Typology of risks of generative text-to-image models
This paper investigates the direct risks and harms associated with modern text-to-image
generative models, such as DALL-E and Midjourney, through a comprehensive literature …
generative models, such as DALL-E and Midjourney, through a comprehensive literature …
Goodtriever: Adaptive toxicity mitigation with retrieval-augmented models
Considerable effort has been dedicated to mitigating toxicity, but existing methods often
require drastic modifications to model parameters or the use of computationally intensive …
require drastic modifications to model parameters or the use of computationally intensive …
Introducing v0. 5 of the ai safety benchmark from mlcommons
This paper introduces v0. 5 of the AI Safety Benchmark, which has been created by the
MLCommons AI Safety Working Group. The AI Safety Benchmark has been designed to …
MLCommons AI Safety Working Group. The AI Safety Benchmark has been designed to …
Responsible AI Considerations in Text Summarization Research: A Review of Current Practices
AI and NLP publication venues have increasingly encouraged researchers to reflect on
possible ethical considerations, adverse impacts, and other responsible AI issues their work …
possible ethical considerations, adverse impacts, and other responsible AI issues their work …
[PDF][PDF] Undesirable biases in NLP: Averting a crisis of measurement
O Van der Wal, D Bachmann, A Leidinger… - arXiv preprint arXiv …, 2022 - pure.uva.nl
Abstract As Large Language Models and Natural Language Processing (NLP) technology
rapidly develops and spreads into daily life, it becomes crucial to anticipate how its use …
rapidly develops and spreads into daily life, it becomes crucial to anticipate how its use …
The gaps between pre-train and downstream settings in bias evaluation and debiasing
The output tendencies of Pre-trained Language Models (PLM) vary markedly before and
after Fine-Tuning (FT) due to the updates to the model parameters. These divergences in …
after Fine-Tuning (FT) due to the updates to the model parameters. These divergences in …
Fft: Towards harmlessness evaluation and analysis for llms with factuality, fairness, toxicity
The widespread of generative artificial intelligence has heightened concerns about the
potential harms posed by AI-generated texts, primarily stemming from factoid, unfair, and …
potential harms posed by AI-generated texts, primarily stemming from factoid, unfair, and …
Healthy immigrant effect or under-detection? Examining undiagnosed and unrecognized late-life depression for racialized immigrants and nonimmigrants in Canada
S Lin - The Journals of Gerontology: Series B, 2024 - academic.oup.com
Abstract Objectives Immigrants to Canada tend to have a lower incidence of diagnosed
depression than nonimmigrants. One theory suggests that this “healthy immigrant effect …
depression than nonimmigrants. One theory suggests that this “healthy immigrant effect …
Mind vs. Mouth: On Measuring Re-judge Inconsistency of Social Bias in Large Language Models
Y Zhao, B Wang, D Zhao, K Huang, Y Wang… - arXiv preprint arXiv …, 2023 - arxiv.org
Recent researches indicate that Pre-trained Large Language Models (LLMs) possess
cognitive constructs similar to those observed in humans, prompting researchers to …
cognitive constructs similar to those observed in humans, prompting researchers to …
“One-Size-Fits-All”? Examining Expectations around What Constitute “Fair” or “Good” NLG System Behaviors
Fairness-related assumptions about what constitute appropriate NLG system behaviors
range from invariance, where systems are expected to behave identically for social groups …
range from invariance, where systems are expected to behave identically for social groups …