Detection and mitigation of misbehaviour in LLMs

Kirtane, Neeraja Kiran

Permalink

https://hdl.handle.net/2142/129254 Copy

Description

Title

Detection and mitigation of misbehaviour in LLMs

Author(s)

Kirtane, Neeraja Kiran

Issue Date

2025-04-28

Director of Research (if dissertation) or Advisor (if thesis)

Peng, Hao

Department of Study

Siebel School Comp & Data Sci

Discipline

Computer Science

Degree Granting Institution

University of Illinois Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

AI safety, Factuality

Abstract

Language models (LMs) hallucinate. We inquire: Can we detect and mitigate hallucinations before they happen? This work answers this research question in the positive, by showing that the internal representations of LMs provide rich signals that can be used for this purpose. We introduce FactCheckmate, which preemptively detects hallucinations by learning a classifier that predicts whether the LM will hallucinate, based on the model’s hidden states produced over the inputs, before decoding begins. If a hallucination is detected, FactCheckmate then intervenes by adjusting the LM’s hidden states such that the model will produce more factual outputs. FactCheckmate provides fresh insights that the inner workings of LMs can be revealed by their hidden states. Practically, both its detection and mitigation models are lightweight, adding little inference overhead; FactCheckmate proves a more efficient approach for mitigating hallucinations compared to many post-hoc alternatives. We evaluate FactCheckmate over LMs of different scales and model families (including Llama, Mistral, Qwen and Gemma), across a variety of QA datasets from different domains. Our results demonstrate the effectiveness of FactCheckmate, achieving over 70% preemptive detection accuracy. On average, outputs generated by LMs with intervention are 34.4% more factual compared to those without.

Graduation Semester

2025-05

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/129254

Copyright and License Information

Copyright 2025 Neeraja Kiran Kirtane

Owning Collections