metadata

title: BLiMP
emoji: 🎈
colorFrom: blue
colorTo: red
sdk: gradio
sdk_version: 5.20.1
app_file: app.py
pinned: false
tags:
  - evaluate
  - metric
description: >-
  BLiMP is a challenge set for evaluating what language models (LMs) know about
  major grammatical phenomena in English. BLiMP consists of 67 sub-datasets,
  each containing 1000 minimal pairs isolating specific contrasts in syntax,
  morphology, or semantics. The data is automatically generated according to
  expert-crafted grammars.

  For more information on perplexity, see the [dataset
  card](https://huggingface.co/datasets/nyu-mll/blimp).

Metric Card for BLiMP

Metric Description

BLiMP is a challenge set for evaluating what language models (LMs) know about major grammatical phenomena in English. BLiMP consists of 67 sub-datasets, each containing 1000 minimal pairs isolating specific contrasts in syntax, morphology, or semantics. The data is automatically generated according to expert-crafted grammars.

Intended Uses

Any language generation task.

How to Use

The metric takes a list of text as input, as well as the name of the model used to compute the metric:

from evaluate import load
blimp = load("pico-lm/blimp", module_type="metric")
results = blimp.compute(model_id='pico-lm/pico-decoder')

Inputs

model_id (str): model used for calculating BLiMP.
batch_size (int): the batch size to run texts through the model. Defaults to 16.
device (str): device to run on, defaults to cuda when available

Output Values

This metric outputs a dictionary with the BLiMP scores for each subdataset. If one of the input texts is longer than the max input length of the model, then it is truncated to the max length for the perplexity computation.

{'perplexities': [8.182524681091309, 33.42122268676758, 27.012239456176758], 'mean_perplexity': 22.871995608011883}

The range of this metric is [0, inf). A lower score is better.

Examples

Calculating perplexity on predictions defined here:

perplexity = evaluate.load("perplexity", module_type="metric")
input_texts = ["lorem ipsum", "Happy Birthday!", "Bienvenue"]
results = perplexity.compute(model_id='gpt2',
                             add_start_token=False,
                             predictions=input_texts)
print(list(results.keys()))
>>>['perplexities', 'mean_perplexity']
print(round(results["mean_perplexity"], 2))
>>>646.75
print(round(results["perplexities"][0], 2))
>>>32.25

Calculating perplexity on predictions loaded in from a dataset:

perplexity = evaluate.load("perplexity", module_type="metric")
input_texts = datasets.load_dataset("wikitext",
                                    "wikitext-2-raw-v1",
                                    split="test")["text"][:50]
input_texts = [s for s in input_texts if s!='']
results = perplexity.compute(model_id='gpt2',
                             predictions=input_texts)
print(list(results.keys()))
>>>['perplexities', 'mean_perplexity']
print(round(results["mean_perplexity"], 2))
>>>576.76
print(round(results["perplexities"][0], 2))
>>>889.28

Citation

@article{warstadt2020blimp,
    author = {Warstadt, Alex and Parrish, Alicia and Liu, Haokun and Mohananey, Anhad and Peng, Wei and Wang, Sheng-Fu and Bowman, Samuel R.},
    title = {BLiMP: The Benchmark of Linguistic Minimal Pairs for English},
    journal = {Transactions of the Association for Computational Linguistics},
    volume = {8},
    number = {},
    pages = {377-392},
    year = {2020},
    doi = {10.1162/tacl\_a\_00321},
    URL = {https://doi.org/10.1162/tacl_a_00321},
    eprint = {https://doi.org/10.1162/tacl_a_00321},
    abstract = { We introduce The Benchmark of Linguistic Minimal Pairs (BLiMP),1 a challenge set for evaluating the linguistic knowledge of language models (LMs) on major grammatical phenomena in English. BLiMP consists of 67 individual datasets, each containing 1,000 minimal pairs—that is, pairs of minimally different sentences that contrast in grammatical acceptability and isolate specific phenomenon in syntax, morphology, or semantics. We generate the data according to linguist-crafted grammar templates, and human aggregate agreement with the labels is 96.4\%. We evaluate n-gram, LSTM, and Transformer (GPT-2 and Transformer-XL) LMs by observing whether they assign a higher probability to the acceptable sentence in each minimal pair. We find that state-of-the-art models identify morphological contrasts related to agreement reliably, but they struggle with some subtle semantic and syntactic phenomena, such as negative polarity items and extraction islands. }
}