Tag
TIAO is a token importance-aware reinforcement learning strategy for text summarization that reweights trajectories based on token dependencies, achieving results comparable to GPT-4 and GPT-5-nano on real-world datasets.
This paper proposes a novel method for extractive text summarization using hypergraph domination, comparing its performance with existing graph-based approaches.
A comparative study of BART, BERT, and RoBERTa for text summarization, examining their architectures and suitability for extractive and abstractive summarization tasks.
This paper examines the concepts of uniformity and linearity in literary summaries, analyzing their prevalence and implications for narrative structure.
This paper defines and quantifies sentiment drift in RLHF-trained summarization models, proposes a Policy Attribution framework to identify causes, and introduces a Sentiment-Aware KL Regularization method to reduce drift.
This paper introduces Reference Abstraction (RA), Summary Abstraction (SA), and Abstraction Ratio (AR) metrics to quantify abstractiveness in text summarization, using harmonic mean of document lengths and cubic non-overlap factor. Empirical evaluation on XSUM with four models shows the metrics effectively discriminate between extractive and abstractive summaries, and flag potential hallucination.
BrainFlow is a tool that turns rambling thoughts into coherent, organized notes.
Presents MASF, a multi-model adaptive selection framework that integrates multiple fine-tuned transformer summarization models and selects the highest-quality summary, achieving 88.63% BERTScore on CNN/DailyMail and outperforming several LLMs.
This paper introduces a KAN-enhanced BiGRU architecture for classifying and summarizing multilingual legal documents from Bangladesh, achieving modest accuracy and ROUGE scores and demonstrating that the KAN block improves classification accuracy over the baseline BiGRU.
This paper proposes a parameter-efficient vocabulary adaptation method for LLM-based text summarization in specialized domains, augmenting pretrained tokenizers with domain-specific tokens and selectively replacing under-trained ones to reduce training time by 35-55% and parameter counts by up to 37%.