Driven by the desire to bridge the gap between human language and machine understanding, I investigate NLP models for tasks such as sentiment analysis, language translation, language generation and conversational AI. My work contributes to the advancement of human-computer interaction and automated understanding of human language.

A critical research area focused on ensuring that artificial intelligence systems behave in ways that are consistent with human values, intentions, and safety requirements. My work explores methods to improve robustness, interpretability, and goal alignment in AI models, especially in uncertain or high-stakes scenarios. A key challenge I address involves preventing misaligned behavior while maintaining model performance and adaptability. I am also interested in developing techniques that incorporate human feedback and ethical constraints into model training and decision-making.

A rapidly evolving research area that studies how humans and artificial intelligence systems collaborate, communicate, and co-adapt in shared tasks. My work focuses on designing AI systems that can interpret human intent more accurately while providing explanations that support trust and effective decision-making. I explore how users perceive, rely on, and challenge AI outputs in dynamic environments, particularly when models are uncertain or make errors. A key research gap I address involves improving transparency and controllability in AI systems so that users can meaningfully guide model behavior during interaction. I am also interested in studying human feedback mechanisms to continuously refine AI performance in real time. Through my research, I aim to build AI systems that are not only intelligent but also cooperative, interpretable, and aligned with human goals in practical settings.

An exciting domain that bridges computer vision and natural language processing to enable machines to understand and interact with multimodal information. My work focuses on developing advanced VLMs for tasks such as image captioning, visual question answering (VQA), and multimodal retrieval, emphasizing creating models that are context-aware, efficient, and generalizable. A critical research gap I address involves enhancing the alignment between visual and textual representations, particularly in scenarios with ambiguous or limited data. I am also keen on tackling challenges related to the semantic grounding of text in complex visual scenes and improving the models’ ability to perform zero-shot or few-shot learning. Through my research, I aim to push the boundaries of multimodal AI, making it more robust and applicable to real-world problems.

With a focus on designing intelligent systems, I am involved in developing models that can learn from and adapt to data. My research in this domain covers supervised and unsupervised learning techniques, neural network architectures, and reinforcement learning applications.

The fusion of natural language processing and speech technologies drives advancements in human-computer communication. My research focuses on developing innovative systems for tasks like automatic speech recognition (ASR), text-to-speech (TTS) synthesis, and spoken language understanding (SLU). A key research gap I address is improving the adaptability of ASR and TTS systems to diverse languages, accents, and speaking styles, particularly in low-resource settings. Additionally, I explore integrating prosody and context modeling to enhance the naturalness of synthesized speech and the accuracy of spoken language interpretation. My work aims to bridge these gaps, creating robust, inclusive, and context-aware speech-driven NLP solutions for real-world applications.