# On the Benefits of Fine-Grained Loss Truncation: A Case Study on Factuality in Summarization

arXiv: https://arxiv.org/abs/2403.05788 · HTML: https://mlsanity.com/paper/2403.05788/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 44 references that are in the map.
- First posted on arXiv 9 Mar 2024.

## Builds on (most cited first, 30 of 32)

- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization](https://mlsanity.com/paper/1808.08745/) (2018)
- [Abstractive Text Summarization Using Sequence-to-Sequence RNNs and Beyond](https://mlsanity.com/paper/1602.06023/) (2016)
- [ScispaCy: Fast and Robust Models for Biomedical Natural Language Processing](https://mlsanity.com/paper/1902.07669/) (2019)
- [QuestEval: Summarization Asks for Fact-based Evaluation](https://mlsanity.com/paper/2103.12693/) (2021)
- [BRIO: Bringing Order to Abstractive Summarization](https://mlsanity.com/paper/2203.16804/) (2022)
- [Detecting Hallucinated Content in Conditional Neural Sequence Generation](https://mlsanity.com/paper/2011.02593/) (2020)
- [On the Origin of Hallucinations in Conversational Models: Is it the Datasets or the Models?](https://mlsanity.com/paper/2204.07931/) (2022)
- [Annotating and Modeling Fine-grained Factuality in Summarization](https://mlsanity.com/paper/2104.04302/) (2021)
- [Hallucinated but Factual! Inspecting the Factuality of Hallucinations in Abstractive Summarization](https://mlsanity.com/paper/2109.09784/) (2021)
- [Truncation Sampling as Language Model Desmoothing](https://mlsanity.com/paper/2210.15191/) (2022)
- [Understanding Factual Errors in Summarization: Errors, Summarizers, Datasets, Error Detectors](https://mlsanity.com/paper/2205.12854/) (2022)
- [ASSET: A Dataset for Tuning and Evaluation of Sentence Simplification Models with Multiple Rewriting Transformations](https://mlsanity.com/paper/2005.00481/) (2020)
- [Controlled Hallucinations: Learning to Generate Faithfully from Noisy Data](https://mlsanity.com/paper/2010.05873/) (2020)
- [Improving Faithfulness in Abstractive Summarization with Contrast Candidate Generation and Selection](https://mlsanity.com/paper/2104.09061/) (2021)
- [Improved Natural Language Generation via Loss Truncation](https://mlsanity.com/paper/2004.14589/) (2020)
- [Mutual Information Alleviates Hallucinations in Abstractive Summarization](https://mlsanity.com/paper/2210.13210/) (2022)
- [Generating Data to Mitigate Spurious Correlations in Natural Language Inference Datasets](https://mlsanity.com/paper/2203.12942/) (2022)
- [Paragraph-level Simplification of Medical Texts](https://mlsanity.com/paper/2104.05767/) (2021)
- [Faithful or Extractive? On Mitigating the Faithfulness-Abstractiveness Trade-off in Abstractive Summarization](https://mlsanity.com/paper/2108.13684/) (2021)
- [EASSE: Easier Automatic Sentence Simplification Evaluation](https://mlsanity.com/paper/1908.04567/) (2019)
- [FactPEGASUS: Factuality-Aware Pre-training and Fine-tuning for Abstractive Summarization](https://mlsanity.com/paper/2205.07830/) (2022)
- [Expertise Style Transfer: A New Task Towards Better Communication between Experts and Laymen](https://mlsanity.com/paper/2005.00701/) (2020)
- [Learning to Revise References for Faithful Summarization](https://mlsanity.com/paper/2204.10290/) (2022)
- [Training Dynamics for Text Summarization Models](https://mlsanity.com/paper/2110.08370/) (2021)
- [Don't Say What You Don't Know: Improving the Consistency of Abstractive Summarization by Constraining Beam Search](https://mlsanity.com/paper/2203.08436/) (2022)
- [Med-EASi: Finely Annotated Dataset and Models for Controllable Simplification of Medical Texts](https://mlsanity.com/paper/2302.09155/) (2023)
- [NapSS: Paragraph-level Medical Text Simplification via Narrative Prompting and Sentence-matching Summarization](https://mlsanity.com/paper/2302.05574/) (2023)
- [Contrastive Error Attribution for Finetuned Language Models](https://mlsanity.com/paper/2212.10722/) (2022)

## Built on by (most cited first, 2 of 2)

- [An Audit on the Perspectives and Challenges of Hallucinations in NLP](https://mlsanity.com/paper/2404.07461/) (2024)
- [How to Learn in a Noisy World? Self-Correcting the Real-World Data Noise in Machine Translation](https://mlsanity.com/paper/2407.02208/) (2024)

## Abstract (arXiv)

Text summarization and simplification are among the most widely used applications of AI. However, models developed for such tasks are often prone to hallucination, which can result from training on unaligned data. One efficient approach to address this issue is Loss Truncation (LT) (Kang and Hashimoto, 2020), an approach to modify the standard log loss to adaptively remove noisy examples during training. However, we find that LT alone yields a considerable number of hallucinated entities on various datasets. We study the behavior of the underlying losses between factual and non-factual examples, to understand and refine the performance of LT. We demonstrate that LT's performance is limited when the underlying assumption that noisy targets have higher NLL loss is not satisfied, and find that word-level NLL among entities provides better signal for distinguishing factuality. We then leverage this to propose a fine-grained NLL loss and fine-grained data cleaning strategies, and observe improvements in hallucination reduction across some datasets. Our work is available at https://https://github.com/yale-nlp/fine-grained-lt.