We introduce a zero-shot merging framework for large language models (LLMs) that consolidates specialized domain experts into a single model without any further training.Our core contribution lies in leveraging relative task vectors-difference representations encoding each expert's unique traits with respect to a shared base model-to guide a principled and efficient merging process.By dissecting parameters into common dimensions (averaged across experts) and complementary dimensions (unique to each expert), we strike an optimal balance between generalization and specialization.We further devise a compression mechanism for the complementary parameters, retaining only principal components and scalar multipliers per expert, thereby minimizing overhead.A dynamic router then selects the most relevant domain at inference, ensuring that domain-specific precision is preserved.Experiments on code generation, mathematical reasoning, medical question answering, and instruction-following benchmarks confirm the versatility and effectiveness of our approach.Altogether, this framework enables truly adaptive and scalable LLMs that seamlessly integrate specialized knowledge for improved zero-shot performance.

Large Language Models (LLMs) are increasingly deployed as chatbots, yet their ability to personalize responses to user preferences remains limited. We introduce PrefEval, a benchmark for evaluating LLMs' ability to infer, memorize and adhere to user preferences in long-context conversational setting. PrefEval comprises 3,000 manually curated user preference and query pairs spanning 20 topics. PrefEval contains user personalization or preference information in both explicit and implicit preference forms, and evaluates LLM performance using a generation and a classification task. With PrefEval, we have evaluated 10 open-sourced and proprietary LLMs in multi-session conversations with varying context lengths up to 100k tokens. We benchmark with various prompting, iterative feedback, and retrieval-augmented generation methods. Our benchmarking effort reveals that state-of-the-art LLMs face significant challenges in following users' preference during conversations. In particular, in zero-shot settings, preference following accuracy falls below 10\% at merely 10 turns (~3k tokens) across most evaluated models. Even with advanced prompting and retrieval methods, preference following still deteriorates in long-context conversations. Furthermore, we show that fine-tuning on PrefEval significantly improves performance. We believe PrefEval serves as a valuable resource for measuring, understanding, and enhancing LLMs' proactive preference following abilities, paving the way for personalized conversational agents.

Yan Xu, Mahdi Namazifar, Devamanyu Hazarika, Aishwarya Padmakumar, Yang Liu, Dilek Hakkani-Tur. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Yen-Ting Lin, Alexandros Papangelis, Seokhwan Kim, Sungjin Lee, Devamanyu Hazarika, Mahdi Namazifar, Di Jin, Yang Liu, Dilek Hakkani-Tur. Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics. 2023.

Instruction-based multitasking has played a critical role in the success of large language models (LLMs) in multi-turn dialog applications.While publicly-available LLMs have shown promising performance, when exposed to complex instructions with multiple constraints, they lag against state-of-the-art models like Chat-GPT.In this work, we hypothesize that the availability of large-scale complex demonstrations is crucial in bridging this gap.Focusing on dialog applications, we propose a novel framework, CESAR, that unifies a large number of dialog tasks in the same format and allows programmatic induction of complex instructions without any manual effort.We apply CESAR on InstructDial, a benchmark for instruction-based dialog tasks.We further enhance InstructDial with new datasets and tasks and utilize CESAR to induce complex tasks with compositional instructions.This results in a new benchmark called InstructDial++, which includes 63 datasets with 86 basic tasks and 68 composite tasks.Through rigorous experiments, we demonstrate the scalability of CESAR in providing rich instructions.Models trained on InstructDial++ can follow compositional prompts, such as prompts that ask for multiple stylistic constraints.

Dot-product attention is a core module in the present generation of neural network models, particularly transformers, and is being leveraged across numerous areas such as natural language processing and computer vision. This attention module is comprised of three linear transformations, namely query, key, and value linear transformations, each of which has a bias term. In this work, we study the role of these bias terms, and mathematically show that the bias term of the key linear transformation is redundant and could be omitted without any impact on the attention module. Moreover, we argue that the bias term of the value linear transformation has a more prominent role than that of the bias term of the query linear transformation. We empirically verify these findings through multiple experiments on language modeling, natural language understanding, and natural language generation tasks.

Chao Zhao, Spandana Gella, Seokhwan Kim, Di Jin, Devamanyu Hazarika, Alexandros Papangelis, Behnam Hedayatnia, Mahdi Namazifar, Yang Liu, Dilek Hakkani-Tur. Proceedings of the 24th Meeting of the Special Interest Group on Discourse and Dialogue. 2023.

Sentiment analysis as a field has come a long way since it was first introduced as a task nearly 20 years ago. It has widespread commercial applications in various domains like marketing, risk management, market research, and politics, to name a few. Given its saturation in specific subtasks — such as sentiment polarity classification — and datasets, there is an underlying perception that this field has reached its maturity. In this article, we discuss this perception by pointing out the shortcomings and under-explored, yet key aspects of this field necessary to attaintruesentiment understanding. We analyze the significant leaps responsible for its current relevance. Further, we attempt to chart a possible course for this field that covers many overlooked and unanswered questions.

Controlling neural network-based models for natural language generation (NLG) to realize desirable attributes in the generated outputs has broad applications in numerous areas such as machine translation, document summarization, and dialog systems. Approaches that enable such control in a zero-shot manner would be of great importance as, among other reasons, they remove the need for additional annotated data and training. In this work, we propose novel approaches for controlling encoder-decoder transformer-based NLG models in zero shot. While zero-shot control has previously been observed in massive models (e.g., GPT3), our method enables such control for smaller models. This is done by applying two control knobs, attention biasing and context augmentation, to these models directly during decoding and without additional training or auxiliary models. These knobs control the generation process by directly manipulating trained NLG models (e.g., biasing cross-attention layers). We show that not only are these NLG models robust to such manipulations but also their behavior could be controlled without an impact on their generation performance.

Abhinav Ramesh Kashyap, Devamanyu Hazarika, Min-Yen Kan, Roger Zimmermann, Soujanya Poria. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

Prefix-tuning, or more generally continuous prompt tuning, has become an essential paradigm of parameter-efficient transfer learning.Using a large pre-trained language model (PLM), prefix-tuning can obtain strong performance by training only a small portion of parameters.In this paper, we propose to understand and further develop prefix-tuning through the kernel lens.Specifically, we make an analogy between prefixes and inducing variables in kernel methods and hypothesize that prefixes serving as inducing variables would improve their overall mechanism.From the kernel estimator perspective, we suggest a new variant of prefix-tuning-inducer-tuning, which shares the exact mechanism as prefix-tuning while leveraging the residual form found in adaptertuning.This mitigates the initialization issue in prefix-tuning.Through comprehensive empirical experiments on natural language understanding and generation tasks, we demonstrate that inducer-tuning can close the performance gap between prefix-tuning and fine-tuning.

Devamanyu Hazarika, Yingting Li, Bo Cheng, Shuai Zhao, Roger Zimmermann, Soujanya Poria. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

Abhinav Ramesh Kashyap, Devamanyu Hazarika, Min-Yen Kan, Roger Zimmermann. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

Cross-domain sentiment analysis has received significant attention in recent years, prompted by the need to combat the domain gap between different applications that make use of sentiment analysis.In this paper, we take a novel perspective on this task by exploring the role of external commonsense knowledge.We introduce a new framework, KinGDOM, which utilizes the ConceptNet knowledge graph to enrich the semantics of a document by providing both domain-specific and domain-general background concepts.These concepts are learned by training a graph convolutional autoencoder that leverages inter-domain concepts in a domain-invariant manner.Conditioning a popular domain-adversarial baseline method with these learned concepts helps improve its performance over state-of-the-art approaches, demonstrating the efficacy of our proposed framework.

Dhanasekar Sundararaman, Shijing Si, Vivek Subramanian, Guoyin Wang, Devamanyu Hazarika, Lawrence Carin. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020.

Multimodal Sentiment Analysis is an active area of research that leverages multimodal signals for affective understanding of user-generated videos. The predominant approach, addressing this task, has been to develop sophisticated fusion techniques. However, the heterogeneous nature of the signals creates distributional modality gaps that pose significant challenges. In this paper, we aim to learn effective modality representations to aid the process of fusion. We propose a novel framework, MISA, which projects each modality to two distinct subspaces. The first subspace is modality-invariant, where the representations across modalities learn their commonalities and reduce the modality gap. The second subspace is modality-specific, which is private to each modality and captures their characteristic features. These representations provide a holistic view of the multimodal data, which is used for fusion that leads to task predictions. Our experiments on popular sentiment analysis benchmarks, MOSI and MOSEI, demonstrate significant gains over state-of-the-art models. We also consider the task of Multimodal Humor Detection and experiment on the recently proposed UR_FUNNY dataset. Here too, our model fares better than strong baselines, establishing MISA as a useful multimodal framework.

Emotion detection in conversations is a necessary step for a number of applications, including opinion mining over chat history, social media threads, debates, argumentation mining, understanding consumer feedback in live conversations, and so on. Currently systems do not treat the parties in the conversation individually by adapting to the speaker of each utterance. In this paper, we describe a new method based on recurrent neural networks that keeps track of the individual party states throughout the conversation and uses this information for emotion classification. Our model outperforms the state-of-the-art by a significant margin on two different datasets.

Santiago Castro, Devamanyu Hazarika, Verónica Pérez-Rosas, Roger Zimmermann, Rada Mihalcea, Soujanya Poria. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. 2019.

Emotion recognition in conversations (ERC) is a challenging task that has recently gained popularity due to its potential applications.Until now, however, there has been no largescale multimodal multi-party emotional conversational database containing more than two speakers per dialogue.To address this gap, we propose the Multimodal EmotionLines Dataset (MELD), an extension and enhancement of EmotionLines.MELD contains about 13,000 utterances from 1,433 dialogues from the TV-series Friends.Each utterance is annotated with emotion and sentiment labels, and encompasses audio, visual, and textual modalities.We propose several strong multimodal baselines and show the importance of contextual and multimodal information for emotion recognition in conversations.The full dataset is available for use at http:// affective-meld.github.io.

With the recent development of deep learning, research in AI has gained new vigor and prominence. While machine learning has succeeded in revitalizing many research fields, such as computer vision, speech recognition, and medical diagnosis, we are yet to witness impressive progress in natural language understanding. One of the reasons behind this unmatched expectation is that, while a bottom-up approach is feasible for pattern recognition, reasoning and understanding often require a top-down approach. In this work, we couple sub-symbolic and symbolic AI to automatically discover conceptual primitives from text and link them to commonsense concepts and named entities in a new three-level knowledge representation for sentiment analysis. In particular, we employ recurrent neural networks to infer primitives by lexical substitution and use them for grounding common and commonsense knowledge by means of multi-dimensional scaling.

The literature in automated sarcasm detection has mainly focused on lexical-, syntactic- and semantic-level analysis of text. However, a sarcastic sentence can be expressed with contextual presumptions, background and commonsense knowledge. In this paper, we propose a ContextuAl SarCasm DEtector (CASCADE), which adopts a hybrid approach of both content- and context-driven modeling for sarcasm detection in online social media discussions. For the latter, CASCADE aims at extracting contextual information from the discourse of a discussion thread. Also, since the sarcastic nature and form of expression can vary from person to person, CASCADE utilizes user embeddings that encode stylometric and personality features of users. When used along with content-based feature extractors such as convolutional neural networks, we see a significant boost in the classification performance on a large Reddit corpus.

Emotion recognition in conversations is crucial for building empathetic machines.Current work in this domain do not explicitly consider the inter-personal influences that thrive in the emotional dynamics of dialogues.To this end, we propose Interactive COnversational memory Network (ICON), a multimodal emotion detection framework that extracts multimodal features from conversational videos and hierarchically models the selfand interspeaker emotional influences into global memories.Such memories generate contextual summaries which aid in predicting the emotional orientation of utterance-videos.Our model outperforms state-of-the-art networks on multiple classification and regression tasks in two benchmark datasets.

Emotion recognition in conversations is crucial for the development of empathetic machines. Present methods mostly ignore the role of inter-speaker dependency relations while classifying emotions in conversations. In this paper, we address recognizing utterance-level emotions in dyadic conversational videos. We propose a deep neural framework, termed conversational memory network, which leverages contextual information from the conversation history. The framework takes a multimodal approach comprising audio, visual and textual features with gated recurrent units to model past utterances of each speaker into memories. Such memories are then merged using attention-based hops to capture inter-speaker dependencies. Experiments show an accuracy improvement of 3-4% over the state of the art.

Soujanya Poria, Erik Cambria, Devamanyu Hazarika, Navonil Majumder, Amir Zadeh, Louis-Philippe Morency. Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2017.

Multimodal sentiment analysis involves identifying sentiment in videos and is a developing field of research. Unlike current works, which model utterances individually, we propose a recurrent model that is able to capture contextual information among utterances. In this paper, we also introduce attentionbased networks for improving both context learning and dynamic feature fusion. Our model shows 6-8% improvement over the state of the art on a benchmark dataset.

Sarcasm detection is a key task for many natural language processing tasks. In sentiment analysis, for example, sarcasm can flip the polarity of an “apparently positive” sentence and, hence, negatively affect polarity detection performance. To date, most approaches to sarcasm detection have treated the task primarily as a text categorization problem. Sarcasm, however, can be expressed in very subtle ways and requires a deeper understanding of natural language that standard text categorization techniques cannot grasp. In this work, we develop models based on a pre-trained convolutional neural network for extracting sentiment, emotion and personality features for sarcasm detection. Such features, along with the network’s baseline features, allow the proposed models to outperform the state of the art on benchmark datasets. We also address the often ignored generalizability issue of classifying data that have not been seen by the models at learning phase.