Back to search

Georgia Institute of Technology

Adapting Models to User Intent for Robust and Controllable Generation

Abstract

dc:description.abstract

As artificial intelligence (AI) is increasingly deployed in high-stakes, open-world environments, the demand for systems that are both highly capable and strictly trustworthy has become paramount. Modern generative models require high test-time plasticity to remain responsive to complex and dynamic human intentions. However, this same architectural flexibility inadvertently leaves them vulnerable to manipulation by bad actors and unintended changes. To ensure AI remains an empowering tool rather than a fragile liability, this dissertation presents a comprehensive framework for adapting models to both benign and malicious user intent. First, this work introduces interpretable test-time mechanisms that empower users to precisely steer model behavior. These interventions are designed to enforce hard constraints in graph diffusion models and guide large language and imitation learning models toward exemplary and directed behavior descriptions. However, because unconstrained plasticity inherently alters a model's risk profile, the research then systematically diagnoses the vulnerabilities in the systems, exposing vulnerabilities in dynamic graph models and multimodal language models. To safely support user-driven controllability without succumbing to these adversarial risks, the final phase demonstrates how structural adaptivity of system prompts can protect a language model's core without compromising its benign behavior. Ultimately, this research works towards resolving the modern stability-plasticity dilemma, transforming vulnerable models into deeply steerable and robust extensions of human intent.

Degree

thesis:*
Name thesis:degree_name
Computer Science, PhD
Department thesis:degree_department
Computational Science and Engineering
Grantor
Georgia Institute of Technology
Year dc:date.issued
2026

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Sharma, Kartik
Advisor dc:contributor.advisor
  • Kumar, Srijan
Committee members dc:contributor.committeemember
  • Rakshit Trivedi
  • Zhang, Chao
  • Dai, Bo
  • Chau, Duen Horng (Polo)

Rights

Language dc:language.iso
English

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/1853/81756

Chain of custody

source
Harvested from
Georgia Tech
Base URL
repository.gatech.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Sharma, Kartik. Adapting Models to User Intent for Robust and Controllable Generation. Georgia Institute of Technology, 2026. https://hdl.handle.net/1853/81756