by Large Language Models: A Benchmark Evaluation (2023, background): “In natural language processing, the above progress that LLMs learn from contextual examples is referred as in-context learning [9].”
• Harms from Increasingly Agentic Algorithmic Systems (2023, background): “In the current scientific paradigm of large-scale deep-learning, one instead provides high-level learning algorithms that tend to be task-agnostic or adapt to new tasks efficiently [Beck et al. 2023; Dong et al. 2022].”
• The rise and potential of large language model based agents: a survey (2025, positive): “Numerous studies indicate that LLMs can perform a variety of complex tasks through in-context learning (ICL), which involves the models’ ability to learn from a few examples within a given context [186].”
• Foundation Models Defining a New Era in Vision: A Survey and Outlook (2025, neutral): “Related Reviews and Differences: In the literature, few recent works have reviewed large language models (LLMs) in natural language processing [2], [20], [21], [22], [23].”
• A Survey on Model Compression for Large Language Models (2023, neutral): “Emergent Abilities encompass several intriguing facets, including In-Context Learning (ICL) [ Dong et al. , 2023; Wang et al. , 2023b ] , Chain-of-Thought (CoT) [ Wei et al. , 2022b; Wang et al. , 2023c; Shi et al. , 2023 ] , and Instruction Following (IF) [ Ouyang et al. , 2022; Brooks et al. ,…”
• Recommender Systems in the Era of Large Language Models (LLMs) (2023, neutral): “According to the definition, prompts can be either discrete ( i.e. , hard) or continuous ( i.e. , soft) that guide LLMs to generate the expected output [96].”
• A Systematic Study and Comprehensive Evaluation of ChatGPT on Benchmark Datasets (2023, methodology): “Intuitively, in-context learning works by learning through analogies drawn from the given demonstration examples (Dong et al., 2023).”
• LLM Self Defense: By Self Examination, LLMs Know They Are Being Tricked (2023, methodology): “To further extend LLM S ELF D EFENSE , we can provide concrete examples of “harm” and use in-context learning as discussed in [7].”
• Large language models empowered agent-based modeling and simulation: a survey and perspectives (2023, neutral): “In addition, the basic rules or domain-speci fi c knowledge, such as the game rule, is also summarized with texts, which are received by the large language model agents with prompt engineering.”
• Unleashing the Power of Edge-Cloud Generative AI in Mobile Networks: A Survey of AIGC Services (2023, neutral): “analyzing the entire context of the input [46], including the dialogue history and user profile, to generate responses that are personalized and tailored to the user’s needs.”
• Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To! (2023, background): “Although alternatives like in-context learning (Dong et al., 2022) and prompt engineering (White et al.”
• What If the TV Was Off? Examining Counterfactual Reasoning Abilities of Multi-modal Language Models (2023, methodology): “We use chain-of-thought prompting [24] and in-context learning [20] to promote the quality of the questions produced by ChatGPT and keep the accuracy of new answers.”
• Predict Land Covers with Transition Modeling and Incremental Learning (2017, background): “Besides, they overlook the contextual information which provides advantage in classification task [23].”
• A Comprehensive Survey of AI-Generated Content (AIGC): A History of Generative AI from GAN to ChatGPT (2023, background): “For instance, in NLP, instead of fine-tuning, people sometimes prefer few-shot prompting [38], which refers to including a few examples selected from the dataset in the prompt, to help the model better understand task requirements.”
• PaLI-X: On Scaling up a Multilingual Vision and Language Model (2023, background): “This is usually referred to as in-context learning [14].”
• A Brief Overview of ChatGPT: The History, Status Quo and Potential Future Development (2023, methodology): “To further improve the model performance on few-shot or zero-shot [24] settings, GPT-3 [25] combines meta-learning [26], [27] with in-context learning [28] so that the generalization ability of the model has been greatly improved with surpassing most of the existing methods on various downstream tas”