paper-with-me

Papers

What's in the Box? A Preliminary Analysis of Undesirable Content in the Common Crawl Corpus

2021-05-06 · Alexandra Sasha Luccioni, Joseph D. Viviano

Whereas much of the success of the current generation of neural language models has been driven by increasingly large training corpora, relatively little research has been dedicated to analyzing these massive sources of textual data. In this exploratory analysis, we delve deeper into the Common Crawl, a colossal web corpus that is extensively used for training language models. We find that it contains a significant amount of undesirable content, including hate speech and sexually explicit content, even after filtering procedures. We discuss the potential impacts of this content on language models and conclude with future research directions and a more mindful approach to corpus collection and analysis.

📄 PDF Abstract BibTeX arXiv:2105.02732

Code (1)

josephdviviano/whatsinthebox 공식 구현

Similar Papers 제목 키워드 기반

What's in the Box? An Analysis of Undesirable Content in the Common Crawl Corpus

2021-08-01 · ACL 2021 5 · Alexandra Luccioni, Joseph Viviano

Whereas much of the success of the current generation of neural language models has been driven by increasingly large training corpora, relatively little research has been dedicated to analyzing these massive sources of …

Epistemic Injustice in Language Models: An Audit of Pretraining Filters and Guardrails

2026-06-04 · Marco Antonio Stranisci, A Pranav, Rossana Damiano, Christian Hardmeier 외 arxiv

Modern language models rely on pretraining filters to remove undesirable content from training corpora and inference-time guardrails to suppress undesirable outputs during deployment. In this paper, we examine how these …

On the probability-quality paradox in language generation

2022-03-31 · Clara Meister, Gian Wiher, Tiago Pimentel, Ryan Cotterell

When generating natural language from neural probabilistic models, high probability does not always coincide with high quality: It has often been observed that mode-seeking decoding methods, i.e., those that produce high…

Text Generation

Building a User-Generated Content North-African Arabizi Treebank: Tackling Hell

2020-07-01 · ACL 2020 6 · Djam{\'e} Seddah, Farah Essaidi, Amal Fethi, Matthieu Futeral 외

We introduce the first treebank for a romanized user-generated content variety of Algerian, a North-African Arabic dialect known for its frequent usage of code-switching. Made of 1500 sentences, fully annotated in morpho…

Dependency ParsingPOSPOS TaggingSentence+1

Risk-Aware Distributional Intervention Policies for Language Models

2025-01-27 · Bao Nguyen, Binh Nguyen, Duy Nguyen, Viet Anh Nguyen

Language models are prone to occasionally undesirable generations, such as harmful or toxic content, despite their impressive capability to produce texts that appear accurate and coherent. This paper presents a new two-s…