# Thesis proposal: Are We Losing Textual Diversity to Natural Language Processing?

arXiv: https://arxiv.org/abs/2409.09568 · HTML: https://mlsanity.com/paper/2409.09568/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 38 of its 75 references that are in the map.
- First posted on arXiv 15 Sep 2024.

## Builds on (most cited first, 30 of 38)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Books](https://mlsanity.com/paper/1506.06724/) (2015)
- [Diffusion-LM Improves Controllable Text Generation](https://mlsanity.com/paper/2205.14217/) (2022)
- [Neural Network Acceptability Judgments](https://mlsanity.com/paper/1805.12471/) (2018)
- [Sequence Level Training with Recurrent Neural Networks](https://mlsanity.com/paper/1511.06732/) (2015)
- [Six Challenges for Neural Machine Translation](https://mlsanity.com/paper/1706.03872/) (2017)
- [Blockwise Parallel Decoding for Deep Autoregressive Models](https://mlsanity.com/paper/1811.03115/) (2018)
- [CometKiwi: IST-Unbabel 2022 Submission for the Quality Estimation Shared Task](https://mlsanity.com/paper/2209.06243/) (2022)
- [Is MAP Decoding All You Need? The Inadequacy of the Mode in Neural Machine Translation](https://mlsanity.com/paper/2005.10283/) (2020)
- [Analyzing Uncertainty in Neural Machine Translation](https://mlsanity.com/paper/1803.00047/) (2018)
- [Future Lens: Anticipating Subsequent Tokens from a Single Hidden State](https://mlsanity.com/paper/2311.04897/) (2023)
- [If beam search is the answer, what was the question?](https://mlsanity.com/paper/2010.02650/) (2020)
- [On NMT Search Errors and Model Errors: Cat Got Your Tongue?](https://mlsanity.com/paper/1908.10090/) (2019)
- [Why Does Surprisal From Larger Transformer-Based Language Models Provide a Poorer Fit to Human Reading Times?](https://mlsanity.com/paper/2212.12131/) (2022)
- [High Quality Rather than High Model Probability: Minimum Bayes Risk Decoding with Neural Metrics](https://mlsanity.com/paper/2111.09388/) (2021)
- [The Curious Decline of Linguistic Diversity: Training Language Models on Synthetic Text](https://mlsanity.com/paper/2311.09807/) (2023)
- [Text Generation with Diffusion Language Models: A Pre-training Approach with Continuous Paragraph Denoise](https://mlsanity.com/paper/2212.11685/) (2022)
- [Large Language Models Suffer From Their Own Output: An Analysis of the Self-Consuming Training Loop](https://mlsanity.com/paper/2311.16822/) (2023)
- [Machine Translationese: Effects of Algorithmic Bias on Linguistic Complexity in Machine Translation](https://mlsanity.com/paper/2102.00287/) (2021)
- [Revisiting the Uniform Information Density Hypothesis](https://mlsanity.com/paper/2109.11635/) (2021)
- [Sampling-Based Approximations to Minimum Bayes Risk Decoding for Neural Machine Translation](https://mlsanity.com/paper/2108.04718/) (2021)
- [Understanding the Properties of Minimum Bayes Risk Decoding in Neural Machine Translation](https://mlsanity.com/paper/2105.08504/) (2021)
- [Lost in Translation: Loss and Decay of Linguistic Richness in Machine Translation](https://mlsanity.com/paper/1906.12068/) (2019)
- [Sentence-Level Fluency Evaluation: References Help, But Can Be Spared!](https://mlsanity.com/paper/1809.08731/) (2018)
- [Identifying Weaknesses in Machine Translation Metrics Through Minimum Bayes Risk Decoding: A Case Study for COMET](https://mlsanity.com/paper/2202.05148/) (2022)
- [CodeFusion: A Pre-trained Diffusion Model for Code Generation](https://mlsanity.com/paper/2310.17680/) (2023)
- [Post-editese: an Exacerbated Translationese](https://mlsanity.com/paper/1907.00900/) (2019)
- [Global Voices: Crossing Borders in Automatic News Summarization](https://mlsanity.com/paper/1910.00421/) (2019)

## Abstract (arXiv)

This thesis argues that the currently widely used Natural Language Processing algorithms possibly have various limitations related to the properties of the texts they handle and produce. With the wide adoption of these tools in rapid progress, we must ask what these limitations are and what are the possible implications of integrating such tools even more deeply into our daily lives. As a testbed, we have chosen the task of Neural Machine Translation (NMT). Nevertheless, we aim for general insights and outcomes, applicable even to current Large Language Models (LLMs). We ask whether the algorithms used in NMT have inherent inductive biases that are beneficial for most types of inputs but might harm the processing of untypical texts. To explore this hypothesis, we define a set of measures to quantify text diversity based on its statistical properties, like uniformity or rhythmicity of word-level surprisal, on multiple scales (sentence, discourse, language). We then conduct a series of experiments to investigate whether NMT systems struggle with maintaining the diversity of such texts, potentially reducing the richness of the language generated by these systems, compared to human translators. We search for potential causes of these limitations rooted in training objectives and decoding algorithms. Our ultimate goal is to develop alternatives that do not enforce uniformity in the distribution of statistical properties in the output and that allow for better global planning of the translation, taking into account the intrinsic ambiguity of the translation task.