Ireland

Harmful or abusive online content has been increasing over time and it has been raising concerns among social media platforms, government agencies, and policymakers. Such harmful or abusive content can have significant negative impact on society, e.g., cyberbullying can lead to suicides, rumors about COVID-19 can cause vaccine hesitance, promotion of fake cures for COVID-19 can cause health harms and deaths. The content that is posted and shared online can be textual, visual, or a combination of both, e.g., in a meme. Here, we describe our experiments in detecting the roles of the entities (hero, villain, victim) in harmful memes, which is part of the CONSTRAINT-2022 shared task, as well as our system for the task. We further provide a comparative analysis on different experimental settings (i.e., unimodal, multimodal, attention, and augmentation). For reproducibility, we make the experimental code publicly available (https://github.com/robi56/harmful_memes_block_fusion)

ACL 2022

Detecting the Role of an Entity in Harmful Memes: Techniques and their Limitations

harmfulness

images

propaganda

fake news

disinformation

multimodality

text

# Welcome everyone to ACL 2022!

The 60th Annual Meeting of the Association for Computational Linguistics is taking place May 22-27, 2022 as a hybrid event, in Dublin and online. We are happy to welcome all of you to this anniversary edition with an almost 50-50 in-person and virtual participation. 
The main conference program features oral presentations, in-person and virtual posters and demo sessions, a plenary session for our best paper presentations and awards, three amazing keynote events and two new initiatives of invited talks: Spotlight Talks for Young Rising Stars (STIRS) and The Next Big Idea Talks. Posters (including Findings of ACL 2022) and demos are grouped by areas for both the in-person and the virtual sessions. For the virtual component, the talks will be on Zoom and the posters and the demos will be in GatherTown. The Student Research Workshop will have an oral session and a poster session as part of Poster Session 1. The program also features eight Tutorials and 28 Workshops. 

 
We wish you a wonderful conference! 
[**The ACL 2022 Organizing Committee**](https://www.2022.aclweb.org/organisers)
 
[**Conference Handbook**](https://drive.google.com/file/d/1_BUCMfhMVrjG9E2e71aHdHeE28KSje0l/view?usp=sharing) 
[**Mini Handbook**](https://drive.google.com/file/d/1qlBKl0wzmlVF1oCeMQl3BahLd9nLP5Ce/view?usp=sharing) 
[**Posters and Demo guides**](https://drive.google.com/file/d/1UucMAoCNncIOaH1rMMDa0owuG9qgvJTG/view?usp=sharing)

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 (NLP). 

workshop paper

We propose our solution to the multimodal semantic role labeling task from the CONSTRAINT'22 workshop. The task aims at classifying entities in memes into classes such as ``hero'' and ``villain''. We use several pre-trained multi-modal models to jointly encode the text and image of the memes, and implement three systems to classify the role of the entities. We propose dynamic sampling strategies to tackle the issue of class imbalance. Finally, we perform a qualitative analysis of the representations of the entities.


Fine-tuning and Sampling Strategies for Multimodal Role Labeling of Entities under Class Imbalance

We present the findings of the shared task at the CONSTRAINT 2022 Workshop: Hero, Villain, and Victim: Dissecting harmful memes for Semantic role labeling of entities. The task aims to delve deeper into the domain of meme comprehension by deciphering the connotations behind the entities present in a meme. In more nuanced terms, the shared task focuses on determining the victimizing, glorifying, and vilifying intentions embedded in meme entities to explicate their connotations. To this end, we curate HVVMemes, a novel meme dataset of about 7000 memes spanning the domains of COVID-19 and US Politics, each containing entities and their associated roles: hero, villain, victim, or none. The shared task attracted 105 participants, but eventually only 6 submissions were made. The successful submissions majorly relied on an ensemble of language and multimodal models. The best submission achieved an F1-score of 58.67.

Findings of the CONSTRAINT 2022 Shared Task on Detecting the Hero, the Villain, and the Victim in Memes

the COVID-19 pandemic, the spread of misinformation on online social media has grown exponentially. Unverified bogus claims on these platforms regularly mislead people, leading them to believe in half-baked truths. The current vogue is to employ manual fact-checkers to verify claims to combat this avalanche of misinformation. However, establishing such claims' veracity is becoming increasingly challenging, partly due to the plethora of information available, which is difficult to process manually. Thus, it becomes imperative to verify claims automatically without human interventions. To cope up with this issue, we propose an automated claim verification solution encompassing two steps -- document retrieval and veracity prediction. For the retrieval module, we employ a hybrid search-based system with BM25 as a base retriever and experiment with recent state-of-the-art transformer-based models for re-ranking. Furthermore, we use a BART-based textual entailment architecture to authenticate the retrieved documents in the later step. We report experimental findings, demonstrating that our retrieval module outperforms the best baseline system by 10.32 NDCG@100 points. We escort a demonstration to assess the efficacy and impact of our suggested solution. As a byproduct of this study, we present an open-source, easily deployable, and user-friendly Python API that the community can adopt. 


Document Retrieval and Claim Verification to Mitigate COVID-19 Misinformation

Analogies play a central role in human commonsense reasoning. The ability to recognize analogies such as “eye is to seeing what ear is to hearing”, sometimes referred to as analogical proportions, shape how we structure knowledge and understand language. Surprisingly, however, the task of identifying such analogies has not yet received much attention in the language model era. In this paper, we analyze the capabilities of transformer-based language models on this unsupervised task, using benchmarks obtained from educational settings, as well as more commonly used datasets. We find that off-the-shelf language models can identify analogies to a certain extent, but struggle with abstract and complex relations, and results are highly sensitive to model architecture and hyperparameters. Overall the best results were obtained with GPT-2 and RoBERTa, while configurations using BERT were not able to outperform word embedding models. Our results raise important questions for future work about how, and to what extent, pre-trained language models capture knowledge about abstract semantic relations.

CIKQA: Learning Commonsense Inference with a Unified Knowledge-in-the-loop QA Paradigm

re-trained language models have been found to capture a surprisingly rich amount of lexical knowledge, ranging from commonsense properties of everyday concepts to detailed factual knowledge about named entities. Among others, this makes it possible to distill high-quality word vectors from pre-trained language models. However, it is currently unclear to what extent it is possible to distill relation embeddings, i.e. vectors that characterize the relationship between two words. Such relation embeddings are appealing because they can, in principle, encode relational knowledge in a more fine-grained way than is possible with knowledge graphs. To obtain relation embeddings from a pre-trained language model, we encode word pairs using a (manually or automatically generated) prompt, and we fine-tune the language model such that relationally similar word pairs yield similar output vectors. We find that the resulting relation embeddings are highly competitive on analogy (unsupervised) and relation classification (supervised) benchmarks, even without any task-specific fine-tuning. Source code to reproduce our experimental results and the model checkpoints are available in the following repository: https://github.com/asahi417/relbert

Distilling Relation Embeddings from Pre-trained Language Models

Pre-trained models (PTMs) have lead to great improvements in natural language generation (NLG). However, it is still unclear how much commonsense knowledge they possess. With the goal of evaluating commonsense knowledge of NLG models, recent work has proposed the problem of generative commonsense reasoning, e.g., to compose a logical sentence given a set of unordered concepts. Existing approaches to this problem hypothesize that PTMs lack sufficient parametric knowledge for this task, which can be overcome by introducing external knowledge or task-specific pre-training objectives. Different from this trend, we argue that PTM's \textit{inherent} ability for generative commonsense reasoning is underestimated due to the order-agnostic property of its input. In particular, we hypothesize that the order of the input concepts can affect the PTM's ability to utilize its commonsense knowledge. To this end, we propose a pre-ordering approach to elaborately manipulate the order of the given concepts before generation. Experiments show that our approach can outperform the more sophisticated models that have access to a lot of external data and resources.

Revisiting Generative Commonsense Reasoning: A Pre-Ordering Approach

Pre-trained language models have shown successful progress in many text understanding benchmarks. This work explores the capability of these models to predict actionable plans in real-world environments. Given a text instruction, we show that language priors encoded in pre-trained models allow us to infer fine-grained subgoal sequences. In contrast to recent methods which make strong assumptions about subgoal supervision, our experiments show that language models can infer detailed subgoal sequences from few training sequences without any fine-tuning. We further propose a simple strategy to re-rank language model predictions based on interaction and feedback from the environment.Combined with pre-trained navigation and visual reasoning components, our approach demonstrates competitive performance on subgoal prediction and task completion in the ALFRED benchmark compared to prior methods that assume more subgoal supervision.

Few-shot Subgoal Planning with Language Models

Commonsense is defined as the knowledge that is shared by everyone. However, certain types of commonsense knowledge are correlated with culture and geographic locations and they are only shared locally. For example, the scenarios of wedding ceremonies vary across regions due to different customs influenced by historical and religious factors. Such regional characteristics, however, are generally omitted in prior work. In this paper, we construct a Geo-Diverse Visual Commonsense Reasoning dataset (GD-VCR) to test vision-and-language models' ability to understand cultural and geo-location-specific commonsense. In particular, we study two state-of-the-art Vision-and-Language models, VisualBERT and ViLBERT trained on VCR, a standard multimodal commonsense benchmark with images primarily from Western regions. We then evaluate how well the trained models can generalize to answering the questions in GD-VCR. We find that the performance of both models for non-Western regions including East Asia, South Asia, and Africa is significantly lower than that for Western region. We analyze the reasons behind the performance disparity and find that the performance gap is larger on QA pairs that: 1) are concerned with culture-related scenarios, e.g., weddings, religious activities, and festivals; 2) require high-level geo-diverse commonsense reasoning rather than low-order perception and recognition.

Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning

Specialized number representations in NLP have shown improvements on numerical reasoning tasks like arithmetic word problems and masked number prediction. But humans also use numeracy to make better sense of world concepts, e.g., you can seat 5 people in your ‘room’ but not 500. Does a better grasp of numbers improve a model’s understanding of other concepts and words? This paper studies the effect of using six different number encoders on the task of masked word prediction (MWP), as a proxy for evaluating literacy. To support this investigation, we develop Wiki-Convert, a 900,000 sentence dataset annotated with numbers and units, to avoid conflating nominal and ordinal number occurrences. We find a significant improvement in MWP for sentences containing numbers, that exponent embeddings are the best number encoders, yielding over 2 points jump in prediction accuracy over a BERT baseline, and that these enhanced literacy skills also generalize to contexts without annotated numbers. We release all code at https://git.io/JuZXn.

Numeracy enhances the Literacy of Language Models

Robots operating in everyday environments need to effectively perceive, model, and infer semantic properties of objects. Existing knowledge reasoning frameworks only model binary relations between an object's class label and its semantic properties, unable to collectively reason about object properties detected by different perception algorithms and grounded in diverse sensory modalities. We bridge the gap between multimodal perception and knowledge reasoning by introducing an n-ary representation that models complex, inter-related object properties. To tackle the problem of collecting n-ary semantic knowledge at scale, we propose a transformer neural network that directly generalizes knowledge from observations of object instances. The learned model can reason at different levels of abstraction, effectively predicting unknown properties of objects in different environmental contexts given different amounts of observed information. We quantitatively validate our approach against five prior methods on LINK, a unique dataset we contribute that contains 1457 situated object instances with 15 multimodal properties types and 200 total properties. Additionally, we apply our work to a mobile manipulation robot, demonstrating its ability to leverage n-ary reasoning to retrieve objects and actively detect object properties.

Detecting the Role of an Entity in Harmful Memes: Techniques and their Limitations

Downloads

Next from ACL 2022

Fine-tuning and Sampling Strategies for Multimodal Role Labeling of Entities under Class Imbalance

Similar lecture

A Comparative Analysis for Troll-Based Meme Classification

Detecting the Role of an Entity in Harmful Memes: Techniques and their Limitations

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from ACL 2022

Fine-tuning and Sampling Strategies for Multimodal Role Labeling of Entities under Class Imbalance

Similar lecture

A Comparative Analysis for Troll-Based Meme Classification

Downloads