http://repository.iiitd.edu.in/xmlui/handle/123456789/15 Thu, 20 Aug 2026 11:24:32 GMT 2026-08-20T11:24:32Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1993 Title: Accessing trust in mental health counselling conversation Authors: Agarwal, Ayush; Akhtar, Md. Shad (Advisor) Abstract: Mental health has emerged as the most critical global healthcare concern which is ac celerated by recent lifestyle changes, evolving individual aspirations, and social media. The use of AI in mental health has been useful in every aspect and stage of mental health- from monitoring, assistance and evaluation to providing a cost-effective, scalable, and qualitative improvement of mental health therapeutic outcomes. However, there is a serious lack of research and assistive tools in providing real-time patient feedback to measure aspects and overall therapeutic outcomes of psychotherapy. To this end, we propose a novel metric of trust in mental health dialogue systems. Specifically, every patient’s utterance in a counseling dialogue is scored on a scale that signals the degree of trust developed in the patient towards the therapist at any point in the therapy ses sion. The contribution of this work is twofold (a) we provide a concretized definition of the novel metric of trust in mental health dialogue systems and a clear annotation framework that scores the strength of trust developed between a patient and therapist during counseling. (b) We discuss various strategies to model trust correctly and finally model it as a knowledge-guided time-series forecasting problem. We perform exten sive experimentation and benchmark trust on various classical and modern methods of time-series forecasting, with our best model achieving an MSE loss of 0.0017. Thu, 18 Jul 2024 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1993 2024-07-18T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1965 Title: Ship marine strategy database access using natural language: an application of LLM-based text-to-SQL model Authors: Ghorai, Arunoday; Goyal, Vikram (Advisor) Abstract: The growing reliance on relational databases across industries and the ability to efficiently query and extract from a structured database has become a crucial skill in the industry. However, the Complexity of SQL Syntax creates a barrier for non-technical uses limiting their ability to interact with databases effectively. Natural Language to SQL (NL-to-SQL) query generation performs a critical task in bridging gap between non-technical users and relational databases and enables intuitive data interaction with out any need for SQL expertise. This thesis first explores various Text-to-SQL approaches, leveraging both proprietary model like Open AI’s GPT-4 and open-source models like RESDSQL, focusing on their performance across benchmark datasets like Spider, CoSQL and SPARC. Additionally, two datasets, MORD and CMEC are prepared from the real world use cases to highlight unique challenges such as hierarchical data structures, string matching operations, and privacy issues. The MORD dataset was queried using GPT-4 integrated with LangChain, to showcase natural language interaction with data and the usability of proprietary models without any tuning to domain specific dataset. Meanwhile the CMEC dataset is a privately curated dataset and access to it needs to be confidential. So we use open source models like RESDSQL that run on local server in order to minimize leakage. The dataset is pre-processed into a relational schema, and RESDSQL is fine tuned on curated NL to SQL pairs to improve performance. String matching techniques are applied to prepare better prompts in order to further enhance the results generated by the model. Sun, 01 Dec 2024 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1965 2024-12-01T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1928 Title: Automatic speech recognition for code-mixed Indian languages Authors: Kumar, Shivam; Akhtar, Md. Shad (Advisor) Abstract: Code-mixing presents significant challenges for Automatic Speech Recognition (ASR), especially for Indian languages, due to homophone ambiguity, domain-specific word identification, and data scarcity. Traditional ASR models struggle with these complexities, often failing to differentiate between phonetically similar words in multilingual contexts. To address this, we propose CLEAR, a novel rescoring model that integrates descriptive prompting and LLM-based rescoring while analyzing the impact of n-best hypotheses across multiple beam widths. CLEAR enhances ASR performance, achieving S-WER of 26.9, P-WER of 26.46, and T- WER of 25.04—improving by 6.9%, 13.47%, and 4.42%, respectively, over the best baseline, i.e., TDNN. These findings demonstrate that CLEAR effectively resolves homophone ambiguities and refines transcriptions, leading to a 13.56% S-WER reduction over fine-tuned Whisper without extensive pretraining. In addition to improving transcription accuracy, CLEAR introduces a principled framework for handling ambiguous hypotheses in low-resource, script-mixed speech. CLEAR is a generic framework that can be adopted for multiple languages apart from Hindi. This work sets the foundation for more linguistically aware ASR systems tailored for multilingual societies. Thu, 01 May 2025 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1928 2025-05-01T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1927 Title: Simulating distributed ML training under heterogeneous network infrastructure Authors: Temura, Arjun; Shah, Rinku (Advisor) Abstract: There has been an increasing demand to train ML models, particularly large language models (LLMs), on multiple GPUs to ensure reduced training time and costs. However, making the correct training configuration choice (for example, the number of GPUs, parallelism technique, and network topology) to ensure minimal training time and maximum resource utilisation remains challenging. Distributed ML simulators help users with capacity planning and selecting optimal configuration knobs before training. However, state-of-the-art simulators assume homogeneous compute and network infrastructure. Distributed ML training infrastructure frequently consists of heterogeneous hardware, arising from generational shifts in devices or resource sharing in cloud environments. Several training plans have been introduced in the last few years to make the best out of the available heterogeneous hardware and improve training performance. However, there are no simulation tools that mimic realistic training environments for these heterogeneity-aware training strategies. Generally, heterogeneity-aware training optimisations make guided training plans considering compute or network heterogeneity. Therefore, we design a heterogeneity-aware distributed ML training simulator that supports compute and network heterogeneity. As part of our preliminary analysis, we study GPU communication flows for popular LLMs (GPT, Mixtral) on existing simulation frameworks under realistic training configurations with network heterogeneity. We observe improvement in the completion time of the median flows under heterogeneous configurations during training. Additionally, we develop ideas for effective model partitioning strategies in light of heterogeneous compute. Finally, we briefly discuss the additional abstractions required for our simulator to leverage heterogeneous hardware effectively. Wed, 21 May 2025 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1927 2025-05-21T00:00:00Z