Md Shad Akhtar

Shivani Kumar, Atharva Kulkarni, Md Shad Akhtar et al. · cmu

Indirect speech such as sarcasm achieves a constellation of discourse goals in human communication. While the indirectness of figurative language warrants speakers to achieve certain pragmatic goals, it is challenging for AI agents to comprehend such idiosyncrasies of human communication. Though sarcasm identification has been a well-explored topic in dialogue analysis, for conversational systems to truly grasp a conversation's innate meaning and generate appropriate responses, simply detecting sarcasm is not enough; it is vital to explain its underlying sarcastic connotation to capture its true essence. In this work, we study the discourse structure of sarcastic conversations and propose a novel task - Sarcasm Explanation in Dialogue (SED). Set in a multimodal and code-mixed setting, the task aims to generate natural language explanations of satirical conversations. To this end, we curate WITS, a new dataset to support our task. We propose MAF (Modality Aware Fusion), a multimodal context-aware attention and global information fusion module to capture multimodality and use it to benchmark WITS. The proposed attention module surpasses the traditional multimodal fusion baselines and reports the best performance on almost all metrics. Lastly, we carry out detailed analyses both quantitatively and qualitatively.

Megha Sundriyal, Atharva Kulkarni, Vaibhav Pulastya et al. · cmu

The widespread diffusion of medical and political claims in the wake of COVID-19 has led to a voluminous rise in misinformation and fake news. The current vogue is to employ manual fact-checkers to efficiently classify and verify such data to combat this avalanche of claim-ridden misinformation. However, the rate of information dissemination is such that it vastly outpaces the fact-checkers' strength. Therefore, to aid manual fact-checkers in eliminating the superfluous content, it becomes imperative to automatically identify and extract the snippets of claim-worthy (mis)information present in a post. In this work, we introduce the novel task of Claim Span Identification (CSI). We propose CURT, a large-scale Twitter corpus with token-level claim spans on more than 7.5k tweets. Furthermore, along with the standard token classification baselines, we benchmark our dataset with DABERTa, an adapter-based variation of RoBERTa. The experimental results attest that DABERTa outperforms the baseline systems across several evaluation metrics, improving by about 1.5 points. We also report detailed error analysis to validate the model's performance along with the ablation studies. Lastly, we release our comprehensive span annotation guidelines for public use.

Sarah Masud, Manjot Bedi, Mohammad Aflah Khan et al.

Curbing online hate speech has become the need of the hour; however, a blanket ban on such activities is infeasible for several geopolitical and cultural reasons. To reduce the severity of the problem, in this paper, we introduce a novel task, hate speech normalization, that aims to weaken the intensity of hatred exhibited by an online post. The intention of hate speech normalization is not to support hate but instead to provide the users with a stepping stone towards non-hate while giving online platforms more time to monitor any improvement in the user's behavior. To this end, we manually curated a parallel corpus - hate texts and their normalized counterparts (a normalized text is less hateful and more benign). We introduce NACL, a simple yet efficient hate speech normalization model that operates in three stages - first, it measures the hate intensity of the original sample; second, it identifies the hate span(s) within it; and finally, it reduces hate intensity by paraphrasing the hate spans. We perform extensive experiments to measure the efficacy of NACL via three-way evaluation (intrinsic, extrinsic, and human-study). We observe that NACL outperforms six baselines - NACL yields a score of 0.1365 RMSE for the intensity prediction, 0.622 F1-score in the span identification, and 82.27 BLEU and 80.05 perplexity for the normalized text generation. We further show the generalizability of NACL across other platforms (Reddit, Facebook, Gab). An interactive prototype of NACL was put together for the user study. Further, the tool is being deployed in a real-world setting at Wipro AI as a part of its mission to tackle harmful content on online platforms.

Shivani Kumar, Ishani Mondal, Md Shad Akhtar et al.

Conversations emerge as the primary media for exchanging ideas and conceptions. From the listener's perspective, identifying various affective qualities, such as sarcasm, humour, and emotions, is paramount for comprehending the true connotation of the emitted utterance. However, one of the major hurdles faced in learning these affect dimensions is the presence of figurative language, viz. irony, metaphor, or sarcasm. We hypothesize that any detection system constituting the exhaustive and explicit presentation of the emitted utterance would improve the overall comprehension of the dialogue. To this end, we explore the task of Sarcasm Explanation in Dialogues, which aims to unfold the hidden irony behind sarcastic utterances. We propose MOSES, a deep neural network, which takes a multimodal (sarcastic) dialogue instance as an input and generates a natural language sentence as its explanation. Subsequently, we leverage the generated explanation for various natural language understanding tasks in a conversational dialogue setup, such as sarcasm detection, humour identification, and emotion recognition. Our evaluation shows that MOSES outperforms the state-of-the-art system for SED by an average of ~2% on different evaluation metrics, such as ROUGE, BLEU, and METEOR. Further, we observe that leveraging the generated explanation advances three downstream tasks for affect classification - an average improvement of ~14% F1-score in the sarcasm detection task and ~2% in the humour identification and emotion recognition task. We also perform extensive analyses to assess the quality of the results.

Ayan Sengupta, Tharun Suresh, Md Shad Akhtar et al.

Being a popular mode of text-based communication in multilingual communities, code-mixing in online social media has became an important subject to study. Learning the semantics and morphology of code-mixed language remains a key challenge, due to scarcity of data and unavailability of robust and language-invariant representation learning technique. Any morphologically-rich language can benefit from character, subword, and word-level embeddings, aiding in learning meaningful correlations. In this paper, we explore a hierarchical transformer-based architecture (HIT) to learn the semantics of code-mixed languages. HIT consists of multi-headed self-attention and outer product attention components to simultaneously comprehend the semantic and syntactic structures of code-mixed texts. We evaluate the proposed method across 6 Indian languages (Bengali, Gujarati, Hindi, Tamil, Telugu and Malayalam) and Spanish for 9 NLP tasks on 17 datasets. The HIT model outperforms state-of-the-art code-mixed representation learning and multilingual language models in all tasks. We further demonstrate the generalizability of the HIT architecture using masked language modeling-based pre-training, zero-shot learning, and transfer learning approaches. Our empirical results show that the pre-training objectives significantly improve the performance on downstream tasks.

Shivani Kumar, Shubham Dudeja, Md Shad Akhtar et al.

In a conversational dialogue, speakers may have different emotional states and their dynamics play an important role in understanding dialogue's emotional discourse. However, simply detecting emotions is not sufficient to entirely comprehend the speaker-specific changes in emotion that occur during a conversation. To understand the emotional dynamics of speakers in an efficient manner, it is imperative to identify the rationale or instigator behind any changes or flips in emotion expressed by the speaker. In this paper, we explore the task called Instigator based Emotion Flip Reasoning (EFR), which aims to identify the instigator behind a speaker's emotion flip within a conversation. For example, an emotion flip from joy to anger could be caused by an instigator like threat. To facilitate this task, we present MELD-I, a dataset that includes ground-truth EFR instigator labels, which are in line with emotional psychology. To evaluate the dataset, we propose a novel neural architecture called TGIF, which leverages Transformer encoders and stacked GRUs to capture the dialogue context, speaker dynamics, and emotion sequence in a conversation. Our evaluation demonstrates state-of-the-art performance (+4-12% increase in F1-score) against five baselines used for the task. Further, we establish the generalizability of TGIF on an unseen dataset in a zero-shot setting. Additionally, we provide a detailed analysis of the competing models, highlighting the advantages and limitations of our neural architecture.

Megha Sundriyal, Harshit Choudhary, Tanmoy Chakraborty et al.

Misinformation spreads rapidly on social media, causing serious damage by influencing public opinion, promoting dangerous behavior, or eroding trust in reliable sources. It spreads too fast for traditional fact-checking, stressing the need for predictive methods. We introduce CROWDSHIELD, a crowd intelligence-based method for early misinformation prediction. We hypothesize that the crowd's reactions to misinformation reveal its accuracy. Furthermore, we hinge upon exaggerated assertions/claims and replies with particular positions/stances on the source post within a conversation thread. We employ Q-learning to capture the two dimensions -- stances and claims. We utilize deep Q-learning due to its proficiency in navigating complex decision spaces and effectively learning network properties. Additionally, we use a transformer-based encoder to develop a comprehensive understanding of both content and context. This multifaceted approach helps ensure the model pays attention to user interaction and stays anchored in the communication's content. We propose MIST, a manually annotated misinformation detection Twitter corpus comprising nearly 200 conversation threads with more than 14K replies. In experiments, CROWDSHIELD outperformed ten baseline systems, achieving an improvement of ~4% macro-F1 score. We conduct an ablation study and error analysis to validate our proposed model's performance. The source code and dataset are available at https://github.com/LCS2-IIITD/CrowdShield.git.

Ayan Sengupta, Md Shad Akhtar, Tanmoy Chakraborty

Code-mixing and script-mixing are prevalent across online social networks and multilingual societies. However, a user's preference toward code-mixing depends on the socioeconomic status, demographics of the user, and the local context, which existing generative models mostly ignore while generating code-mixed texts. In this work, we make a pioneering attempt to develop a persona-aware generative model to generate texts resembling real-life code-mixed texts of individuals. We propose a Persona-aware Generative Model for Code-mixed Generation, PARADOX, a novel Transformer-based encoder-decoder model that encodes an utterance conditioned on a user's persona and generates code-mixed texts without monolingual reference data. We propose an alignment module that re-calibrates the generated sequence to resemble real-life code-mixed texts. PARADOX generates code-mixed texts that are semantically more meaningful and linguistically more valid. To evaluate the personification capabilities of PARADOX, we propose four new metrics -- CM BLEU, CM Rouge-1, CM Rouge-L and CM KS. On average, PARADOX achieves 1.6 points better CM BLEU, 47% better perplexity and 32% better semantic coherence than the non-persona-based counterparts.

Megha Sundriyal, Md Shad Akhtar, Tanmoy Chakraborty

The expansion of online social media platforms has led to a surge in online content consumption. However, this has also paved the way for disseminating false claims and misinformation. As a result, there is an escalating demand for a substantial workforce to sift through and validate such unverified claims. Currently, these claims are manually verified by fact-checkers. Still, the volume of online content often outweighs their potency, making it difficult for them to validate every single claim in a timely manner. Thus, it is critical to determine which assertions are worth fact-checking and prioritize claims that require immediate attention. Multiple factors contribute to determining whether a claim necessitates fact-checking, encompassing factors such as its factual correctness, potential impact on the public, the probability of inciting hatred, and more. Despite several efforts to address claim check-worthiness, a systematic approach to identify these factors remains an open challenge. To this end, we introduce a new task of fine-grained claim check-worthiness, which underpins all of these factors and provides probable human grounds for identifying a claim as check-worthy. We present CheckIt, a manually annotated large Twitter dataset for fine-grained claim check-worthiness. We benchmark our dataset against a unified approach, CheckMate, that jointly determines whether a claim is check-worthy and the factors that led to that conclusion. We compare our suggested system with several baseline systems. Finally, we report a thorough analysis of results and human assessment, validating the efficacy of integrating check-worthiness factors in detecting claims worth fact-checking.

Shivani Kumar, Rishabh Gupta, Md Shad Akhtar et al.

In conversational settings, individuals exhibit unique behaviors, rendering a one-size-fits-all approach insufficient for generating responses by dialogue agents. Although past studies have aimed to create personalized dialogue agents using speaker persona information, they have relied on the assumption that the speaker's persona is already provided. However, this assumption is not always valid, especially when it comes to chatbots utilized in industries like banking, hotel reservations, and airline bookings. This research paper aims to fill this gap by exploring the task of Speaker Profiling in Conversations (SPC). The primary objective of SPC is to produce a summary of persona characteristics for each individual speaker present in a dialogue. To accomplish this, we have divided the task into three subtasks: persona discovery, persona-type identification, and persona-value extraction. Given a dialogue, the first subtask aims to identify all utterances that contain persona information. Subsequently, the second task evaluates these utterances to identify the type of persona information they contain, while the third subtask identifies the specific persona values for each identified type. To address the task of SPC, we have curated a new dataset named SPICE, which comes with specific labels. We have evaluated various baselines on this dataset and benchmarked it with a new neural model, SPOT, which we introduce in this paper. Furthermore, we present a comprehensive analysis of SPOT, examining the limitations of individual modules both quantitatively and qualitatively.

Shivani Kumar, Ramaneswaran S, Md Shad Akhtar et al.

Understanding emotions during conversation is a fundamental aspect of human communication, driving NLP research for Emotion Recognition in Conversation (ERC). While considerable research has focused on discerning emotions of individual speakers in monolingual dialogues, understanding the emotional dynamics in code-mixed conversations has received relatively less attention. This motivates our undertaking of ERC for code-mixed conversations in this study. Recognizing that emotional intelligence encompasses a comprehension of worldly knowledge, we propose an innovative approach that integrates commonsense information with dialogue context to facilitate a deeper understanding of emotions. To achieve this, we devise an efficient pipeline that extracts relevant commonsense from existing knowledge graphs based on the code-mixed input. Subsequently, we develop an advanced fusion technique that seamlessly combines the acquired commonsense information with the dialogue representation obtained from a dedicated dialogue understanding module. Our comprehensive experimentation showcases the substantial performance improvement obtained through the systematic incorporation of commonsense in ERC. Both quantitative assessments and qualitative analyses further corroborate the validity of our hypothesis, reaffirming the pivotal role of commonsense integration in enhancing ERC.

Aseem Srivastava, Smriti Joshi, Tanmoy Chakraborty et al.

In mental health counseling, condensing dialogues into concise and relevant summaries (aka counseling notes) holds pivotal significance. Large Language Models (LLMs) exhibit remarkable capabilities in various generative tasks; however, their adaptation to domain-specific intricacies remains challenging, especially within mental health contexts. Unlike standard LLMs, mental health experts first plan to apply domain knowledge in writing summaries. Our work enhances LLMs' ability by introducing a novel planning engine to orchestrate structuring knowledge alignment. To achieve high-order planning, we divide knowledge encapsulation into two major phases: (i) holding dialogue structure and (ii) incorporating domain-specific knowledge. We employ a planning engine on Llama-2, resulting in a novel framework, PIECE. Our proposed system employs knowledge filtering-cum-scaffolding to encapsulate domain knowledge. Additionally, PIECE leverages sheaf convolution learning to enhance its understanding of the dialogue's structural nuances. We compare PIECE with 14 baseline methods and observe a significant improvement across ROUGE and Bleurt scores. Further, expert evaluation and analyses validate the generation quality to be effective, sometimes even surpassing the gold standard. We further benchmark PIECE with other LLMs and report improvement, including Llama-2 (+2.72%), Mistral (+2.04%), and Zephyr (+1.59%), to justify the generalizability of the planning engine.

Shivani Kumar, Md Shad Akhtar, Erik Cambria et al.

We present SemEval-2024 Task 10, a shared task centred on identifying emotions and finding the rationale behind their flips within monolingual English and Hindi-English code-mixed dialogues. This task comprises three distinct subtasks - emotion recognition in conversation for code-mixed dialogues, emotion flip reasoning for code-mixed dialogues, and emotion flip reasoning for English dialogues. Participating systems were tasked to automatically execute one or more of these subtasks. The datasets for these tasks comprise manually annotated conversations focusing on emotions and triggers for emotion shifts (The task data is available at https://github.com/LCS2-IIITD/EDiReF-SemEval2024.git). A total of 84 participants engaged in this task, with the most adept systems attaining F1-scores of 0.70, 0.79, and 0.76 for the respective subtasks. This paper summarises the results and findings from 24 teams alongside their system descriptions.

Megha Sundriyal, Md Shad Akhtar, Tanmoy Chakraborty

A significant increase in content creation and information exchange has been made possible by the quick development of online social media platforms, which has been very advantageous. However, these platforms have also become a haven for those who disseminate false information, propaganda, and fake news. Claims are essential in forming our perceptions of the world, but sadly, they are frequently used to trick people by those who spread false information. To address this problem, social media giants employ content moderators to filter out fake news from the actual world. However, the sheer volume of information makes it difficult to identify fake news effectively. Therefore, it has become crucial to automatically identify social media posts that make such claims, check their veracity, and differentiate between credible and false claims. In response, we presented CLAIMSCAN in the 2023 Forum for Information Retrieval Evaluation (FIRE'2023). The primary objectives centered on two crucial tasks: Task A, determining whether a social media post constitutes a claim, and Task B, precisely identifying the words or phrases within the post that form the claim. Task A received 40 registrations, demonstrating a strong interest and engagement in this timely challenge. Meanwhile, Task B attracted participation from 28 teams, highlighting its significance in the digital era of misinformation.

Ayan Sengupta, Shantanu Dixit, Md Shad Akhtar et al.

Knowledge distillation (KD) methods are pivotal in compressing large pre-trained language models into smaller models, ensuring computational efficiency without significantly dropping performance. Traditional KD techniques assume homogeneity in modalities between the teacher (source) and the student (target) models. On the other hand, existing multimodal knowledge distillation methods require modality-specific pre-training of the teacher model, which is computationally infeasible in most cases. In this paper, we introduce ARMADA, an efficient cross-modal knowledge distillation framework designed to transfer knowledge from large vision-language models, including black-box models, to language-only models. Unlike existing KD techniques that rely on the internal structures of multimodal teachers or require computationally expensive pre-training, ARMADA leverages novel alignment techniques to distil knowledge without altering the teacher model, ensuring efficiency and scalability. We empirically validate ARMADA on twelve natural language understanding, eight complex generative reasoning and five instruction-tuning tasks, demonstrating consistent performance improvements in large models such as DeBERTa-v2-1.4B, OPT-1.3B, LLaMA-{3B, 7B, 8B}. ARMADA achieves up to 3.4% improvement on language understanding tasks and 2.6% boost in generative reasoning, all without requiring expensive multimodal pre-training or fine-tuning of the teacher model. Our findings challenge conventional knowledge distillation paradigms by demonstrating that even vision-language models, despite lacking direct textual understanding, can significantly enhance language models when distilled appropriately.

Ayan Sengupta, Md Shad Akhtar, Tanmoy Chakraborty

Multi-head self-attention-based Transformers have shown promise in different learning tasks. Albeit these models exhibit significant improvement in understanding short-term and long-term contexts from sequences, encoders of Transformers and their variants fail to preserve layer-wise contextual information. Transformers usually project tokens onto sparse manifolds and fail to preserve mathematical equivalence among the token representations. In this work, we propose TransJect, an encoder model that guarantees a theoretical bound for layer-wise distance preservation between a pair of tokens. We propose a simple alternative to dot-product attention to ensure Lipschitz continuity. This allows TransJect to learn injective mappings to transform token representations to different manifolds with similar topology and preserve Euclidean distance between every pair of tokens in subsequent layers. Evaluations across multiple benchmark short- and long-sequence classification tasks show maximum improvements of 6.8% and 5.9%, respectively, over the variants of Transformers. Additionally, TransJect displays 79% better performance than Transformer on the language modeling task. We further highlight the shortcomings of multi-head self-attention from the statistical physics viewpoint. Although multi-head self-attention was incepted to learn different abstraction levels within the networks, our empirical analyses suggest that different attention heads learn randomly and unorderly. In contrast, TransJect adapts a mixture of experts for regularization; these experts are more orderly and balanced and learn different sparse representations from the input sequences. TransJect exhibits very low entropy and can be efficiently scaled to larger depths.

Parth Patwa, Shivam Sharma, Srinivas Pykl et al.

Along with COVID-19 pandemic we are also fighting an `infodemic'. Fake news and rumors are rampant on social media. Believing in rumors can cause significant harm. This is further exacerbated at the time of a pandemic. To tackle this, we curate and release a manually annotated dataset of 10,700 social media posts and articles of real and fake news on COVID-19. We benchmark the annotated dataset with four machine learning baselines - Decision Tree, Logistic Regression, Gradient Boost, and Support Vector Machine (SVM). We obtain the best performance of 93.46% F1-score with SVM. The data and code is available at: https://github.com/parthpatwa/covid19-fake-news-dectection

Rohan Charudatt Salvi, Chirag Chawla, Dhruv Jain et al.

Automatic medical text simplification plays a key role in improving health literacy by making complex biomedical research accessible to diverse readers. However, most existing resources assume a single generic audience, overlooking the wide variation in medical literacy and information needs across user groups. To address this limitation, we introduce PERCS (Persona-guided Controllable Summarization), a dataset of biomedical abstracts paired with summaries tailored to four personas: Laypersons, Premedical Students, Non-medical Researchers, and Medical Experts. These personas represent different levels of medical literacy and information needs, emphasizing the need for targeted, audience-specific summarization. Each summary in PERCS was reviewed by physicians for factual accuracy and persona alignment using a detailed error taxonomy. Technical validation shows clear differences in readability, vocabulary, and content depth across personas. Along with describing the dataset, we benchmark four large language models on PERCS using automatic evaluation metrics that assess comprehensiveness, readability, and faithfulness, establishing baseline results for future research. The dataset, annotation guidelines, and evaluation materials are publicly available to support research on persona-specific communication and controllable biomedical summarization.

Siddhant Agarwal, Adya Dhuler, Polly Ruhnke et al.

Over the past years, memes have evolved from being exclusively a medium of humorous exchanges to one that allows users to express a range of emotions freely and easily. With the ever-growing utilization of memes in expressing depressive sentiments, we conduct a study on identifying depressive symptoms exhibited by memes shared by users of online social media platforms. We introduce RESTOREx as a vital resource for detecting depressive symptoms in memes on social media through the Large Language Model (LLM) generated and human-annotated explanations. We introduce MAMAMemeia, a collaborative multi-agent multi-aspect discussion framework grounded in the clinical psychology method of Cognitive Analytic Therapy (CAT) Competencies. MAMAMemeia improves upon the current state-of-the-art by 7.55% in macro-F1 and is established as the new benchmark compared to over 30 methods.

Amey Hengle, Aswini Kumar, Sahajpreet Singh et al.

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.

Kartik Kartik, Sanjana Soni, Anoop Kunchukuttan et al. · amazon-science, meta-ai

The widespread online communication in a modern multilingual world has provided opportunities to blend more than one language (aka code-mixed language) in a single utterance. This has resulted a formidable challenge for the computational models due to the scarcity of annotated data and presence of noise. A potential solution to mitigate the data scarcity problem in low-resource setup is to leverage existing data in resource-rich language through translation. In this paper, we tackle the problem of code-mixed (Hinglish and Bengalish) to English machine translation. First, we synthetically develop HINMIX, a parallel corpus of Hinglish to English, with ~4.2M sentence pairs. Subsequently, we propose RCMT, a robust perturbation based joint-training model that learns to handle noise in the real-world code-mixed text by parameter sharing across clean and noisy words. Further, we show the adaptability of RCMT in a zero-shot setup for Bengalish to English translation. Our evaluation and comprehensive analyses qualitatively and quantitatively demonstrate the superiority of RCMT over state-of-the-art code-mixed and robust translation methods.

Abdullah Mazhar, Zuhair hasan shaik, Aseem Srivastava et al. · microsoft-research

The expression of mental health symptoms through non-traditional means, such as memes, has gained remarkable attention over the past few years, with users often highlighting their mental health struggles through figurative intricacies within memes. While humans rely on commonsense knowledge to interpret these complex expressions, current Multimodal Language Models (MLMs) struggle to capture these figurative aspects inherent in memes. To address this gap, we introduce a novel dataset, AxiOM, derived from the GAD anxiety questionnaire, which categorizes memes into six fine-grained anxiety symptoms. Next, we propose a commonsense and domain-enriched framework, M3H, to enhance MLMs' ability to interpret figurative language and commonsense knowledge. The overarching goal remains to first understand and then classify the mental health symptoms expressed in memes. We benchmark M3H against 6 competitive baselines (with 20 variations), demonstrating improvements in both quantitative and qualitative metrics, including a detailed human evaluation. We observe a clear improvement of 4.20% and 4.66% on weighted-F1 metric. To assess the generalizability, we perform extensive experiments on a public dataset, RESTORE, for depressive symptom identification, presenting an extensive ablation study that highlights the contribution of each module in both datasets. Our findings reveal limitations in existing models and the advantage of employing commonsense to enhance figurative understanding.

Aseem Srivastava, Zuhair Hasan Shaik, Tanmoy Chakraborty et al.

In mental health counseling, a variety of earlier studies have focused on dialogue modeling. However, most of these studies give limited to no emphasis on the quality of interaction between a patient and a therapist. The therapeutic bond between a patient and a therapist directly correlates with effective mental health counseling. It involves developing the patient's trust on the therapist over the course of counseling. To assess the therapeutic bond in counseling, we introduce trust as a therapist-assistive metric. Our definition of trust involves patients' willingness and openness to express themselves and, consequently, receive better care. We conceptualize it as a dynamic trajectory observable through textual interactions during the counseling. To facilitate trust modeling, we present MENTAL-TRUST, a novel counseling dataset comprising manual annotation of 212 counseling sessions with first-of-its-kind seven expert-verified ordinal trust levels. We project our problem statement as an ordinal classification task for trust quantification and propose a new benchmark, TrustBench, comprising a suite of classical and state-of-the-art language models on MENTAL-TRUST. We evaluate the performance across a suite of metrics and lay out an exhaustive set of findings. Our study aims to unfold how trust evolves in therapeutic interactions.

Abdullah Mazhar, Het Riteshkumar Shah, Aseem Srivastava et al.

The increasing use of large language models in mental health applications calls for principled evaluation frameworks that assess alignment with psychotherapeutic best practices beyond surface-level fluency. While recent systems exhibit conversational competence, they lack structured mechanisms to evaluate adherence to core therapeutic principles. In this paper, we study the problem of evaluating AI-generated therapist-like responses for clinically grounded appropriateness and effectiveness. We assess each therapists utterance along six therapeutic principles: non-judgmental acceptance, warmth, respect for autonomy, active listening, reflective understanding, and situational appropriateness using a fine-grained ordinal scale. We introduce FAITH-M, a benchmark annotated with expert-assigned ordinal ratings, and propose CARE, a multi-stage evaluation framework that integrates intra-dialogue context, contrastive exemplar retrieval, and knowledge-distilled chain-of-thought reasoning. Experiments show that CARE achieves an F-1 score of 63.34 versus the strong baseline Qwen3 F-1 score of 38.56 which is a 64.26 improvement, which also serves as its backbone, indicating that gains arise from structured reasoning and contextual modeling rather than backbone capacity alone. Expert assessment and external dataset evaluations further demonstrate robustness under domain shift, while highlighting challenges in modelling implicit clinical nuance. Overall, CARE provides a clinically grounded framework for evaluating therapeutic fidelity in AI mental health systems.

Akshara Nair, Zeba Afroz, Md Shad Akhtar

Incongruence detection in eyewitness narratives is critical for understanding the reliability of testimonies, yet traditional approaches often fail to address the nuanced inconsistencies inherent in such accounts. In this paper, we introduce a novel task of incongruence detection in eyewitness testimonies. Given a pair of testimonies containing of multiple pairs of question and answer by two subjects, we identify contextually related incongruence between the two subjects. We also mark the span of incongruences in the utterances. To achieve this, we developed MIND(MultI-EyewitNess Deception) - a comprehensive dataset consisting of 2927 pairs of contextually related answers designed to capture both explicit and implicit contradictions. INstruction - TunEd iNcongruity Detection framework based on 6W and multi-hop reasoning approach, aka. INTEND. Drawing from investigative techniques, INTEND address the task as a close-style problem, contradicting on the who, what, when, where and why aspect of the content. Our findings shows that prompt tuning, especially when utilizing our framework, enhances the detection of incongruences by a margin of +5.63 percent. We compare our approach with multiple fine-tuning and prompt tuning techniques on MLMs and LLMs. Emperical results demonstrate convincing performance improvement in F1-score over fine-tuned and regular prompt-tuning techniques, highlighting the effectiveness of our approach.

Joykirat Singh, Sehban Fazili, Rohan Jain et al.

Privacy policy documents have a crucial role in educating individuals about the collection, usage, and protection of users' personal data by organizations. However, they are notorious for their lengthy, complex, and convoluted language especially involving privacy-related entities. Hence, they pose a significant challenge to users who attempt to comprehend organization's data usage policy. In this paper, we propose to enhance the interpretability and readability of policy documents by using controlled abstractive summarization -- we enforce the generated summaries to include critical privacy-related entities (e.g., data and medium) and organization's rationale (e.g.,target and reason) in collecting those entities. To achieve this, we develop PD-Sum, a policy-document summarization dataset with marked privacy-related entity labels. Our proposed model, EROS, identifies critical entities through a span-based entity extraction model and employs them to control the information content of the summaries using proximal policy optimization (PPO). Comparison shows encouraging improvement over various baselines. Furthermore, we furnish qualitative and human evaluations to establish the efficacy of EROS.

Sarah Masud, Mohammad Aflah Khan, Vikram Goyal et al.

Despite the widespread adoption, there is a lack of research into how various critical aspects of pretrained language models (PLMs) affect their performance in hate speech detection. Through five research questions, our findings and recommendations lay the groundwork for empirically investigating different aspects of PLMs' use in hate speech detection. We deep dive into comparing different pretrained models, evaluating their seed robustness, finetuning settings, and the impact of pretraining data collection time. Our analysis reveals early peaks for downstream tasks during pretraining, the limited benefit of employing a more recent pretraining corpus, and the significance of specific layers during finetuning. We further call into question the use of domain-specific models and highlight the need for dynamic datasets for benchmarking hate speech detection.

Zuhair Hasan Shaik, Abdullah Mazhar, Aseem Srivastava et al.

Large Language Models have demonstrated impressive fluency across diverse tasks, yet their tendency to produce toxic content remains a critical challenge for AI safety and public trust. Existing toxicity mitigation approaches primarily manipulate individual neuron activations, but these methods suffer from instability, context dependence, and often compromise the model's core language abilities. To address these shortcomings, we investigate three key questions: the stability of neuron-level toxicity indicators, the advantages of structural (layer-wise) representations, and the interpretability of mechanisms driving toxic generation. Through extensive experiments on Jigsaw and ToxiCN datasets, we show that aggregated layer-wise features provide more robust signals than single neurons. Moreover, we observe conceptual limitations in prior works that conflate toxicity detection experts and generation experts within neuron-based interventions. To mitigate this, we propose a novel principled intervention technique, EigenShift, based on eigen-decomposition of the language model's final output layer. This method selectively targets generation-aligned components, enabling precise toxicity suppression without impairing linguistic competence. Our method requires no additional training or fine-tuning, incurs minimal computational cost, and is grounded in rigorous theoretical analysis.

Bhagesh Gaur, Karan Gupta, Aseem Srivastava et al.

Online Mental Health Communities (OMHCs) provide crucial peer and expert support, yet many posts remain unanswered due to missing support attributes that signal the need for help. We present a novel framework that identifies these gaps and prompts users to enrich their posts, thereby improving engagement. To support this, we introduce REDDME, a new dataset of 4,760 posts from mental health subreddits annotated for the span and intensity of three key support attributes: event what happened?, effect what did the user experience?, and requirement what support they need?. Next, we devise a hierarchical taxonomy, CueTaxo, of support attributes for controlled question generation. Further, we propose MH-COPILOT, a reinforcement learning-based system that integrates (a) contextual attribute-span identification, (b) support attribute intensity classification, (c) controlled question generation via a hierarchical taxonomy, and (d) a verifier for reward modeling. Our model dynamically assesses posts for the presence/absence of support attributes, and generates targeted prompts to elicit missing information. Empirical results across four notable language models demonstrate significant improvements in attribute elicitation and user engagement. A human evaluation further validates the model's effectiveness in real-world OMHC settings.

Palaash Goel, Dushyant Singh Chauhan, Md Shad Akhtar

Sarcasm is a linguistic phenomenon that intends to ridicule a target (e.g., entity, event, or person) in an inherent way. Multimodal Sarcasm Explanation (MuSE) aims at revealing the intended irony in a sarcastic post using a natural language explanation. Though important, existing systems overlooked the significance of the target of sarcasm in generating explanations. In this paper, we propose a Target-aUgmented shaRed fusion-Based sarcasm explanatiOn model, aka. TURBO. We design a novel shared-fusion mechanism to leverage the inter-modality relationships between an image and its caption. TURBO assumes the target of the sarcasm and guides the multimodal shared fusion mechanism in learning intricacies of the intended irony for explanations. We evaluate our proposed TURBO model on the MORE+ dataset. Comparison against multiple baselines and state-of-the-art models signifies the performance improvement of TURBO by an average margin of $+3.3\%$. Moreover, we explore LLMs in zero and one-shot settings for our task and observe that LLM-generated explanation, though remarkable, often fails to capture the critical nuances of the sarcasm. Furthermore, we supplement our study with extensive human evaluation on TURBO's generated explanations and find them out to be comparatively better than other systems.

Neemesh Yadav, Sarah Masud, Vikram Goyal et al.

Employing language models to generate explanations for an incoming implicit hate post is an active area of research. The explanation is intended to make explicit the underlying stereotype and aid content moderators. The training often combines top-k relevant knowledge graph (KG) tuples to provide world knowledge and improve performance on standard metrics. Interestingly, our study presents conflicting evidence for the role of the quality of KG tuples in generating implicit explanations. Consequently, simpler models incorporating external toxicity signals outperform KG-infused models. Compared to the KG-based setup, we observe a comparable performance for SBIC (LatentHatred) datasets with a performance variation of +0.44 (+0.49), +1.83 (-1.56), and -4.59 (+0.77) in BLEU, ROUGE-L, and BERTScore. Further human evaluation and error analysis reveal that our proposed setup produces more precise explanations than zero-shot GPT-3.5, highlighting the intricate nature of the task.

Aseem Srivastava, Tanya Gupta, Alison Cerezo et al.

Online Mental Health Communities (OMHCs), such as Reddit, have witnessed a surge in popularity as go-to platforms for seeking information and support in managing mental health needs. Platforms like Reddit offer immediate interactions with peers, granting users a vital space for seeking mental health assistance. However, the largely unregulated nature of these platforms introduces intricate challenges for both users and society at large. This study explores the factors that drive peer engagement within counseling threads, aiming to enhance our understanding of this critical phenomenon. We introduce BeCOPE, a novel behavior encoded Peer counseling dataset comprising over 10,118 posts and 58,279 comments sourced from 21 mental health-specific subreddits. The dataset is annotated using three major fine-grained behavior labels: (a) intent, (b) criticism, and (c) readability, along with the emotion labels. Our analysis indicates the prominence of ``self-criticism'' as the most prevalent form of criticism expressed by help-seekers, accounting for a significant 43% of interactions. Intriguingly, we observe that individuals who explicitly express their need for help are 18.01% more likely to receive assistance compared to those who present ``surveys'' or engage in ``rants.'' Furthermore, we highlight the pivotal role of well-articulated problem descriptions, showing that superior readability effectively doubles the likelihood of receiving the sought-after support. Our study emphasizes the essential role of OMHCs in offering personalized guidance and unveils behavior-driven engagement patterns.

Poorav Desai, Tanmoy Chakraborty, Md Shad Akhtar

Sarcasm is a pervading linguistic phenomenon and highly challenging to explain due to its subjectivity, lack of context and deeply-felt opinion. In the multimodal setup, sarcasm is conveyed through the incongruity between the text and visual entities. Although recent approaches deal with sarcasm as a classification problem, it is unclear why an online post is identified as sarcastic. Without proper explanation, end users may not be able to perceive the underlying sense of irony. In this paper, we propose a novel problem -- Multimodal Sarcasm Explanation (MuSE) -- given a multimodal sarcastic post containing an image and a caption, we aim to generate a natural language explanation to reveal the intended sarcasm. To this end, we develop MORE, a new dataset with explanation of 3510 sarcastic multimodal posts. Each explanation is a natural language (English) sentence describing the hidden irony. We benchmark MORE by employing a multimodal Transformer-based architecture. It incorporates a cross-modal attention in the Transformer's encoder which attends to the distinguishing features between the two modalities. Subsequently, a BART-based auto-regressive decoder is used as the generator. Empirical results demonstrate convincing results over various baselines (adopted for MuSE) across five evaluation metrics. We also conduct human evaluation on predictions and obtain Fleiss' Kappa score of 0.4 as a fair agreement among 25 evaluators.

Ganeshan Malhotra, Abdul Waheed, Aseem Srivastava et al.

The onset of the COVID-19 pandemic has brought the mental health of people under risk. Social counselling has gained remarkable significance in this environment. Unlike general goal-oriented dialogues, a conversation between a patient and a therapist is considerably implicit, though the objective of the conversation is quite apparent. In such a case, understanding the intent of the patient is imperative in providing effective counselling in therapy sessions, and the same applies to a dialogue system as well. In this work, we take forward a small but an important step in the development of an automated dialogue system for mental-health counselling. We develop a novel dataset, named HOPE, to provide a platform for the dialogue-act classification in counselling conversations. We identify the requirement of such conversation and propose twelve domain-specific dialogue-act (DAC) labels. We collect 12.9K utterances from publicly-available counselling session videos on YouTube, extract their transcripts, clean, and annotate them with DAC labels. Further, we propose SPARTA, a transformer-based architecture with a novel speaker- and time-aware contextual learning for the dialogue-act classification. Our evaluation shows convincing performance over several baselines, achieving state-of-the-art on HOPE. We also supplement our experiments with extensive empirical and qualitative analyses of SPARTA.

Shraman Pramanick, Shivam Sharma, Dimitar Dimitrov et al.

Internet memes have become powerful means to transmit political, psychological, and socio-cultural ideas. Although memes are typically humorous, recent days have witnessed an escalation of harmful memes used for trolling, cyberbullying, and abuse. Detecting such memes is challenging as they can be highly satirical and cryptic. Moreover, while previous work has focused on specific aspects of memes such as hate speech and propaganda, there has been little work on harm in general. Here, we aim to bridge this gap. We focus on two tasks: (i)detecting harmful memes, and (ii)identifying the social entities they target. We further extend a recently released HarMeme dataset, which covered COVID-19, with additional memes and a new topic: US politics. To solve these tasks, we propose MOMENTA (MultimOdal framework for detecting harmful MemEs aNd Their tArgets), a novel multimodal deep neural network that uses global and local perspectives to detect harmful memes. MOMENTA systematically analyzes the local and the global perspective of the input meme (in both modalities) and relates it to the background context. MOMENTA is interpretable and generalizable, and our experiments show that it outperforms several strong rivaling approaches.

Megha Sundriyal, Parantak Singh, Md Shad Akhtar et al.

The formulation of a claim rests at the core of argument mining. To demarcate between a claim and a non-claim is arduous for both humans and machines, owing to latent linguistic variance between the two and the inadequacy of extensive definition-based formalization. Furthermore, the increase in the usage of online social media has resulted in an explosion of unsolicited information on the web presented as informal text. To account for the aforementioned, in this paper, we proposed DESYR. It is a framework that intends on annulling the said issues for informal web-based text by leveraging a combination of hierarchical representation learning (dependency-inspired Poincare embedding), definition-based alignment, and feature projection. We do away with fine-tuning computer-heavy language models in favor of fabricating a more domain-centric but lighter approach. Experimental results indicate that DESYR builds upon the state-of-the-art system across four benchmark claim datasets, most of which were constructed with informal texts. We see an increase of 3 claim-F1 points on the LESA-Twitter dataset, an increase of 1 claim-F1 point and 9 macro-F1 points on the Online Comments(OC) dataset, an increase of 24 claim-F1 points and 17 macro-F1 points on the Web Discourse(WD) dataset, and an increase of 8 claim-F1 points and 5 macro-F1 points on the Micro Texts(MT) dataset. We also perform an extensive analysis of the results. We make a 100-D pre-trained version of our Poincare-variant along with the source code.

Ayan Sengupta, Sourabh Kumar Bhattacharjee, Tanmoy Chakraborty et al.

Understanding linguistics and morphology of resource-scarce code-mixed texts remains a key challenge in text processing. Although word embedding comes in handy to support downstream tasks for low-resource languages, there are plenty of scopes in improving the quality of language representation particularly for code-mixed languages. In this paper, we propose HIT, a robust representation learning method for code-mixed texts. HIT is a hierarchical transformer-based framework that captures the semantic relationship among words and hierarchically learns the sentence-level semantics using a fused attention mechanism. HIT incorporates two attention modules, a multi-headed self-attention and an outer product attention module, and computes their weighted sum to obtain the attention weights. Our evaluation of HIT on one European (Spanish) and five Indic (Hindi, Bengali, Tamil, Telugu, and Malayalam) languages across four NLP tasks on eleven datasets suggests significant performance improvement against various state-of-the-art systems. We further show the adaptability of learned representation across tasks in a transfer learning setup (with and without fine-tuning).

Manjot Bedi, Shivani Kumar, Md Shad Akhtar et al.

Sarcasm detection and humor classification are inherently subtle problems, primarily due to their dependence on the contextual and non-verbal information. Furthermore, existing studies in these two topics are usually constrained in non-English languages such as Hindi, due to the unavailability of qualitative annotated datasets. In this work, we make two major contributions considering the above limitations: (1) we develop a Hindi-English code-mixed dataset, MaSaC, for the multi-modal sarcasm detection and humor classification in conversational dialog, which to our knowledge is the first dataset of its kind; (2) we propose MSH-COMICS, a novel attention-rich neural architecture for the utterance classification. We learn efficient utterance representation utilizing a hierarchical attention mechanism that attends to a small portion of the input sentence at a time. Further, we incorporate dialog-level contextual attention mechanism to leverage the dialog history for the multi-modal classification. We perform extensive experiments for both the tasks by varying multi-modal inputs and various submodules of MSH-COMICS. We also conduct comparative analysis against existing approaches. We observe that MSH-COMICS attains superior performance over the existing models by > 1 F1-score point for the sarcasm detection and 10 F1-score points in humor classification. We diagnose our model and perform thorough analysis of the results to understand the superiority and pitfalls.

Shraman Pramanick, Md Shad Akhtar, Tanmoy Chakraborty

Today's Internet is awash in memes as they are humorous, satirical, or ironic which make people laugh. According to a survey, 33% of social media users in age bracket [13-35] send memes every day, whereas more than 50% send every week. Some of these memes spread rapidly within a very short time-frame, and their virality depends on the novelty of their (textual and visual) content. A few of them convey positive messages, such as funny or motivational quotes; while others are meant to mock/hurt someone's feelings through sarcastic or offensive messages. Despite the appealing nature of memes and their rapid emergence on social media, effective analysis of memes has not been adequately attempted to the extent it deserves. In this paper, we attempt to solve the same set of tasks suggested in the SemEval'20-Memotion Analysis competition. We propose a multi-hop attention-based deep neural network framework, called MHA-MEME, whose prime objective is to leverage the spatial-domain correspondence between the visual modality (an image) and various textual segments to extract fine-grained feature representations for classification. We evaluate MHA-MEME on the 'Memotion Analysis' dataset for all three sub-tasks - sentiment classification, affect classification, and affect class quantification. Our comparative study shows sota performances of MHA-MEME for all three tasks compared to the top systems that participated in the competition. Unlike all the baselines which perform inconsistently across all three tasks, MHA-MEME outperforms baselines in all the tasks on average. Moreover, we validate the generalization of MHA-MEME on another set of manually annotated test samples and observe it to be consistent. Finally, we establish the interpretability of MHA-MEME.

Shivani Kumar, Anubhav Shrimal, Md Shad Akhtar et al.

Efficient discovery of a speaker's emotional states in a multi-party conversation is significant to design human-like conversational agents. During a conversation, the cognitive state of a speaker often alters due to certain past utterances, which may lead to a flip in their emotional state. Therefore, discovering the reasons (triggers) behind the speaker's emotion-flip during a conversation is essential to explain the emotion labels of individual utterances. In this paper, along with addressing the task of emotion recognition in conversations (ERC), we introduce a novel task - Emotion-Flip Reasoning (EFR), that aims to identify past utterances which have triggered one's emotional state to flip at a certain time. We propose a masked memory network to address the former and a Transformer-based network for the latter task. To this end, we consider MELD, a benchmark emotion recognition dataset in multi-party conversations for the task of ERC, and augment it with new ground-truth labels for EFR. An extensive comparison with five state-of-the-art models suggests improved performances of our models for both tasks. We further present anecdotal evidence and both qualitative and quantitative error analyses to support the superiority of our models compared to the baselines.

Shreya Gupta, Parantak Singh, Megha Sundriyal et al.

The conceptualization of a claim lies at the core of argument mining. The segregation of claims is complex, owing to the divergence in textual syntax and context across different distributions. Another pressing issue is the unavailability of labeled unstructured text for experimentation. In this paper, we propose LESA, a framework which aims at advancing headfirst into expunging the former issue by assembling a source-independent generalized model that captures syntactic features through part-of-speech and dependency embeddings, as well as contextual features through a fine-tuned language model. We resolve the latter issue by annotating a Twitter dataset which aims at providing a testing ground on a large unstructured dataset. Experimental results show that LESA improves upon the state-of-the-art performance across six benchmark claim datasets by an average of 3 claim-F1 points for in-domain experiments and by 2 claim-F1 points for general-domain experiments. On our dataset too, LESA outperforms existing baselines by 1 claim-F1 point on the in-domain experiments and 2 claim-F1 points on the general-domain experiments. We also release comprehensive data annotation guidelines compiled during the annotation phase (which was missing in the current literature).

Mohit Bhardwaj, Md Shad Akhtar, Asif Ekbal et al.

In this paper, we present a novel hostility detection dataset in Hindi language. We collect and manually annotate ~8200 online posts. The annotated dataset covers four hostility dimensions: fake news, hate speech, offensive, and defamation posts, along with a non-hostile label. The hostile posts are also considered for multi-label tags due to a significant overlap among the hostile classes. We release this dataset as part of the CONSTRAINT-2021 shared task on hostile post detection.

Kumar Shikhar Deep, Md Shad Akhtar, Asif Ekbal et al.

Expressing the polarity of sentiment as 'positive' and 'negative' usually have limited scope compared with the intensity/degree of polarity. These two tasks (i.e. sentiment classification and sentiment intensity prediction) are closely related and may offer assistance to each other during the learning process. In this paper, we propose to leverage the relatedness of multiple tasks in a multi-task learning framework. Our multi-task model is based on convolutional-Gated Recurrent Unit (GRU) framework, which is further assisted by a diverse hand-crafted feature set. Evaluation and analysis suggest that joint-learning of the related tasks in a multi-task framework can outperform each of the individual tasks in the single-task frameworks.

Md Shad Akhtar, Dushyant Singh Chauhan, Deepanway Ghosal et al.

Related tasks often have inter-dependence on each other and perform better when solved in a joint framework. In this paper, we present a deep multi-task learning framework that jointly performs sentiment and emotion analysis both. The multi-modal inputs (i.e., text, acoustic and visual frames) of a video convey diverse and distinctive information, and usually do not have equal contribution in the decision making. We propose a context-level inter-modal attention framework for simultaneously predicting the sentiment and expressed emotions of an utterance. We evaluate our proposed approach on CMU-MOSEI dataset for multi-modal sentiment and emotion analysis. Evaluation results suggest that multi-task learning framework offers improvement over the single-task framework. The proposed approach reports new state-of-the-art performance for both sentiment analysis and emotion analysis.

Md Shad Akhtar, Deepanway Ghosal, Asif Ekbal et al.

In this paper, through multi-task ensemble framework we address three problems of emotion and sentiment analysis i.e. "emotion classification & intensity", "valence, arousal & dominance for emotion" and "valence & arousal} for sentiment". The underlying problems cover two granularities (i.e. coarse-grained and fine-grained) and a diverse range of domains (i.e. tweets, Facebook posts, news headlines, blogs, letters etc.). The ensemble model aims to leverage the learned representations of three deep learning models (i.e. CNN, LSTM and GRU) and a hand-crafted feature representation for the predictions. Experimental results on the benchmark datasets show the efficacy of our proposed multi-task ensemble frameworks. We obtain the performance improvement of 2-3 points on an average over single-task systems for most of the problems and domains.