-
TESS: Text-to-Text Self-Conditioned Simplex Diffusion
Diffusion models have emerged as a power-ful paradigm for generation, obtaining strong performance in various continuous domains. However, applying continuous diffusion models... -
How do large language models capture the ever-changing world knowledge?
This paper presents a review of recent advances in large language models' ability to capture ever-changing world knowledge. -
Masked Acoustic Unit for Mispronunciation Detection and Correction
The proposed method uses the acoustic unit (AU) as the intermediary feature for both mispronunciation detection and correction. -
Language models are few-shot learners
A language model that demonstrates capabilities in processing and generating human-like text. -
Self-Supervised Alignment with Mutual Information
The dataset is used for training a language model to follow behavioral principles without the use of preference labels, demonstrations, or human oversight. -
NLPositionality
NLPositionality is a framework for characterizing design biases and quantifying the positionality of NLP datasets and models. -
Anthropic Persona Dataset
The Persona dataset contains 99 different personas, each entailing 500 statements that align and 500 statements that disagree with the persona trait. -
LyricCanvas
The LyricCanvas dataset is a large-scale collection of lyrics with noisy visual descriptions that represent their implicit meaning. -
English and Luganda datasets for ASR-free keyword spotting
South African English and Luganda datasets -
Feature learning for efficient ASR-free keyword spotting in low-resource lang...
ASR-free keyword spotting in low-resource languages -
Tensor Trust Dataset
A dataset of prompt injection attacks for evaluating the effectiveness of Tensor Trust in detecting prompt injection attacks. -
SPML Dataset
A dataset of system prompts and user prompts for evaluating the effectiveness of SPML in detecting prompt injection attacks. -
Natural Questions
The Natural Questions dataset consists of questions extracted from web queries, with each question accompanied by a corresponding Wikipedia article containing the answer. -
Examining the State-of-the-Art in News Timeline Summarization
Examining the state-of-the-art in news timeline summarization.
