Zobrazeno 1 - 10
of 282
pro vyhledávání: '"Wingate, David"'
Sparse Autoencoders (SAEs) are a promising approach for extracting neural network representations by learning a sparse and overcomplete decomposition of the network's internal activations. However, SAEs are traditionally trained considering only acti
Externí odkaz:
http://arxiv.org/abs/2411.10397
Autor:
Rytting, Christopher Michael, Sorensen, Taylor, Argyle, Lisa, Busby, Ethan, Fulda, Nancy, Gubler, Joshua, Wingate, David
Researchers often rely on humans to code (label, annotate, etc.) large sets of texts. This kind of human coding forms an important part of social science research, yet the coding process is both resource intensive and highly variable from application
Externí odkaz:
http://arxiv.org/abs/2306.02177
Autor:
Argyle, Lisa P., Busby, Ethan, Gubler, Joshua, Bail, Chris, Howe, Thomas, Rytting, Christopher, Wingate, David
A rapidly increasing amount of human conversation occurs online. But divisiveness and conflict can fester in text-based interactions on social media platforms, in messaging apps, and on other digital forums. Such toxicity increases polarization and,
Externí odkaz:
http://arxiv.org/abs/2302.07268
While large language models (LLMs) like GPT-3 have achieved impressive results on multiple choice question answering (MCQA) tasks in the zero, one, and few-shot settings, they generally lag behind the MCQA state of the art (SOTA). MCQA tasks have tra
Externí odkaz:
http://arxiv.org/abs/2210.12353
We explore the idea of compressing the prompts used to condition language models, and show that compressed prompts can retain a substantive amount of information about the original prompt. For severely compressed prompts, while fine-grained informati
Externí odkaz:
http://arxiv.org/abs/2210.03162
Autor:
Argyle, Lisa P., Busby, Ethan C., Fulda, Nancy, Gubler, Joshua, Rytting, Christopher, Wingate, David
We propose and explore the possibility that language models can be studied as effective proxies for specific human sub-populations in social science research. Practical and research applications of artificial intelligence tools have sometimes been li
Externí odkaz:
http://arxiv.org/abs/2209.06899
Autor:
Sorensen, Taylor, Robinson, Joshua, Rytting, Christopher Michael, Shaw, Alexander Glenn, Rogers, Kyle Jeffrey, Delorey, Alexia Pauline, Khalil, Mahmoud, Fulda, Nancy, Wingate, David
Pre-trained language models derive substantial linguistic and factual knowledge from the massive corpora on which they are trained, and prompt engineering seeks to align these models to specific tasks. Unfortunately, existing prompt engineering metho
Externí odkaz:
http://arxiv.org/abs/2203.11364
Large natural language models (such as GPT-3 or T5) demonstrate impressive abilities across a range of general NLP tasks. Here, we show that the knowledge embedded in such models provides a useful inductive bias, not just on traditional NLP tasks, bu
Externí odkaz:
http://arxiv.org/abs/2110.02370
Publikováno v:
Neural Information Processing Systems 33 (2020) 17416-17428
It is notoriously difficult to control the behavior of artificial neural networks such as generative neural language models. We recast the problem of controlling natural language generation as that of learning to interface with a pretrained language
Externí odkaz:
http://arxiv.org/abs/2012.05983
Autor:
Dawkins, Laura C., Brown, Kate, Bernie, Dan J., Lowe, Jason A., Economou, Theodoros, Grassie, Duncan, Schwartz, Yair, Godoy-Shimizu, Daniel, Korolija, Ivan, Mumovic, Dejan, Wingate, David, Dyer, Emma
Publikováno v:
In Climate Risk Management 2024 44
