idx
int64
20k
20.7k
premise
stringlengths
16
707
hypothesis
stringlengths
17
336
label
class label
0
2
20,035
The forecasting challenges retailers confront have been amplified in recent years by product proliferation in almost every category .
Forecasting has been easier recently due to the updated process we have today .
2contradiction
20,036
After a lifetime of trials , Donna not only earned her GED at Goodwill , she earned a job here .
Donna went through Goodwill to get her GED , but was still unemployed .
2contradiction
20,037
The hijackers had planned to take flights scheduled to depart at 7:45 ( American 11 ) , 8:00 ( United 175 and United 93 ) , and 8:10 ( American 77 ) .
The hijackers planned to take late flights because fewer people would be on board .
2contradiction
20,038
The most recent hijacking that involved U.S. air traffic controllers , FAA management , and military coordination had occurred in 1993 .
The last hijacking that involved military coordination had been in May of 1993 .
1neutral
20,039
Wha , yeah , very hard compared to high school .
I much preferred the easy ride of high school .
1neutral
20,040
For example , close to 50 percent of the value of shipments for men 's shirts comes from the cost of materials , while 25 percent arises from compensation costs .
90 percent of the value of shipments for men 's shirts comes from compensation costs .
2contradiction
20,041
Of those beginning prenatal care in the first trimester , Indiana mothers rank in the lower third of individuals receiving such care nationwide .
There are Indiana mothers who do not receive prenatal care in the first trimester .
0entailment
20,042
Did you have lots of storybooks at your house that were read to you ?
Were a lot of the books in your house read to you ?
0entailment
20,043
Basic items also represent a major percentage of all apparel goods sold .
Prom dresses make up the largest percentage of apparel goods sold .
2contradiction
20,044
Therefore , it is not surprising that most replenishment products are basics or fashion-basics with relatively stable Manufacturers are unwilling to hold speculative stock to meet replenishment requests from retailers for fashion products because the risk of holding those fashion goods in finished goods inventory is too high .
Finished goods inventory has a very direct bearing on a Manufacturers ' bottom line .
1neutral
20,045
For example , increasing the number of products offered in a category by a factor of four ( say from fifty items to 200 ) without increasing total demand in the category would increase the coefficient of variation for each individual product by a factor of two .
An increasing number of products offered in a category will definitely not increase the total demand for the item in the category .
1neutral
20,046
It is a pivotal time in the life of a young girl-when small changes in her behavior , or the choices she makes , can affect her future .
It is important to help young girls with the choices that they make .
0entailment
20,047
But , um , It 's not like I 'll never come back and visit or anything like that .
I will never come back and visit .
2contradiction
20,048
But , um , It 's not like I 'll never come back and visit or anything like that .
I will likely come back and visit .
2contradiction
20,049
Perhaps $ 100 is more to your liking or , perhaps , you would be interested in making several smaller gifts during the year .
If you give us $ 100 now , we wo n't ever ask you to give another gift again .
1neutral
20,050
Um , Can we turn that off while I think ?
Can we turn that on while I think ?
2contradiction
20,051
But the power of words is such that I contend their impact is totally lost if not treated somehow , there is a difference between Someone had written shit on the wall and Some asshole had written shit on the wall : the writer of the former has more credibility ; the writer of the latter was , very likely , the very asshole who had written shit on the wall .
A girl wrote some shit on the wall .
1neutral
20,052
Denouncing waywardness among the faithful , some clerics have appealed for a return to observance of the literal teachings of the Qur'an and Hadith .
The clerics have made their best efforts to stress the importance of taking the passages in the books with a grain of salt .
2contradiction
20,053
By the late 1980s , a growing number of retailers had started changing the way they did business .
Retailers began selling different stuff in the late 1980s .
1neutral
20,054
to which the other replied , Yes , put it in it , and pull it down .
During a long conversation , one person stated `` Yes , put it in it , and pull down . ''
1neutral
20,055
It would be so damn foggy you could not see the yellow line on that divided highway in front of you and we to turn left out onto this highway .
The fog did not allow us to turn right .
1neutral
20,056
He worked , he 's a teacher , and at that time he worked as the principal of that school , of that school , because it was a , like a high school , there was , from first ( grade ) to high school .
The man is a teacher , but also a principal of a place of a learning .
0entailment
20,057
Would you say that um , your mom or you dad or whoever is telling the story , tells that part of the story , the , the last part when you came to the door and she saw the glow of light , would you say that they tell that part of the story more than any other part or do they tell the whole thing from beginning to end ?
Do you think that your mom and dad feel more sad or angry when telling the story ?
1neutral
20,058
I moved to San Diego .
I was born and raised in San Diego , but moved somewhere else .
2contradiction
20,059
Now I am writing to ask if you would continue your support with a further donation for 1990 .
You donated in the past , and so we would like you to donate even more in the future .
0entailment
20,060
Nobody who knew him would have expected him to do anything else , whether he had been sent away in disgrace or not'but , as it was thought he had left the country , his return did have a dramatic quality .
Nobody he knew expected him to do anything else by he surprised them .
0entailment
20,061
Therefore , I stand uncorrected .
Thus , I have been shown my ideas are without correction and am in the right .
0entailment
20,062
We are the only statewide not-for-profit organization whose primary mission is child abuse prevention .
There are many not-for-profit organizations whose primary mission is child abuse prevention .
2contradiction
20,063
That fund would allow us to grant a $ 500 scholarship annually to meritorious applicants with financial need .
Resources in that account will give us the opportunity to give grants every year to worthy applicants who need help with costs .
0entailment
20,064
Now where did you , when you were hearing these stories , your mother was telling them to you , where were you living at the time ?
Your mother was the one who knew all the stories .
1neutral
20,065
They assumed the fighter escort would be discreet , vectored to a position five miles directly behind the hijacked aircraft , where it could perform its mission to monitor the aircraft 's flight path .
The fighter jets stayed far enough away to track the plane .
0entailment
20,066
Also at 8:25 , and again at 8:29 , Amy Sweeney got through to the American Flight Services Office in Boston but was cut off after she reported someone was hurt aboard the flight .
Amy Sweeney got through to the American Flight Services Office in Boston at 8:25 and again at 8:29 but was disconnected as soon as someone said hello .
2contradiction
20,067
) for the ironic title of his dystopic novel .
For the title to his book about a utopian society is it is non-ironic .
2contradiction
20,068
That is nonsense , of course ; but even if the users remain keenly aware that 1669 is the date only of the earliest written evidence , of what use or importance is that to them ?
For what purpose should the users remember that 1559 is the date of the earliest written proof ?
2contradiction
20,069
Societies seldom see what is central in their own cultures , having much less trouble identifying it in the cultures of others .
Societies often see what is central in their own cultures .
2contradiction
20,070
Considering the circulation of The New York Times on Sundays , his column is probably the most widely read commentary on contemporary English in the world ; that places more than one uncommon burden on a he must do his utmost to be accurate ; he must try to select subjects likely to be of interest to his readers ; and he must write well .
He needs to write well and choose engrossing subjects to his readers .
0entailment
20,071
Because teaching a man to fish will keep him fed for his entire life .
It is better to help a person to become self-sufficient rather than reliant on others
0entailment
20,072
As surveys have shown that etymological information is the least often sought after , making the average user wade through the etymology before getting to the definition has always seemed pointless and irritating to me ; besides , it is unlikely that the serious , consistent seeker of etymological information is likely to use anything but the OED or a major etymological dictionary .
Surveys have shown that the definition is the information least sought after .
2contradiction
20,073
By 1920 , over half of the spinning and weaving capacity was in the South , leading industrialization there .
Most of the weaving capacity was in the South .
0entailment
20,074
The YMCA of the USA also named it an Outstanding International Program in 1997 .
In 1994 , the YMCA of the USA also named it an Outstanding International Program .
2contradiction
20,075
But stories I remember hearing one of uh , I believe it was Marsha uh , the wind caught the storm door there she was , she was uh , hanging on to the back storm door and I guess pulled it open a little bit to see what was going on and the wind caught it and threw her right out in the yard .
The best story about storms I have ever heard was about Marsha getting thrown out into the yard when the wind caught the door she was opening .
0entailment
20,076
It has been a busy time , for we have an ambitious agenda for growth and development .
It 's been a very slow time because everything is the same .
2contradiction
20,077
We 've helped people break free of the welfare cycle .
People have gotten off of welfare because of our help .
0entailment
20,078
Prior to 9/11 , it was understood that an order to shoot down a commercial aircraft would have to be issued by the National Command Authority ( a phrase used to describe the president and secretary of defense ) .
Aircrafts ca n't be shot down by just anyone .
1neutral
20,079
Groundhog hole found the front tire of the motorcycle and I managed to fly over the handlebars .
I fell off my motorcycle due to a hole made by a groundhog .
0entailment
20,080
The thin plastic sheet effectively cuts off the flow of room air through the fabric .
Room air can destroy the fabric if it 's not controlled .
1neutral
20,081
Perhaps the Proximity people thought that they had got round that little problem by giving the same definition for each of the items in the list ; but we know that only very rarely are two synonyms bi-unique ( which is another way of saying that if A = B , B does not necessarily always equal A ) , an ineluctable fact of language .
Just because x equals y , you ca n't assume that y and x are always identical .
0entailment
20,082
We are pleased to tell you of a very exciting development with the fund , which has reached a market value of $ 750,000 .
The fund now has a market value of three dollars and fifty nine cents .
2contradiction
20,083
Why would n't you want to stay on Charlotte ?
What 's the reason for your refusal to stay in Charlotte ?
0entailment
20,084
This is demonstrated in a variety of a shared literacy and the assumptions and biases that go with it ; an awareness of what books are for ; consciousness and use of innumerable cultural allusions and educated idioms ; and a relative ease in reading a periodical called VERBATIM , with sections called OBITER DICTA and EPISTOLAE .
VERBATIM was an article written once in the New York Times in the late 1960s .
2contradiction
20,085
I always had to get up early , milk the cows , and , uh , run , run them , as we say , because it 's a , to the pastures , until times got pretty bad , and one day , I sent my daughter to , to the pasture to bring in the cows .
I seldom wake up early as there is much for me to do .
2contradiction
20,086
In the apt description of one witness , It drops below the radar screen and it 's just continually hovering in your imagination ; you do n't know where it is or what happens to it .
And in that moment , a sudden realization sets in , and your stomach starts to sink .
1neutral
20,087
I read everything I could get my hands on .
There was nothing I could n't get my hands on .
1neutral
20,088
The Command Center executed the request and , a minute later , Cleveland Center reported that United 93 may have a bomb on board .
The Command Center executed the request for the fighter pilots to search for a lost plane and a minute later the Cleveland Center reported that United 93 may have a bomb on board .
1neutral
20,089
Lower-level officials improvised-for example , the FAA 's Boston Center bypassed the chain of command and directly contacted NEADS after the first hijacking .
When the chain of command was circumvented by the FAA 's Boston Center contacting NEADS directly , no one stood in their way .
1neutral
20,090
Nobody who knew him would have expected him to do anything else , whether he had been sent away in disgrace or not'but , as it was thought he had left the country , his return did have a dramatic quality .
Nobody he knew expected him to do anything else .
0entailment
20,091
Therefore , this age group continues to need more adult supervision and positive programs during after school hours .
It was an age group that could n't get by without constant adult supervision .
1neutral
20,092
Instead , they obey their caregivers and go to bed .
They went to bed and did n't disobey their caregivers .
0entailment
20,093
Retail buyers of old grappled with this problem as they do today .
The buyers then and now both have no answers to the problem .
1neutral
20,094
Gifts of $ 100.00 are more are recognized through various ways by the school .
Donations of $ 100 dollars are appreciated more by the school .
0entailment
20,095
It is safe to say that lean retailers have also come out ahead , given their rapid growth in relation to , and at the expense of , traditional retailers in many different retail channels .
Lean retailers have grown at a very slow , but consistent pace .
2contradiction
20,096
However , the simultaneous loss of radio and transponder signal would be a rare and alarming occurrence , and would normally indicate a catastrophic system failure or an aircraft crash .
All planes that lose signal while losing radio connection are due to damage to the plane .
2contradiction
20,097
Nobody who knew him would have expected him to do anything else , whether he had been sent away in disgrace or not'but , as it was thought he had left the country , his return did have a dramatic quality .
Nobody he knew expected him to stop .
0entailment
20,098
Having spreading , computer-cutting , fusing , and inspection in the same areas also allows teams of cross-trained workers to prepare the order for sewing , finishing , final inspection , and packaging .
Keeping many operations in the same area will help other operations be prepared .
0entailment
20,099
It has the original meaning , a `sprite or hobgoblin that plays tricks , especially at night in people 's houses ; a poltergeist .
Originally the word meant police officers or tax collectors .
2contradiction
20,100
Let me assure you that we would appreciate receiving one million dollars from you .
Let me tell you that we would be angry if you gave us a million dollars .
2contradiction
20,101
If you would like information , please enter your response in the appropriate area and your name , address and /or phone number in the space provided at the end .
Responding in the appropriate area at the end is the only way to get information .
1neutral
20,102
And uh , my parents and I had stopped many times in a little town called New Market , VA which was just a // little , // yeah it seemed like a halfway point .
We 'd stop in California on the way to visiting our grandparents .
2contradiction
20,103
The cockpit voice recorder captured the sounds of the passenger assault muffled by the intervening cockpit door .
The cockpit voice recorder had no problems that day .
1neutral
20,104
It 's a relatively painless experience and not very cumbersome to deal with , not bad .
It was pretty good and really simple to deal with .
0entailment
20,105
once in a blue moon , listed under moon in L ) may be listed elsewhere ( under blue moon in W ) .
The phrase `` once in a blue moon '' may be listed in the F column .
1neutral
20,106
Spreading the cloth many plies thick without stressing the cloth is , again , one of the quality steps of getting ready to sew .
One important step in sewing is spreading the cloth many piles thick and not stressing it .
0entailment
20,107
Through conversing with others , we try to make puzzling events and behaviors understandable .
Through talking with others , we try to understand funny events .
2contradiction
20,108
These social encounters oer children many opportunities to hear people refer to their own mental states and those of others and , therefore , to observe dierent points of view .
The social encounters give kids a lot of chances to hear people talk about their mental states .
0entailment
20,109
So we quick think and I go , well grab a paper towel and wet it and throw a couple of ice cubes in and put it on his hand , get him up to his feet and haul him off to the emergency room .
Quickly , we need to get something to cool off his head injury and get him to an emergency room fast !
2contradiction
20,110
I moved to San Diego .
I lived in San Diego , but used to live somewhere else .
0entailment
20,111
I 'd be happy to answer any questions you may have about this program -- please call me at 317/274-5625 .
I will gladly answer any questions you have concerning the program .
0entailment
20,112
Tough question .
Not so easy to solve
0entailment
20,113
He became even more concerned as its route changed , moving into another sector 's airspace .
The plane never should have gone that way .
1neutral
20,114
How was it going to high school in Charlotte ?
What was it like being homeschooled for the entirety of your education ?
2contradiction
20,115
Onomatopoeia has a long , honorable , and pleasurable history'from Aristophanes ' brek-ke-kex ko-ax ko-ax to Kermit 's ribbet , ribbet 'from majestic lines in ancient epic poetry to simple , homey words like plop , splash , and murmur . In every language and time , the sound and look of some words have given clues to their their sound has suggested their sense .
A philosophical meaning can be taken from onomatopoeia ,
1neutral
20,116
We have found no evidence indicating that one of the hijackers , or anyone else , sat there on this flight .
We found evidence that someone sat there on this flight .
2contradiction
20,117
[ A message to businessmen from the Pasco County ( Fla . )
A statement for people working at a business in Florida .
0entailment
20,118
As you review the gifts you 've made this year I please consider a year-end gift to Herron School of Art .
Gifts can only be given at the end of the year .
2contradiction
20,119
Our 90th Birthday celebration began in July and will continue through February .
We had no celebration for turning 70 as a company .
1neutral
20,120
Well , probably the favorite book I had , this is a little different I had The Giant Golden Book of Astronomy , which I got when I was in the second grade and I read it cover to cover till I wore it out .
I read The Giant Golden Book of Astronomy many times .
0entailment
20,121
What kind of , do you ever tell stories to your little sister ?
Your sister can read , so I 'm sure you do n't have to read her stories .
2contradiction
20,122
Men are seldom assigned to review them , possibly because the editor fears that they will be either ignorant of or unsympathetic to the issues raised , if not biased against them , or because the editor is a woman .
Men usually do n't assigned their work to review because editors get nervous about it .
0entailment
20,123
And we do n't waste money .
We never waste any money .
0entailment
20,124
Now you 're in , in what grade ?
In which grade are you now ? I think you 're in fourth grade .
1neutral
20,125
Finally , I would remind Mr. Henn that the point of the Vietnam war has nothing to do with whether we called the country 'Nam or Vietnam . It is that today Russian military personnel operate their ships and planes out of our former base complex at Cam Ranh Bay while the political language Ho Chi Minh imported from Moscow is now current through all of Indo-China .
I stopped talking to Mr. Henn years ago , he 's beyond help .
2contradiction
20,126
But to deny that the oblique senses of man are still very much with us is mere optimistic folly .
The oblique senses of females are still very much with us as well .
1neutral
20,127
Assume that the manufacturing lead time'the amount of time required to go from a production order to a product ready for shipping to retail'is less than one week .
Lead times often exceed 30 days or more .
2contradiction
20,128
The Student American Dental Hygienist 's Association Hearty Smiles project at Riley Hospital has been successful and rewarding to the children as well as the hygiene students .
Both the children as well as the hygiene students have found value in the project .
0entailment
20,129
I will call you in a few weeks to discuss your level of interest .
In a few weeks I will call you to talk about your level of interest , I hope you will donate generously .
1neutral
20,130
However , our sample also reveals that some manufacturers began to adjust to the new demands of lean By 1992 , the median dropped to ten hours , reflecting their desire to reduce the amount of work-in-process at this beginning stage of apparel assembly .
Our sample shows that some manufacturers fail to adjust to the new demands .
2contradiction
20,131
Ironically , she notes , economically less well o parents find it easier to establish a viable dividing line between workplace and home .
She notes that economically less well off parents have the toughest time separating their jobs from their families .
2contradiction
20,132
Chased him all the way up to the upstairs , I do n't think the bird ever even pooped in the house not that we found --
The bird was always calm and very housebroken .
1neutral
20,133
I 'd be happy to answer any questions you may have about this program -- please call me at 317/274-5625 .
If you have questions , someone else will need to help you .
2contradiction
20,134
How should I start ?
Would you like me to go first ?
2contradiction
YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/datasets-cards)

Overview

Original dataset is available in the original Github repo.

This dataset is a collection of NLI benchmarks constructed as described in the paper An Empirical Study on Model-agnostic Debiasing Strategies for Robust Natural Language Inference published at CoNLL 2020.

Dataset curation

No specific curation for this dataset. Label encoding follows exactly what is reported in the paper by the authors. Also, from the paper:

all the following datasets are collected based on the public available resources proposed by their authors, thus the experimental results in this paper are comparable to the numbers reported in the original papers and the other papers that use these datasets

Most of the datasets included follow the custom 3-class NLI convention {"entailment": 0, "neutral": 1, "contradiction": 2}. However, the following datasets have a particular label mapping

  • IS-SD: {"non-entailment": 0, "entailment": 1}

  • LI_TS: {"non-contradiction": 0, "contradiction": 1}

Dataset structure

This benchmark dataset includes 10 adversarial datasets. To provide more insights on how the adversarial datasets attack the models, the authors categorized them according to the bias(es) they test and they renamed them accordingly. More details in section 2 of the paper. A mapping with the original dataset names is provided below

Name Original Name Original Paper Original Curation
0 PI-CD SNLI-Hard Gururangan et al. (2018) SNLI test sets instances that cannot be correctly classified by a neural classifier (fastText) trained on only the hypothesis sentences.
1 PI-SP MNLI-Hard Liu et al. (2020) MNLI-mismatched dev sets instances that cannot be correctly classified by surface patterns that are highly correlated with the labels.
2 IS-SD HANS McCoy et al. (2019) Dataset that tests lexical overlap, subsequence, and constituent heuristics between the hypothesis and premises sentences.
3 IS-CS SoSwap-AddAMod Nie et al. (2019) Pairs of sentences whose logical relations cannot be extracted from lexical information alone. Premise are taken from SNLI dev set and modified. The original paper assigns a Lexically Misleading Scores (LMS) to each instance. Here, only the subset with LMS > 0.7 is reported.
4 LI-LI Stress tests (antonym) Naik et al. (2018) and Glockner et al. (2018) Merge of the 'antonym' category in Naik et al. (2018) (from MNLI matched and mismatched dev sets) and Glockner et al. (2018) (SNLI training set).
5 LI-TS Created by the authors Created by the authors Swap the two sentences in the original MultiNLI mismatched dev sets. If the gold label is 'contradiction', the corresponding label in the swapped instance remains unchanged, otherwise it becomes 'non-contradicted'.
6 ST-WO Word overlap Naik et al. (2018) 'Word overlap' category in Naik et al. (2018).
7 ST-NE Negation Naik et al. (2018) 'Negation' category in Naik et al. (2018).
8 ST-LM Length mismatch Naik et al. (2018) 'Length mismatch' category in Naik et al. (2018).
9 ST-SE Spelling errors Naik et al. (2018) 'Spelling errors' category in Naik et al. (2018).

Code to create the dataset


import pandas as pd
from datasets import Dataset, ClassLabel, Value, Features, DatasetDict


Tri_dataset = ["IS_CS", "LI_LI", "PI_CD", "PI_SP", "ST_LM", "ST_NE", "ST_SE", "ST_WO"]
Ent_bin_dataset = ["IS_SD"]
Con_bin_dataset = ["LI_TS"]


# read data
with open("<path to file>/robust_nli.txt", encoding="utf-8", mode="r") as fl:
    f = fl.read().strip().split("\n")
    f = [eval(i) for i in f]
df = pd.DataFrame.from_dict(f)

# rename to map common names
df = df.rename(columns={"prem": "premise", "hypo": "hypothesis"})

# reorder columns
df = df.loc[:, ["idx", "split", "premise", "hypothesis", "label"]]

# create split-specific features
Tri_features = Features(
    {
        "idx": Value(dtype="int64"),
        "premise": Value(dtype="string"),
        "hypothesis": Value(dtype="string"),
        "label": ClassLabel(num_classes=3, names=["entailment", "neutral", "contradiction"]),
    }
)

Ent_features = Features(
    {
        "idx": Value(dtype="int64"),
        "premise": Value(dtype="string"),
        "hypothesis": Value(dtype="string"),
        "label": ClassLabel(num_classes=2, names=["non-entailment", "entailment"]),
    }
)

Con_features = Features(
    {
        "idx": Value(dtype="int64"),
        "premise": Value(dtype="string"),
        "hypothesis": Value(dtype="string"),
        "label": ClassLabel(num_classes=2, names=["non-contradiction", "contradiction"]),
    }
)

# convert to datasets
dataset_splits = {}

for split in df["split"].unique():
    print(split)
    df_split = df.loc[df["split"] == split].copy()
    
    if split in Tri_dataset:
        df_split["label"] = df_split["label"].map({"entailment": 0, "neutral": 1, "contradiction": 2})
        ds = Dataset.from_pandas(df_split, features=Tri_features)
    
    elif split in Ent_bin_dataset:
        df_split["label"] = df_split["label"].map({"non-entailment": 0, "entailment": 1})
        ds = Dataset.from_pandas(df_split, features=Ent_features)
    
    elif split in Con_bin_dataset:
        df_split["label"] = df_split["label"].map({"non-contradiction": 0, "contradiction": 1})
        ds = Dataset.from_pandas(df_split, features=Con_features)

    else:
        print("ERROR:", split)
    dataset_splits[split] = ds
datasets = DatasetDict(dataset_splits)    
datasets.push_to_hub("pietrolesci/robust_nli", token="<your token>")


# check overlap between splits
from itertools import combinations
for i, j in combinations(datasets.keys(), 2):
    print(
        f"{i} - {j}: ",
        pd.merge(
            datasets[i].to_pandas(), 
            datasets[j].to_pandas(), 
            on=["premise", "hypothesis", "label"], 
            how="inner",
        ).shape[0],
    )
#> PI_SP - ST_LM:  0
#> PI_SP - ST_NE:  0
#> PI_SP - IS_CS:  0
#> PI_SP - LI_TS:  1
#> PI_SP - LI_LI:  0
#> PI_SP - ST_SE:  0
#> PI_SP - PI_CD:  0
#> PI_SP - IS_SD:  0
#> PI_SP - ST_WO:  0
#> ST_LM - ST_NE:  0
#> ST_LM - IS_CS:  0
#> ST_LM - LI_TS:  0
#> ST_LM - LI_LI:  0
#> ST_LM - ST_SE:  0
#> ST_LM - PI_CD:  0
#> ST_LM - IS_SD:  0
#> ST_LM - ST_WO:  0
#> ST_NE - IS_CS:  0
#> ST_NE - LI_TS:  0
#> ST_NE - LI_LI:  0
#> ST_NE - ST_SE:  0
#> ST_NE - PI_CD:  0
#> ST_NE - IS_SD:  0
#> ST_NE - ST_WO:  0
#> IS_CS - LI_TS:  0
#> IS_CS - LI_LI:  0
#> IS_CS - ST_SE:  0
#> IS_CS - PI_CD:  0
#> IS_CS - IS_SD:  0
#> IS_CS - ST_WO:  0
#> LI_TS - LI_LI:  0
#> LI_TS - ST_SE:  0
#> LI_TS - PI_CD:  0
#> LI_TS - IS_SD:  0
#> LI_TS - ST_WO:  0
#> LI_LI - ST_SE:  0
#> LI_LI - PI_CD:  0
#> LI_LI - IS_SD:  0
#> LI_LI - ST_WO:  0
#> ST_SE - PI_CD:  0
#> ST_SE - IS_SD:  0
#> ST_SE - ST_WO:  0
#> PI_CD - IS_SD:  0
#> PI_CD - ST_WO:  0
#> IS_SD - ST_WO:  0
Downloads last month
150
Edit dataset card

Models trained or fine-tuned on pietrolesci/robust_nli