Down and Across: Introducing Crossword-Solving as a New NLP Benchmark

TL;DR

This paper introduces crossword puzzle solving as a new NLP benchmark, providing a large dataset, diverse clues, and multiple baseline models to evaluate language understanding and reasoning capabilities.

Contribution

It presents a comprehensive crossword puzzle dataset, a new NLP task, and baseline models, establishing a benchmark for reasoning and knowledge in language understanding.

Findings

Baseline models show varying performance on the puzzle-solving task.

The dataset enables diverse reasoning and knowledge evaluation.

The proposed framework offers multiple metrics for comprehensive assessment.

Abstract

Solving crossword puzzles requires diverse reasoning capabilities, access to a vast amount of knowledge about language and the world, and the ability to satisfy the constraints imposed by the structure of the puzzle. In this work, we introduce solving crossword puzzles as a new natural language understanding task. We release the specification of a corpus of crossword puzzles collected from the New York Times daily crossword spanning 25 years and comprised of a total of around nine thousand puzzles. These puzzles include a diverse set of clues: historic, factual, word meaning, synonyms/antonyms, fill-in-the-blank, abbreviations, prefixes/suffixes, wordplay, and cross-lingual, as well as clues that depend on the answers to other clues. We separately release the clue-answer pairs from these puzzles as an open-domain question answering dataset containing over half a million unique clue-answer pairs. For the question answering task, our baselines include several sequence-to-sequence and retrieval-based generative models. We also introduce a non-parametric constraint satisfaction baseline for solving the entire crossword puzzle. Finally, we propose an evaluation framework which consists of several complementary performance metrics.