-
OpenAssistant Conversations– Democratizing Large Language Model Alignment
OpenAssistant Conversations– Democratizing Large Language Model Alignment -
Sparse Watermarking in LLMs with Enhanced Text Quality
The dataset used in the paper is not explicitly described, but it is mentioned that the authors used the ELI5, FinanceQA, MultiNews, and QMSum datasets. -
Universal and transferable adversarial attacks on aligned language models
AdvBench is a dataset for evaluating the safety of large language models. -
Hate Speech Detection using Large Language Models
The dataset used for probing LLMs for hate speech detection, including HateXplain, implicit hate, and ToxicSpans datasets. -
TruthX: Alleviating Hallucinations by Editing Large Language Models
TruthX: Alleviating Hallucinations by Editing Large Language Models -
Inducing Anxiety in Large Language Models Increases Exploration and Bias
The Inducing Anxiety in Large Language Models Increases Exploration and Bias dataset contains anxiety-inducing scenarios for large language models. -
MoralChoice
The MoralChoice survey dataset contains 1767 moral decision-making scenarios. Every moral scenario consists of a triple (context, action 1, action 2) and a set of auxiliary labels. -
Jailbreaking Large Language Models
The dataset used in the paper is a jailbreaking dataset for large language models, containing pairs of provocative questions and their corresponding affirmative responses. -
Xiezhi Benchmark
Xiezhi comprises multiple-choice questions across 516 diverse disciplines ranging from 13 different subjects with 249,587 questions and accompanied by Xiezhi-Specialty with... -
Xiezhi: An Ever-Updating Benchmark for Holistic Domain Knowledge Evaluation
New Natural Language Process (NLP) benchmarks are urgently needed to align with the rapid development of large language models (LLMs). We present Xiezhi, the most comprehensive... -
Conceptual Captions
The dataset used in the paper "Scaling Laws of Synthetic Images for Model Training". The dataset is used for supervised image classification and zero-shot classification tasks. -
GraphEval2000
GraphEval2000 is a graph dataset designed to evaluate the graph reasoning abilities of large language models (LLMs) through coding challenges. -
Free-Bloom: Zero-Shot Text-to-Video Generator
Text-to-video is a rapidly growing research area that aims to generate a semantic, identical, and temporal coherence sequence of frames that accurately align with the input text... -
VideoStreaming
A novel approach to tackle the complexities of long video understanding with large language models (LLMs). Our proposed memory-propagated streaming encoding architecture... -
TOPA: Extend Large Language Models for Video Understanding via Text-Only Pre-...
TOPA is a text-only pre-alignment framework for extending large language models for video understanding without the need for pre-training on real video data. -
Orca: Progressive Learning from Complex Explanation Traces
The Orca approach involves leveraging explanation tuning to generate detailed responses from a large language model. -
Evol-Instruct: A Pipeline for Automatically Evolving Instruction Datasets
The Evol-Instruct pipeline involves automatically evolving instruction datasets using large language models.
