Date of Award

Spring 2026

Document Type

Open Access Dissertation

Department

Computer Science and Engineering

First Advisor

Biplav Srivastava

Second Advisor

Amit Sheth

Abstract

Planning is a fundamental capability for intelligent systems, yet classical planners remain difcult to scale and generalize across domains due to their reliance on hand- engineered models, brittle search procedures, and limited adaptability. These limita- tions constrain real-world applications that require fexible reasoning, robust decision- making, and the ability to operate across diverse environments.

Large language models have emerged as powerful learners supported by their ar- chitectures and large-scale training data, demonstrating strong capabilities in tasks such as question answering, code generation, summarization, and reasoning. Moti- vated by their broad generalization abilities, I investigate whether language models can be systematically leveraged to advance automated planning.

I begin by conducting a comprehensive study of existing literature to categorize how language models are being used for automated planning, resulting in a structured taxonomy that captures the full landscape of techniques, objectives, and use cases, along with a semi-automated platform for tracking the evolution of this rapidly grow- ing area. Building on this foundation, I examine how language models can be used for efective plan generation by evaluating their pretrained capabilities and quantifying the benefts of fne-tuning on planning-specifc data.

To alleviate the limitations of relying solely on language models, I introduce a complementary approach for plan generation by training a compact foundational model from scratch. This approach adopts a state-centric perspective to generalized planning, learning transition dynamics over graph-based state representations. In addition, I develop a planning ontology to systematically capture metadata, enabling  more structured data representation and facilitating improved training of future mod- els. I also introduce plan summarization as a downstream task, demonstrating how these trained models can be efectively leveraged for concise and structured plan understanding.

Next, I develop neurosymbolic architectures, grounded in the SOFAI “Thinking, Fast and Slow” paradigm, that integrate language models with symbolic planning under metacognitive control to achieve more robust and generalizable plan gener- ation. These architectures address key limitations of language models in planning and demonstrate how symbolic reasoning can complement learning-based models. Finally, I evaluate the resulting generalized planners in real-world applications in- cluding dialog-based information retrieval, trustworthy conversational AI for sensitive domains, and adaptive replanning in stochastic manufacturing environments, show- ing how generalized planners combining language models and symbolic methods are practical tools for deployment.

Collectively, my contributions advance the scientifc understanding of how lan- guage models can support, improve, and generalize automated planning, ofering a coherent path toward planning systems that combine the strengths of learning and symbolic reasoning.

Rights

© 2026, Vishal Pallagani

Download

Share

COinS