Do Androids Laugh at Electric Sheep? Humor “Understanding” Benchmarks from The New Yorker Caption Contest

Large neural networks can now generate jokes, but do they really “understand” humor? We challenge AI models with three tasks derived from the New Yorker Cartoon Caption Contest: matching a joke to a cartoon, identifying a winning caption, and explaining why a winning caption is funny. These tasks encapsulate progressively more sophisticated aspects of “understanding” a cartoon; key elements are the complex, often surprising relationships between images and captions and the frequent inclusion of indirect and playful allusions to human experience and culture. We investigate both multimodal and language-only models: the former are challenged with the cartoon images directly, while the latter are given multifaceted descriptions of the visual scene to simulate human-level visual understanding. We find that both types of models struggle at all three tasks. For example, our best multimodal models fall 30 accuracy points behind human performance on the matching task, and, even when provided ground-truth visual scene descriptors, human-authored explanations are preferred head-to-head over the best machine-authored ones (few-shot GPT-4) in more than 2/3 of cases. We release models, code, leaderboard, and corpus, which includes newly-gathered annotations describing the image’s locations/entities, what’s unusual in the scene, and an explanation of the joke.

[1]  S. Gu,et al.  Large Language Models are Zero-Shot Reasoners , 2022, NeurIPS.

[2]  Andrew M. Dai,et al.  PaLM: Scaling Language Modeling with Pathways , 2022, J. Mach. Learn. Res..

[3]  Adrian S. Wong,et al.  Socratic Models: Composing Zero-Shot Multimodal Reasoning with Language , 2022, ICLR.

[4]  Jingren Zhou,et al.  OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework , 2022, ICML.

[5]  Dale Schuurmans,et al.  Chain of Thought Prompting Elicits Reasoning in Large Language Models , 2022, NeurIPS.

[6]  Mark O. Riedl,et al.  Reframing Human-AI Collaboration for Generating Free-Text Explanations , 2021, NAACL.

[7]  Matthew E. Peters,et al.  Few-Shot Self-Rationalization with Natural Language Prompts , 2021, NAACL-HLT.

[8]  Chenhao Tan On the Diversity and Limits of Human Explanations , 2021, NAACL.

[9]  Louis-Philippe Morency,et al.  Humor Knowledge Enriched Transformer for Understanding Multimodal Humor , 2021, AAAI.

[10]  Ilya Sutskever,et al.  Learning Transferable Visual Models From Natural Language Supervision , 2021, ICML.

[11]  Vinay P. Namboodiri,et al.  Multimodal Humor Dataset: Predicting Laughter tracks for Sitcoms , 2021, 2021 IEEE Winter Conference on Applications of Computer Vision (WACV).

[12]  S. Gelly,et al.  An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale , 2020, ICLR.

[13]  Jack Hessel,et al.  Does My Multimodal Model Learn Cross-modal Interactions? It’s Harder to Tell than You Might Think! , 2020, EMNLP.

[14]  Olatunji Ruwase,et al.  DeepSpeed: System Optimizations Enable Training Deep Learning Models with Over 100 Billion Parameters , 2020, KDD.

[15]  Mark Chen,et al.  Language Models are Few-Shot Learners , 2020, NeurIPS.

[16]  Issa Annamoradnejad,et al.  ColBERT: Using BERT Sentence Embedding for Humor Detection , 2020, ArXiv.

[17]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[18]  Colin Raffel,et al.  Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer , 2019, J. Mach. Learn. Res..

[19]  Lysandre Debut,et al.  HuggingFace's Transformers: State-of-the-art Natural Language Processing , 2019, ArXiv.

[20]  Philip R. Smith ‘Getting Arno’: the New Yorker cartoon caption competition , 2019, Journal of Graphic Novels and Comics.

[21]  Iryna Gurevych,et al.  Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks , 2019, EMNLP.

[22]  Pavel Braslavski,et al.  Large Dataset and Language Model Fun-Tuning for Humor Recognition , 2019, ACL.

[23]  Verónica Pérez-Rosas,et al.  Towards Multimodal Sarcasm Detection (An _Obviously_ Perfect Paper) , 2019, ACL.

[24]  Kazuhiro Fukui,et al.  News2meme: An Automatic Content Generator from News Based on Word Subspaces from Text and Image , 2019, 2019 16th International Conference on Machine Vision Applications (MVA).

[25]  Yejin Choi,et al.  The Curious Case of Neural Text Degeneration , 2019, ICLR.

[26]  Louis-Philippe Morency,et al.  UR-FUNNY: A Multimodal Language Dataset for Understanding Humor , 2019, EMNLP.

[27]  Fallianda Fallianda,et al.  Analyzing Humor in Newspaper Comic Strips Using Verbal-Visual Analysis , 2018, Lingua Cultura.

[28]  Adam Trischler,et al.  How Reasonable are Common-Sense Reasoning Tasks: A Case-Study on the Winograd Schema Challenge and SWAG , 2018, EMNLP.

[29]  Oriol Vinyals,et al.  Representation Learning with Contrastive Predictive Coding , 2018, ArXiv.

[30]  Kota Yoshida,et al.  Neural Joking Machine : Humorous image captioning , 2018, ArXiv.

[31]  Rachel Rudinger,et al.  Hypothesis Only Baselines in Natural Language Inference , 2018, *SEMEVAL.

[32]  Matt Post,et al.  A Call for Clarity in Reporting BLEU Scores , 2018, WMT.

[33]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[34]  Devi Parikh,et al.  Punny Captions: Witty Wordplay in Image Descriptions , 2017, NAACL.

[35]  Ashwin K. Vijayakumar,et al.  We are Humor Beings: Understanding and Predicting Visual Humor , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[36]  Lalit Jain,et al.  NEXT: A System for Real-World Development, Evaluation, and Application of Active Learning , 2015, NIPS.

[37]  Dafna Shahaf,et al.  Inside Jokes: Identifying Humorous Cartoon Captions , 2015, KDD.

[38]  Dragomir R. Radev,et al.  Humor in Collective Discourse: Unsupervised Funniness Detection in the New Yorker Cartoon Caption Contest , 2015, LREC.

[39]  K. Gwet Handbook of Inter-Rater Reliability: The Definitive Guide to Measuring the Extent of Agreement Among Raters , 2014 .

[40]  John A. Bateman,et al.  Text and Image , 2014 .

[41]  Pietro Perona,et al.  Microsoft COCO: Common Objects in Context , 2014, ECCV.

[42]  Dragomir R. Radev,et al.  Random Walk Factoid Annotation for Collective Discourse , 2013, ACL.

[43]  M. Burkhard,et al.  Preface , 2010, IOP Conference Series: Materials Science and Engineering.

[44]  V. Tsakona Language and image interaction in cartoons: Towards a multimodal theory of humor , 2009 .

[45]  Rada Mihalcea,et al.  Characterizing Humour: An Exploration of Features in Humorous Texts , 2009, CICLing.

[46]  Salvatore Attardo,et al.  A primer for the linguistics of humor , 2008 .

[47]  T. Lombrozo The structure and function of explanations , 2006, Trends in Cognitive Sciences.

[48]  Carlo Strapparava,et al.  Technologies That Make You Smile: Adding Humor to Text-Based Applications , 2006, IEEE Intelligent Systems.

[49]  Carlo Strapparava,et al.  Making Computers Laugh: Investigations in Automatic Humor Recognition , 2005, HLT.

[50]  Michael Billig,et al.  Laughter and Ridicule: Towards a Social Critique of Humour , 2005 .

[51]  Chin-Yew Lin,et al.  ROUGE: A Package for Automatic Evaluation of Summaries , 2004, ACL 2004.

[52]  Carlo Strapparava,et al.  Getting serious about the development of computational humor , 2003, IJCAI 2003.

[53]  Antinus Nijholt,et al.  Humour Research: State of Art , 2002 .

[54]  Salim Roukos,et al.  Bleu: a Method for Automatic Evaluation of Machine Translation , 2002, ACL.

[55]  Christiane Fellbaum,et al.  Book Reviews: WordNet: An Electronic Lexical Database , 1999, CL.

[56]  Kim Binsted,et al.  An Implemented Model of Punning Riddles , 1994, AAAI.

[57]  Victor Raskin,et al.  Semantic mechanisms of humor , 1984 .

[58]  W. Fry Sweet Madness: A Study of Humor , 1968 .

[59]  A. Linear-probe,et al.  Learning Transferable Visual Models From Natural Language Supervision , 2021 .

[60]  Miriam Amin,et al.  A Survey on Approaches to Computational Humor Generation , 2020, LATECHCLFL.

[61]  Robert Nowak,et al.  A KL-LUCB algorithm for Large-Scale Crowdsourcing , 2017, NIPS.

[62]  S. Hirsch Image Music Text , 2016 .

[63]  William Yang Wang,et al.  I Can Has Cheezburger? A Nonparanormal Approach to Combining Textual and Visual Information for Predicting and Generating Popular Meme Descriptions , 2015, NAACL.

[64]  Alessandro Valitutti,et al.  How Many Jokes are Really Funny? Towards a New Approach to the Evaluation of Computational Humour Generators , 2011 .

[65]  James M. Jones,et al.  Interaction effects of picture and caption on humor ratings of cartoons , 1979 .

[66]  Charles R. Gruner,et al.  Understanding Laughter: The Workings of Wit and Humor , 1978 .

[67]  T. Shultz A Cognitive-Developmental Analysis of Humour , 1976 .

[68]  Harvey Mindess,et al.  Laughter and liberation , 1971 .

[69]  S. Freud The Standard Edition of the Complete Psychological Works of Sigmund Freud , 1953 .