-
EC AI Platform
The dataset used in the paper is not explicitly described, but it is mentioned that the authors evaluated GPT-4 against three applications built with the EC AI platform for... -
FP6-LLM: Efficiently Serving Large Language Models Through FP6-Centric Algori...
Six-bit quantization can effectively reduce the size of large language models and preserve the model quality consistently across varied applications. -
LDC2014T12
The dataset used in the paper is the Linguistic Data Consortium AMR corpus release 1.0 (LDC2014T12), consisting of 13,050 AMR/English sentence pairs. -
Exponential Family Embeddings
Word embeddings are a powerful approach for capturing semantic similarity among terms in a vocabulary. In this paper, we develop exponential family embeddings, a class of... -
Towards Improving Selective Prediction Ability of NLP Systems
SNLI, MNLI, Stress Test, Matched Mismatched, Competence, Distraction, and Noise datasets -
Neural Language Correction with Character-Based Attention
Neural language correction with character-based attention. -
Stanford Neural Machine Translation Systems for Spoken Language Domain
Stanford neural machine translation systems for spoken language domain. -
Corpora Generation for Grammatical Error Correction
Two approaches for generating large parallel datasets for Grammatical Error Correction (GEC) using publicly available Wikipedia data. -
MNLI, QQP, and SST-2
The dataset used in this paper consists of three tasks: Multi-Genre Natural Language Inference (MNLI), Quora Question Pairs (QQP), and Stanford Sentiment Treebank (SST-2). -
Are Larger Pretrained Language Models Uniformly Better? Comparing Performance...
Larger language models have higher accu- racy on average, but are they better on ev- ery single instance (datapoint)? -
Learning to summarize with human feedback
The paper presents a study on the impact of synthetic data on large language models (LLMs) and proposes a method to steer LLMs towards desirable non-differentiable attributes. -
Reward Model Ensembles
The authors used three datasets: TL;DR, HELPFULNESS, and XSUM/NLI. -
STAMP 4 NLP
STAMP 4 NLP is an instantiable, iterative, and incremental process model for developing natural language processing applications with a focus on quality, business value, and... -
Detecting Hallucinated Content in Conditional Neural Sequence Generation
Neural sequence models can generate highly fluent sentences, but recent studies have also shown that they are also prone to hallucinate additional content not supported by the... -
A general theoretical paradigm to understand learning from human preferences
The paper proposes a novel approach to aligning language models with human preferences, focusing on the use of preference optimization in reward-free RLHF. -
Llama: Open and efficient foundation language models
The LLaMA dataset is a large language model dataset used in the paper. -
Mixtral of Experts
The dataset used in the paper for instruction following task -
Toward an Architecture for Never-ending Language Learning
Toward an Architecture for Never-ending Language Learning. -
MISMATCH: Fine-grained Evaluation of Machine-generated Text
The dataset used in the paper for fine-grained evaluation of machine-generated text with mismatch error types. -
BERT: Pre-training of deep bidirectional transformers for language understanding
This paper proposes BERT, a pre-trained deep bidirectional transformer for language understanding.
