Technical Workshop - Evaluation on GANBART for Lecture Summarization(2nd Offering) by: Thennavan Karuppaiah

Thursday, July 31, 2025 - 12:00

School of Computer Science -Technical Workshop Series

 

Evaluation on GANBART for Lecture Summarization (2nd Offering)

Presenter:  Thennavan Karuppaiah

Date: Thursday, July 31st, 2025

Time:  12:00 pm

Location: 4th Floor (Workshop space) at 300 Ouellette Avenue (School of Computer Science Advanced Computing Hub)

 

 

Abstract: 

This report presents our investigation into GANBART, a novel adversarial framework designed for lecture summarization. We began with a publicly available Lecture Summarization dataset and addressed data scarcity by augmenting the corpus through translation and back-translation. Our model leverages LoRA fine-tuning on BART to reduce training parameters, paired with a GAN architecture wherein the BART-based Generator produces candidate summaries, and a BERT-based Discriminator refines them via adversarial feedback. We compare our approach against BARTLarge-CNN as a baseline, evaluating system outputs using ROUGE. Results demonstrate that GANBART yields improved coverage (higher ROUGE) in generated summaries, indicating its efficacy in producing more coherent and informative summaries from a limited-resource lecture dataset.

 

Workshop Outline:
  1. Welcome & Goals
  2. Why Lecture Summarization is Hard
  3. Dataset Tour & Augmentation
  4. LoRA Fine- tuning Primer
  5. Building GANBART

 

Prerequisites:
  • Basic NLP pipeline & tokenization
  • Transformer encoder–decoder anatomy (queries/keys/values)
  • GAN intuition (generator vs. discriminator)

 

Biography: 

Thennavan Karuppaiah is a Master of Applied Computing candidate at the University of Windsor. Their research probes low-resource text-summarization and parameter-efficient fine-tuning of large language models.

 

Registration Link (only MAC students need to pre-register)