What are the best Systems? New Perspectives on NLP Benchmarking

Overview

What are the best Systems? New Perspectives on NLP Benchmarking

In Machine Learning, a benchmark refers to an ensemble of datasets associated with one or multiple metrics together with a way to aggregate different systems performances. They are instrumental in {\it (i)} assessing the progress of new methods along different axes and {\it (ii)} selecting the best systems for practical use. This is particularly the case for NLP with the development of large pre-trained models (\textit{e.g.} GPT, BERT) that are expected to generalize well on a variety of tasks. While the community mainly focused on developing new datasets and metrics, there has been little interest in the aggregation procedure, which is often reduced to a simple average over various performance measures. However, this procedure can be problematic when the metrics are on a different scale, which may lead to spurious conclusions. This paper proposes a new procedure to rank systems based on their performance across different tasks. Motivated by the social choice theory, the final system ordering is obtained through aggregating the rankings induced by each task and is theoretically grounded. We conduct extensive numerical experiments (on over 270k scores) to assess the soundness of our approach both on synthetic and real scores (\textit{e.g.} GLUE, EXTREM, SEVAL, TAC, FLICKR). In particular, we show that our method yields different conclusions on state-of-the-art systems than the mean-aggregation procedure while being both more reliable and robust.

Authors:

Goal :

This repository deals with automatic evaluation of NLG and addresses the special case of reference based evaluation. The goal is to build a metric m: where is the space of sentences. An example is given below:

Overview

Limitations of Mean Aggregation

Counter Example

Kemeny Conscensus based Aggregation

Kemeny Conscensus

Aggregation when Task Level Information is available

Fig1. Production value and quantity of the 10 top commodities

SuperGLUE

XTREM

Toy Data

Toy Example

Aggregation when Instance Level Information is available

draw_table_2-1

two_level_ranking_DIALOG_pc csv-1 two_level_ranking_FLICKR csv-1 two_level_ranking_REAL_SUM csv-1

Reproducing the paper results

See notebooks.

References

If you find this repo useful, please cite our papers:

@article{,
  title={},
  author={},
  journal={},
  year={2022}
}

Usage

Python Function

Running our ranking is require a simple cpu.

We provide example inputs under <>.py. For example for BaryScore


Command Line Interface (CLI)

We provide a command line interface (CLI) of BERTScore as well as a python module. For the CLI, you can use it as follows:

# TASK LEVEL INFORMATION
export PATH_TO_DF_TO_RANK=sample_df/glue.csv
export MODE=task_level

python ranking_cli.py --df_to_rank=$PATH_TO_DF_TO_RANK --mode=$MODE

# INSTANCE LEVEL INFORMATION
export PATH_TO_DF_TO_RANK=sample_df/TAC_08.csv
export MODE=instance_level
python ranking_cli.py --df_to_rank=$PATH_TO_DF_TO_RANK --mode=$MODE

See more options by python score_cli.py -h.

Acknowledgements

This work was granted access to the HPC resources of IDRIS under the allocation 2021- 101838 made by GENCI. Nathan is funded by the projet ANR LIMPID.

Owner
Pierre Colombo
Pierre Colombo
Unsupervised intent recognition

INTENT author: steeve LAQUITAINE description: deployment pattern: currently batch only Setup & run git clone https://github.com/slq0/intent.git bash

sl 1 Apr 08, 2022
This codebase facilitates fast experimentation of differentially private training of Hugging Face transformers.

private-transformers This codebase facilitates fast experimentation of differentially private training of Hugging Face transformers. What is this? Why

Xuechen Li 73 Dec 28, 2022
Extract city and country mentions from Text like GeoText without regex, but FlashText, a Aho-Corasick implementation.

flashgeotext ⚡ 🌍 Extract and count countries and cities (+their synonyms) from text, like GeoText on steroids using FlashText, a Aho-Corasick impleme

Ben 57 Dec 16, 2022
🎐 a python library for doing approximate and phonetic matching of strings.

jellyfish Jellyfish is a python library for doing approximate and phonetic matching of strings. Written by James Turk James Turk 1.8k Dec 21, 2022

A fast and lightweight python-based CTC beam search decoder for speech recognition.

pyctcdecode A fast and feature-rich CTC beam search decoder for speech recognition written in Python, providing n-gram (kenlm) language model support

Kensho 315 Dec 21, 2022
基于pytorch_rnn的古诗词生成

pytorch_peot_rnn 基于pytorch_rnn的古诗词生成 说明 config.py里面含有训练、测试、预测的参数,更改后运行: python main.py 预测结果 if config.do_predict: result = trainer.generate('丽日照残春')

西西嘛呦 3 May 26, 2022
Sample data associated with the Aurora-BP study

The Aurora-BP Study and Dataset This repository contains sample code, sample data, and explanatory information for working with the Aurora-BP dataset

Microsoft 16 Dec 12, 2022
Stanford CoreNLP provides a set of natural language analysis tools written in Java

Stanford CoreNLP Stanford CoreNLP provides a set of natural language analysis tools written in Java. It can take raw human language text input and giv

Stanford NLP 8.8k Jan 07, 2023
Search msDS-AllowedToActOnBehalfOfOtherIdentity

前言 现在进行RBCD的攻击手段主要是搜索mS-DS-CreatorSID,如果机器的创建者是我们可控的话,那就可以修改对应机器的msDS-AllowedToActOnBehalfOfOtherIdentity,利用工具SharpAllowedToAct-Modify 那我们索性也试试搜索所有计算机

Jumbo 26 Dec 05, 2022
Contains the code and data for our #ICSE2022 paper titled as "CodeFill: Multi-token Code Completion by Jointly Learning from Structure and Naming Sequences"

CodeFill This repository contains the code for our paper titled as "CodeFill: Multi-token Code Completion by Jointly Learning from Structure and Namin

Software Analytics Lab 11 Oct 31, 2022
A toolkit for document-level event extraction, containing some SOTA model implementations

Document-level Event Extraction via Heterogeneous Graph-based Interaction Model with a Tracker Source code for ACL-IJCNLP 2021 Long paper: Document-le

84 Dec 15, 2022
SAVI2I: Continuous and Diverse Image-to-Image Translation via Signed Attribute Vectors

SAVI2I: Continuous and Diverse Image-to-Image Translation via Signed Attribute Vectors [Paper] [Project Website] Pytorch implementation for SAVI2I. We

Qi Mao 44 Dec 30, 2022
Library of deep learning models and datasets designed to make deep learning more accessible and accelerate ML research.

Tensor2Tensor Tensor2Tensor, or T2T for short, is a library of deep learning models and datasets designed to make deep learning more accessible and ac

12.9k Jan 07, 2023
Refactored version of FastSpeech2

Refactored version of FastSpeech2. An implementation of Microsoft's "FastSpeech 2: Fast and High-Quality End-to-End Text to Speech"

ILJI CHOI 10 May 26, 2022
Model parallel transformers in JAX and Haiku

Table of contents Mesh Transformer JAX Updates Pretrained Models GPT-J-6B Links Acknowledgments License Model Details Zero-Shot Evaluations Architectu

Ben Wang 4.9k Jan 04, 2023
A Multilingual Latent Dirichlet Allocation (LDA) Pipeline with Stop Words Removal, n-gram features, and Inverse Stemming, in Python.

Multilingual Latent Dirichlet Allocation (LDA) Pipeline This project is for text clustering using the Latent Dirichlet Allocation (LDA) algorithm. It

Artifici Online Services inc. 74 Oct 07, 2022
KoBERT - Korean BERT pre-trained cased (KoBERT)

KoBERT KoBERT Korean BERT pre-trained cased (KoBERT) Why'?' Training Environment Requirements How to install How to use Using with PyTorch Using with

SK T-Brain 1k Jan 02, 2023
A 30000+ Chinese MRC dataset - Delta Reading Comprehension Dataset

Delta Reading Comprehension Dataset 台達閱讀理解資料集 Delta Reading Comprehension Dataset (DRCD) 屬於通用領域繁體中文機器閱讀理解資料集。 本資料集期望成為適用於遷移學習之標準中文閱讀理解資料集。 本資料集從2,108篇

272 Dec 15, 2022
Extracting Summary Knowledge Graphs from Long Documents

GraphSum This repo contains the data and code for the G2G model in the paper: Extracting Summary Knowledge Graphs from Long Documents. The other basel

Zeqiu (Ellen) Wu 10 Oct 21, 2022
Pangu-Alpha for Transformers

Pangu-Alpha for Transformers Usage Download MindSpore FP32 weights for GPU from here to data/Pangu-alpha_2.6B.ckpt Activate MindSpore environment and

One 5 Oct 01, 2022