# A Large Self-Annotated Corpus for Sarcasm

arXiv: https://arxiv.org/abs/1704.05579 · HTML: https://mlsanity.com/paper/1704.05579/

## Key facts

- Cited by 64 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Apr 2017.

## Built on by (most cited first, 30 of 64)

- [BIG-Bench Extra Hard](https://mlsanity.com/paper/2502.19187/) (2025)
- [Do LLMs Understand Social Knowledge? Evaluating the Sociability of Large Language Models with SocKET Benchmark](https://mlsanity.com/paper/2305.14938/) (2023)
- [Do Humans Trust Advice More if it Comes from AI? An Analysis of Human-AI Interactions](https://mlsanity.com/paper/2107.07015/) (2021)
- [A Multimodal Corpus for Emotion Recognition in Sarcasm](https://mlsanity.com/paper/2206.02119/) (2022)
- [Human-Aligned Calibration for AI-Assisted Decision Making](https://mlsanity.com/paper/2306.00074/) (2023)
- [Uncalibrated Models Can Improve Human-AI Collaboration](https://mlsanity.com/paper/2202.05983/) (2022)
- [A Survey of Multimodal Sarcasm Detection](https://mlsanity.com/paper/2410.18882/) (2024)
- [A Dual-Channel Framework for Sarcasm Recognition by Detecting Sentiment Conflict](https://mlsanity.com/paper/2109.03587/) (2021)
- [Pragmatics in Language Grounding: Phenomena, Tasks, and Modeling Approaches](https://mlsanity.com/paper/2211.08371/) (2022)
- [Source Attribution for Large Language Model-Generated Data](https://mlsanity.com/paper/2310.00646/) (2023)
- [Both Matter: Enhancing the Emotional Intelligence of Large Language Models without Compromising the General Intelligence](https://mlsanity.com/paper/2402.10073/) (2024)
- [SoMeLVLM: A Large Vision Language Model for Social Media Processing](https://mlsanity.com/paper/2402.13022/) (2024)
- [Large Language Models for Subjective Language Understanding: A Survey](https://mlsanity.com/paper/2508.07959/) (2025)
- [On Sarcasm Detection with OpenAI GPT-based Models](https://mlsanity.com/paper/2312.04642/) (2023)
- [Reddit Entity Linking Dataset](https://mlsanity.com/paper/2101.01228/) (2021)
- [Sarcasm in Sight and Sound: Benchmarking and Expansion to Improve Multimodal Sarcasm Detection](https://mlsanity.com/paper/2310.01430/) (2023)
- [DocMSU: A Comprehensive Benchmark for Document-level Multimodal Sarcasm Understanding](https://mlsanity.com/paper/2312.16023/) (2023)
- [Latent-Optimized Adversarial Neural Transfer for Sarcasm Detection](https://mlsanity.com/paper/2104.09261/) (2021)
- [Sarcasm Detection: A Comparative Study](https://mlsanity.com/paper/2107.02276/) (2021)
- [Systematic Literature Review: Computational Approaches for Humour Style Classification](https://mlsanity.com/paper/2402.01759/) (2024)
- [Generalizable Sarcasm Detection Is Just Around The Corner, Of Course!](https://mlsanity.com/paper/2404.06357/) (2024)
- [BESSTIE: A Benchmark for Sentiment and Sarcasm Classification for Varieties of English](https://mlsanity.com/paper/2412.04726/) (2024)
- [Cyclical Curriculum Learning](https://mlsanity.com/paper/2202.05531/) (2022)
- [Meta-Tuning LLMs to Leverage Lexical Knowledge for Generalizable Language Style Understanding](https://mlsanity.com/paper/2305.14592/) (2023)
- [ContextLeak: Auditing Leakage in Private In-Context Learning Methods](https://mlsanity.com/paper/2512.16059/) (2025)
- [A Survey on Automated Sarcasm Detection on Twitter](https://mlsanity.com/paper/2202.02516/) (2022)
- [ConFiguRe: Exploring Discourse-level Chinese Figures of Speech](https://mlsanity.com/paper/2209.07678/) (2022)
- [An Evaluation of State-of-the-Art Large Language Models for Sarcasm Detection](https://mlsanity.com/paper/2312.03706/) (2023)
- [Perspective Transition of Large Language Models for Solving Subjective Tasks](https://mlsanity.com/paper/2501.09265/) (2025)
- [Impact of Target Word and Context on End-to-End Metonymy Detection](https://mlsanity.com/paper/2112.03256/) (2021)

## Abstract (arXiv)

We introduce the Self-Annotated Reddit Corpus (SARC), a large corpus for sarcasm research and for training and evaluating systems for sarcasm detection. The corpus has 1.3 million sarcastic statements -- 10 times more than any previous dataset -- and many times more instances of non-sarcastic statements, allowing for learning in both balanced and unbalanced label regimes. Each statement is furthermore self-annotated -- sarcasm is labeled by the author, not an independent annotator -- and provided with user, topic, and conversation context. We evaluate the corpus for accuracy, construct benchmarks for sarcasm detection, and evaluate baseline methods.