Canada

Humor is a central aspect of human communication that has not been solved for artificial agents so far. Large language models (LLMs) are increasingly able to capture implicit and contextual information. Especially, OpenAI’s ChatGPT recently gained immense public attention. The GPT3-based model almost seems to communicate on a human level and can even tell jokes. Humor is an essential component of human communication. But is ChatGPT really funny?
We put ChatGPT’s sense of humor to the test. In a series of exploratory experiments around jokes, i.e., generation, explanation, and detection, we seek to understand ChatGPT’s capability to grasp and reproduce human humor. Since the model itself is not accessible, we applied prompt-based experiments. 
Our empirical evidence indicates that jokes are not hard-coded but mostly also not newly generated by the model. Over 90% of 1008 generated jokes were the same 25 Jokes. The system accurately explains valid jokes but also comes up with fictional explanations for invalid jokes. Joke-typical characteristics can mislead ChatGPT in the classification of jokes. ChatGPT has not solved computational humor yet but it
can be a big leap toward “funny” machines.

ACL 2023

ChatGPT is fun, but it is not funny! Humor is still challenging Large Language Models

chatgpt

computational humor

large language models

workshop paper

### Welcome to ACL 2023, the 61st Annual Meeting of the Association for Computational Linguistics! 
 The conference will be held in Toronto, Canada, July 9-14, 2023. 
Following the succession of the recent conferences in our field, ACL 2023 will adopt a hybrid format.
While the impact of Covid has considerably diminished in terms of traveling, obtaining visas to Canada
entails a very long process. Moreover, the global economic conditions pose challenges for many individuals to travel to conferences. Recognizing these circumstances, we know many participants may not be
able to attend the conference in person. Therefore, we are committed to providing a great virtual platform
so everyone has the opportunity to interact with other participants and enjoy the conference. Based on the
current registered participants, approxiately 30% have chosen to attend the conference virtually. Whether
you join us in person or virtually, we sincerely hope everyone has a remarkable conference experience. 
This General Chair’s message is where I express my gratitude to the many individuals who have made
enormous contributions to the conference over the past year.

Read [**ACL 2023 General Chair's message**](https://docs.google.com/document/d/1WobYM7norbG4dI48s75HfJoD89qgX5a_F-6U8AteLSA/edit?usp=sharing/) in full.

##### **[Conference Handbook](https://2023.aclweb.org/downloads/acl2023-handbook.pdf)**

The Association for Computational Linguistics (ACL) is the premier international scientific and professional society for people working on computational problems involving human language, a field often referred to as either computational linguistics or natural language processing.

Recent advances in the development of large Pretrained Language Models, such as GPT, BERT and Bloom, have achieved remarkable performance on a wide range of different NLP tasks. However, when used for text generation tasks, these models still have limitations when it comes to controlling the content and style of the generated text, often producing content that is incorrect, irrelevant, or inappropriate in the context of a given task. In this survey paper, we explore methods for controllable text generation with a focus on sentiment control. We systematically collect papers from the ACL Anthology, create a categorisation scheme based on different control techniques and controlled attributes, and use the scheme to categorise and compare methods. The result is a detailed and comprehensive overview of state-of-the-art techniques for sentiment-controlled text generation categorised on the basis of how the control is implemented and what attributes are controlled and providing a clear idea of their relative strengths and weaknesses.

How to Control Sentiment in Text Generation: A Survey of the State-of-the-Art in Sentiment-Control Techniques

Emotional reactions to Online Social Network posts have recently gained importance in the study of the online ecosystem. Prior to post publication, the number of received reactions can be predicted based on either the textual content of the post or the related metadata. 
However, existing approaches suffer from both the lack of semantic-aware language understanding models and the limited explainability of the prediction models. To overcome these issues, we present a new transformer-based method to predict the number of emotional reactions of different types to social posts. It leverages the attention mechanism to capture arbitrary semantic textual relations neglected by prior works. Furthermore, it also provides end-users with textual explanations of the predictions. The results achieved on a large collection of Facebook posts confirm the applicability of the presented methodology.

Transformer-based Prediction of Emotional Reactions to Online Social Network Posts

Monolinguals make up a minority of the world's speakers, and yet most language technologies lag behind in handling linguistic behaviours produced by bilingual and multilingual speakers. A commonly observed phenomenon in such communities is code-mixing, which is prevalent on social media, and thus requires attention in NLP research. In this presentation, I talk about the ability of pretrained language models to handle code-mixed data, with a focus on the impact of languages present in pretraining on the downstream performance of the model as measured on the task of sentiment analysis. The discussion leads to how we found that pretraining language has little effect on performance when the model sees code-mixed data during downstream finetuning. Differences in model performance that can be attributed to the pretraining languages come out when evaluating the models on code-mixed data in a zero-shot setting, after task-specific finetuning on a monolingual dataset. I present an analysis of these findings that also looks at model performance based on the composition of participating languages in the code-mixed datasets.

Transfer Learning for Code-Mixed Data: Do Pretraining Languages Matter?

This study evaluated ChatGPT’s ability to understand causal language in science papers and news by testing its accuracy in a task of labeling the strength of a claim as causal, conditional causal, correlational, or no relationship. The results show that ChatGPT is still behind the existing fine-tuned BERT models by a large margin. ChatGPT also had difficulty understanding conditional causal claims mitigated by hedges. However, its weakness may be utilized to improve the clarity of human annotation guideline. Chain-of-thought prompting was faithful and helpful for improving prompt performance, but finding the optimal prompt is difficult with inconsistent results and the lack of effective method to establish cause-effect between prompts and outcomes, suggesting caution when generalizing prompt engineering results across tasks or models.

Can ChatGPT Understand Causal Language in Science Claims?

Very large language models (LLMs) perform extremely well on a spectrum of NLP tasks in a zero-shot setting. However, little is known about their performance on human-level NLP problems which rely on understanding psychological concepts, such as assessing personality traits. In this work, we investigate the zero-shot ability of GPT-3 to estimate the Big 5 personality traits from users' social media posts. Through a set of systematic experiments, we find that zero-shot GPT-3 performance is somewhat close to an existing pre-trained SotA for broad classification upon injecting knowledge about the trait in the prompts. However, when prompted to provide fine-grained classification, its performance drops to close to a simple most frequent class (MFC) baseline. We further analyze where GPT-3 performs better, as well as worse, than a pretrained lexical model, illustrating systematic errors that suggest ways to improve LLMs on human-level NLP tasks. The code for this project is available on Github.

Systematic Evaluation of GPT-3 for Zero-Shot Personality Estimation

Emerging psychopathology studies are showing that patterns of changes in emotional state --- emotion dynamics --- are associated with overall well-being and mental health. More recently, there has been some work in tracking emotion dynamics through one's utterances, allowing for data to be collected on a larger scale across time and people. However, several questions about how emotion dynamics change with age, especially in children, and when determined through children's writing, remain unanswered. In this work, we use both a lexicon and a machine learning based approach to quantify characteristics of emotion dynamics determined from poems written by children of various ages. We show that both approaches point to similar trends: consistent increasing intensities for some emotions (e.g., anger, fear, joy, sadness, arousal, and dominance) with age and a consistent decreasing valence with age. We also find increasing emotional variability, rise rates (i.e., emotional reactivity), and recovery rates (i.e., emotional regulation) with age. These results act as a useful baselines for further research in how patterns of emotions expressed by children change with age, and their association with mental health.

Utterance Emotion Dynamics in Children's Poems: Emotional Changes across Age

In this paper, we present a dataset of subjective views (beliefs and attitudes) held by individuals or groups. We analyze the usefulness of the dataset by training a neural classifier that identifies belief-containing sentences that are relevant for our broader project of interest---scientific modeling of complex systems. We also explore and discuss difficulties related to annotation of subjective views and propose ways of addressing them.

Annotating and Training for Population Subjective Views

Stance Detection is the task of identifying the position of an author of a text towards an issue or a target. Previous studies on Stance Detection indicate that the existing systems are non-robust to the variations and errors in input sentences. Our proposed methodology uses Contrastive Learning to learn sentence representations by bringing semantically similar sentences and sentences implying the same stance closer to each other in the embedding space. We compare our approach to a pretrained transformer model directly finetuned with the stance datasets. We use char-level and word-level adversarial perturbation attacks to measure the resilience of the models and we show that our approach achieves better performances and is more robust to the different adversarial perturbations introduced to the test data. The results indicate that our approach performs better on small-sized and class-imbalanced stance datasets.

Exploration of Contrastive Learning Strategies toward more Robust Stance Detection

Social media is an extremely potent tool for influencing public opinion, particularly during important events such as elections, pandemics, and national conflicts. Emotions are a crucial aspect of this influence, but detecting them accurately in the political domain is a significant challenge due to the lack of suitable emotion labels and training datasets. In this paper, we present a generalized approach to emotion detection that can be adapted to the political domain with minimal performance sacrifice. Our approach is designed to be easily integrated into existing models without the need for additional training or fine-tuning. We demonstrate the zero-shot and few-shot performance of our model on the 2017 French presidential elections and propose efficient emotion groupings that would aid in effectively analyzing influence campaigns and agendas on social media.

Adapting Emotion Detection to Analyze Influence Campaigns on Social Media

Where do the meaning of emoji come from? Though it is often assumed that emoji are fully iconic, with meanings derived from their visual forms, we argue that this is only one component of their meaning. We surveyed users and non-users of the Chinese social media platform WeChat for their interpretations of emoji specific to WeChat. We find that some emoji show significant differences in their interpretations between users and non-users, as well as how familiar a person is with the specific emoji’s use. We argue that this reflects a more complex process for building the meaning of emoji on a platform than pure iconicity.

Premium content

ChatGPT is fun, but it is not funny! Humor is still challenging Large Language Models

Downloads

Next from ACL 2023

How to Control Sentiment in Text Generation: A Survey of the State-of-the-Art in Sentiment-Control Techniques

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES