# Decrypting Cryptic Crosswords: Semantically Complex Wordplay Puzzles as a Target for NLP

arXiv: https://arxiv.org/abs/2104.08620 · HTML: https://mlsanity.com/paper/2104.08620/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 43 references that are in the map.
- First posted on arXiv 17 Apr 2021.

## Builds on (most cited first, 18 of 18)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing](https://mlsanity.com/paper/1808.06226/) (2018)
- [Adafactor: Adaptive Learning Rates with Sublinear Memory Cost](https://mlsanity.com/paper/1804.04235/) (2018)
- [A Primer in BERTology: What we know about how BERT works](https://mlsanity.com/paper/2002.12327/) (2020)
- [Experience Grounds Language](https://mlsanity.com/paper/2004.10151/) (2020)
- [The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence](https://mlsanity.com/paper/2002.06177/) (2020)
- [Muppet: Massive Multi-task Representations with Pre-Finetuning](https://mlsanity.com/paper/2101.11038/) (2021)
- [Injecting Numerical Reasoning Skills into Language Models](https://mlsanity.com/paper/2004.04487/) (2020)
- [QuaRTz: An Open-Domain Dataset of Qualitative Relationship Questions](https://mlsanity.com/paper/1909.03553/) (2019)
- [LIME: Learning Inductive Bias for Primitives of Mathematical Reasoning](https://mlsanity.com/paper/2101.06223/) (2021)
- [Pun Generation with Surprise](https://mlsanity.com/paper/1904.06828/) (2019)
- [Automatically Composing Representation Transformations as a Means for Generalization](https://mlsanity.com/paper/1807.04640/) (2018)
- [Pun-GAN: Generative Adversarial Network for Pun Generation](https://mlsanity.com/paper/1910.10950/) (2019)
- [Cryptonite: A Cryptic Crossword Benchmark for Extreme Ambiguity in Language](https://mlsanity.com/paper/2103.01242/) (2021)
- [Dr.Fill: Crosswords and an Implemented Solver for Singly Weighted CSPs](https://mlsanity.com/paper/1401.4597/) (2014)

## Built on by (most cited first, 16 of 16)

- [Evaluating Human-Language Model Interaction](https://mlsanity.com/paper/2212.09746/) (2022)
- [Puzzle Solving using Reasoning of Large Language Models: A Survey](https://mlsanity.com/paper/2402.11291/) (2024)
- [What do tokens know about their characters and how do they know it?](https://mlsanity.com/paper/2206.02608/) (2022)
- [Inducing Character-level Structure in Subword-based Language Models with Type-level Interchange Intervention Training](https://mlsanity.com/paper/2212.09897/) (2022)
- [VisualSphinx: Large-Scale Synthetic Vision Logic Puzzles for RL](https://mlsanity.com/paper/2505.23977/) (2025)
- [Automated Crossword Solving](https://mlsanity.com/paper/2205.09665/) (2022)
- [Missed Connections: Lateral Thinking Puzzles for Large Language Models](https://mlsanity.com/paper/2404.11730/) (2024)
- [Solving and Generating NPR Sunday Puzzles with Large Language Models](https://mlsanity.com/paper/2306.12255/) (2023)
- [Are LLMs Good Cryptic Crossword Solvers?](https://mlsanity.com/paper/2403.12094/) (2024)
- [On the Effect of (Near) Duplicate Subwords in Language Modelling](https://mlsanity.com/paper/2404.06508/) (2024)
- [Language Models are Crossword Solvers](https://mlsanity.com/paper/2406.09043/) (2024)
- [Non Verbis, Sed Rebus: Large Language Models are Weak Solvers of Italian Rebuses](https://mlsanity.com/paper/2408.00584/) (2024)
- [left|,circlearrowright,boxedBUS,right|: A Large and Diverse Multimodal Benchmark for evaluating the ability of Vision-Language Models to understand Rebus Puzzles](https://mlsanity.com/paper/2511.01340/) (2025)
- [Perturbation: A simple and efficient adversarial tracer for representation learning in language models](https://mlsanity.com/paper/2603.23821/) (2026)
- [Speaking of Language: Reflections on Metalanguage Research in NLP](https://mlsanity.com/paper/2604.02645/) (2026)
- [What Makes Cryptic Crosswords Challenging for LLMs?](https://mlsanity.com/paper/2412.09012/) (2024)

## Abstract (arXiv)

Cryptic crosswords, the dominant crossword variety in the UK, are a promising target for advancing NLP systems that seek to process semantically complex, highly compositional language. Cryptic clues read like fluent natural language but are adversarially composed of two parts: a definition and a wordplay cipher requiring character-level manipulations. Expert humans use creative intelligence to solve cryptics, flexibly combining linguistic, world, and domain knowledge. In this paper, we make two main contributions. First, we present a dataset of cryptic clues as a challenging new benchmark for NLP systems that seek to process compositional language in more creative, human-like ways. After showing that three non-neural approaches and T5, a state-of-the-art neural language model, do not achieve good performance, we make our second main contribution: a novel curriculum approach, in which the model is first fine-tuned on related tasks such as unscrambling words.We also introduce a challenging data split, examine the meta-linguistic capabilities of subword-tokenized models, and investigate model systematicity by perturbing the wordplay part of clues, showing that T5 exhibits behavior partially consistent with human solving strategies. Although our curricular approach considerably improves on the T5 baseline, our best-performing model still fails to generalize to the extent that humans can. Thus, cryptic crosswords remain an unsolved challenge for NLP systems and a potential source of future innovation.