Intent-conditioned and Non-toxic Counterspeech Generation using Multi-Task Instruction Tuning with RLAIF

Counterspeech, defined as a response to mitigate online hate speech, is increasingly used as a non-censorial solution. Addressing hate speech effectively involves dispelling the stereotypes, prejudices, and biases often subtly implied in brief, single-sentence statements or abuses. These implicit expressions challenge language models, especially in seq2seq tasks, as model performance typically excels with longer contexts. Our study introduces CoARL, a novel framework enhancing counterspeech generation by modeling the pragmatic implications underlying social biases in hateful statements. CoARL's first two phases involve sequential multi-instruction tuning, teaching the model to understand intents, reactions, and harms of offensive statements, and then learning task-specific low-rank adapter weights for generating intent-conditioned counterspeech. The final phase uses reinforcement learning to fine-tune outputs for effectiveness and non-toxicity. CoARL outperforms existing benchmarks in intent-conditioned counterspeech generation, showing an average improvement of 3 points in intent-conformity and 4 points in argument-quality metrics. Extensive human evaluation supports CoARL's efficacy in generating superior and more context-appropriate responses compared to existing systems, including prominent LLMs like ChatGPT.

Counterspeeches up my sleeve! Intent Distribution Learning and Persistent Fusion for Intent-Conditioned Counterspeech Generation

Counterspeech has been demonstrated to be an efficacious approach for combating hate speech. While various conventional and controlled approaches have been studied in recent years to generate counterspeech, a counterspeech with a certain intent may not be sufficient in every scenario. Due to the complex and multifaceted nature of hate speech, utilizing multiple forms of counter-narratives with varying intents may be advantageous in different circumstances. In this paper, we explore intent-conditioned counterspeech generation. At first, we develop IntentCONAN, a diversified intent-specific counterspeech dataset with 6831 counterspeeches conditioned on five intents, i.e., informative, denouncing, question, positive, and humour. Subsequently, we propose QUARC, a two-stage framework for intent-conditioned counterspeech generation. QUARC leverages vector-quantized representations learned for each intent category along with PerFuMe, a novel fusion module to incorporate intent-specific information into the model. Our evaluation demonstrates that QUARC outperforms several baselines by an average of 10% across evaluation metrics. An extensive human evaluation supplements our hypothesis of better and more appropriate responses than comparative systems.

LAHM : Large Annotated Dataset for Multi-Domain and Multilingual Hate Speech Identification

Current research on hate speech analysis is typically oriented towards monolingual and single classification tasks. In this paper, we present a new multilingual hate speech analysis dataset for English, Hindi, Arabic, French, German and Spanish languages for multiple domains across hate speech - Abuse, Racism, Sexism, Religious Hate and Extremism. To the best of our knowledge, this paper is the first to address the problem of identifying various types of hate speech in these five wide domains in these six languages. In this work, we describe how we created the dataset, created annotations at high level and low level for different domains and how we use it to test the current state-of-the-art multilingual and multitask learning approaches. We evaluate our dataset in various monolingual, cross-lingual and machine translation classification settings and compare it against open source English datasets that we aggregated and merged for this task. Then we discuss how this approach can be used to create large scale hate-speech datasets and how to leverage our annotations in order to improve hate speech detection and classification in general.

Hatemongers ride on echo chambers to escalate hate speech diffusion

Recent years have witnessed a swelling rise of hateful and abusive content over online social networks. While detection and moderation of hate speech have been the early go-to countermeasures, the solution requires a deeper exploration of the dynamics of hate generation and propagation. We analyze more than 32 million posts from over 6.8 million users across three popular online social networks to investigate the interrelations between hateful behavior, information dissemination, and polarised organization mediated by echo chambers. We find that hatemongers play a more crucial role in governing the spread of information compared to singled-out hateful content. This observation holds for both the growth of information cascades as well as the conglomeration of hateful actors. Dissection of the core-wise distribution of these networks points towards the fact that hateful users acquire a more well-connected position in the social network and often flock together to build up information cascades. We observe that this cohesion is far from mere organized behavior; instead, in these networks, hatemongers dominate the echo chambers -- groups of users actively align themselves to specific ideological positions. The observed dominance of hateful users to inflate information cascades is primarily via user interactions amplified within these echo chambers. We conclude our study with a cautionary note that popularity-based recommendation of content is susceptible to be exploited by hatemongers given their potential to escalate content popularity via echo-chambered interactions.

Logically at Factify 2023: A Multi-Modal Fact Checking System Based on Evidence Retrieval techniques and Transformer Encoder Architecture

In this paper, we present the Logically submissions to De-Factify 2 challenge (DE-FACTIFY 2023) on the task 1 of Multi-Modal Fact Checking. We describes our submissions to this challenge including explored evidence retrieval and selection techniques, pre-trained cross-modal and unimodal models, and a cross-modal veracity model based on the well established Transformer Encoder (TE) architecture which is heavily relies on the concept of self-attention. Exploratory analysis is also conducted on this Factify 2 data set that uncovers the salient multi-modal patterns and hypothesis motivating the architecture proposed in this work. A series of preliminary experiments were done to investigate and benchmarking different pre-trained embedding models, evidence retrieval settings and thresholds. The final system, a standard two-stage evidence based veracity detection system, yields weighted avg. 0.79 on both val set and final blind test set on the task 1, which achieves 3rd place with a small margin to the top performing system on the leaderboard among 9 participants.

Logically at the Factify 2022: Multimodal Fact Verification

This paper describes our participant system for the multi-modal fact verification (Factify) challenge at AAAI 2022. Despite the recent advance in text based verification techniques and large pre-trained multimodal models cross vision and language, very limited work has been done in applying multimodal techniques to automate fact checking process, particularly considering the increasing prevalence of claims and fake news about images and videos on social media. In our work, the challenge is treated as multimodal entailment task and framed as multi-class classification. Two baseline approaches are proposed and explored including an ensemble model (combining two uni-modal models) and a multi-modal attention network (modeling the interaction between image and text pair from claim and evidence document). We conduct several experiments investigating and benchmarking different SoTA pre-trained transformers and vision models in this work. Our best model is ranked first in leaderboard which obtains a weighted average F-measure of 0.77 on both validation and test set. Exploratory analysis of dataset is also carried out on the Factify data set and uncovers salient patterns and issues (e.g., word overlapping, visual entailment correlation, source bias) that motivates our hypothesis. Finally, we highlight challenges of the task and multimodal dataset for future research.