StylePTB: A Compositional Benchmark for Fine-grained Controllable Text Style Transfer

Lyu, Yiwei; Liang, Paul Pu; Pham, Haï Vu; Hovy, Eduard; Póczos, Barnabás; Salakhutdinov, Ruslan; Morency, Louis–Philippe

doi:10.48550/arxiv.2104.05196

Cited by 3 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…deleting peripheral information irrelevant to the target domain) (Alva-Manchego et al 2020). Our approach is similar to the textto-text models that exert controllability using task-specific prompts (Keskar et al 2019;Dathathri et al 2019;Kariuk and Karamshuk 2020;Brown et al 2020;Reif et al 2021;Lyu et al 2021;Xu, Peng, and Liu 2022).…”

Section: Related Workmentioning

confidence: 99%

Med-EASi: Finely Annotated Dataset and Models for Controllable Simplification of Medical Texts

Basu

Rosni

Yasunaga

et al. 2023

AAAI

View full text Add to dashboard Cite

Automatic medical text simplification can assist providers with patient-friendly communication and make medical texts more accessible, thereby improving health literacy. But curating a quality corpus for this task requires the supervision of medical experts. In this work, we present Med-EASi (Medical dataset for Elaborative and Abstractive Simplification), a uniquely crowdsourced and finely annotated dataset for supervised simplification of short medical texts. Its expert-layman-AI collaborative annotations facilitate controllability over text simplification by marking four kinds of textual transformations: elaboration, replacement, deletion, and insertion. To learn medical text simplification, we fine-tune T5-large with four different styles of input-output combinations, leading to two control-free and two controllable versions of the model. We add two types of controllability into text simplification, by using a multi-angle training approach: position-aware, which uses in-place annotated inputs and outputs, and position-agnostic, where the model only knows the contents to be edited, but not their positions. Our results show that our fine-grained annotations improve learning compared to the unannotated baseline. Furthermore, our position-aware control enhances the model's ability to generate better simplification than the position-agnostic version. The data and code are available at https://github.com/Chandrayee/CTRL-SIMP.

show abstract

Section: Related Workmentioning

confidence: 99%

Med-EASi: Finely Annotated Dataset and Models for Controllable Simplification of Medical Texts

Basu

Rosni

Yasunaga

et al. 2023

AAAI

View full text Add to dashboard Cite

show abstract

“…(Frieder et al 2023;Sobania et al 2023;Srivastava et al 2022). The seeming breadth of knowledge capabilities, combined with excellent grammar and fluency, some evidence of reasoning capabilities (Brown et al 2020;Qin et al 2023;Wei et al 2022b;Yao et al 2023), as well as the ability to change styles and personas (Deshpande et al 2023;Lyu et al 2021), have led some to think that AGI will soon be upon us (Bubeck et al 2023;Zhang et al 2023).…”

Section: Why Embodiment Is Key For Agimentioning

confidence: 99%

The Path to AGI Goes through Embodiment

Tan,

Jaiswal

2023

AAAI-SS

View full text Add to dashboard Cite

Recent advances in large language models have raised the question of whether these language models alone could lead to artificial general intelligence (AGI). In this short position essay, we argue that embodiment is not only required for achieving AGI, but also that embodiment is the key to convincingly demonstrate AGI capabilities. There is no single widely-accepted, objective test for AGI, so therefore whether a system has achieved AGI is a subjective judgement. We argue that a language-only system or one that cannot demonstrate success in the real world would not be convincing.

show abstract

Med-EASi: Finely Annotated Dataset and Models for Controllable Simplification of Medical Texts

Basu¹,

Rosni²,

Yasunaga³

et al. 2023

Preprint

View full text Add to dashboard Cite

Automatic medical text simplification can assist providers with patient-friendly communication and make medical texts more accessible, thereby improving health literacy. But curating a quality corpus for this task requires the supervision of medical experts. In this work, we present Med-EASi (Medical dataset for Elaborative and Abstractive Simplification), a uniquely crowdsourced and finely annotated dataset for supervised simplification of short medical texts. Its expert-layman-AI collaborative annotations facilitate controllability over text simplification by marking four kinds of textual transformations: elaboration, replacement, deletion, and insertion. To learn medical text simplification, we fine-tune T5-large with four different styles of inputoutput combinations, leading to two control-free and two controllable versions of the model. We add two types of controllability into text simplification, by using a multi-angle training approach: position-aware, which uses in-place annotated inputs and outputs, and position-agnostic, where the model only knows the contents to be edited, but not their positions. Our results show that our fine-grained annotations improve learning compared to the unannotated baseline. Furthermore, position-aware control generates better simplification than the position-agnostic one. The data and code are available at https://github.com/Chandrayee/CTRL-SIMP.

show abstract

StylePTB: A Compositional Benchmark for Fine-grained Controllable Text Style Transfer

Cited by 3 publications

References 21 publications

Med-EASi: Finely Annotated Dataset and Models for Controllable Simplification of Medical Texts

Med-EASi: Finely Annotated Dataset and Models for Controllable Simplification of Medical Texts

The Path to AGI Goes through Embodiment

Med-EASi: Finely Annotated Dataset and Models for Controllable Simplification of Medical Texts

Contact Info

Product

Resources

About