Hugging Face's logo

Datasets:
ZurichNLP
/
rsd-ists-2016

Tasks:
Token Classification
Modalities:
Text
Formats:
parquet
Languages:
English
German
Spanish
Size:
10K - 100K
ArXiv:
Libraries:
Datasets
pandas
License:
Dataset card Data Studio Files Community
1
Dataset Viewer
Auto-converted to Parquet
tokens_a
sequencelengths
4
28
tokens_b
sequencelengths
3
28
labels_a
sequencelengths
4
28
labels_b
sequencelengths
3
28
lang_a
stringclasses
1 value
lang_b
stringclasses
1 value
subset
stringclasses
2 values
id
stringlengths
9
12
alignments
stringclasses
1 value
[ "China", "'s", "Peace", "Ark", "departs", "for", "the", "Philippines", "Thursday" ]
[ "China", "schickt", "Hilfsteam", "auf", "die", "Philippinen" ]
[ 0, 0.2, 0.2, 0.2, 0.4, 0, 0, 0, 1 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
1-forward
[ "China", "sends", "aid", "team", "to", "Philippines" ]
[ "Chinas", "Peace", "Ark", "bricht", "am", "Donnerstag", "zu", "den", "Philippinen", "auf" ]
[ 0, 0.4, 0.2, 0.2, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
1-backward
[ "Conservatives", "sweep", "to", "Australia", "election", "victory" ]
[ "Konservative", "Opposition", "gewinnt", "Wahlen", "in", "Australien" ]
[ 0.2, 1, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
2-forward
[ "Conservative", "opposition", "wins", "Australian", "election" ]
[ "Konservative", "erringen", "Wahlsieg", "in", "Australien" ]
[ 0.2, 0.2, 0, 0, 0 ]
[ -1, -1, -1, -1, -1 ]
en
de
headlines
2-backward
[ "Abe", "'s", "shrine", "visit", "draws", "wide", "condemnation", ",", "criticism" ]
[ "Besuch", "des", "japanischen", "Premierministers", "Shinzo", "Abe", "am", "Kriegsschrein", "verärgert", "die", "Nachbarn" ]
[ 0.2, 0.2, 0.2, 0.2, 0.2, 0.2, 0.2, -1, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
3-forward
[ "Japanese", "PM", "Shinzo", "Abe", "'s", "war", "shrine", "visit", "angers", "neighbours" ]
[ "Abes", "Schreinbesuch", "stößt", "auf", "breite", "Verurteilung", "und", "Kritik" ]
[ 0.2, 0.2, 0.2, 0.2, 0.2, 0.2, 0.2, 0.2, 0.2, 1 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
3-backward
[ "Ukraine", "revokes", "anti-protest", "law", ",", "PM", "resigns" ]
[ "Ukraine", "will", "Anti", "-", "Protest", "-", "Gesetze", "aufheben" ]
[ 0, 0, 0.2, 0.2, -1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
4-forward
[ "Ukraine", "to", "repeal", "anti-protest", "laws" ]
[ "Ukraine", "hebt", "Anti", "-", "Protest", "-", "Gesetz", "auf", ",", "Premierminister", "tritt", "zurück" ]
[ 0, 0, 0, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
4-backward
[ "Russia", "says", "detects", "missiles", "fired", "from", "Mediterranean", "Sea" ]
[ "Russland", "sagt", ",", "ballistische", "Raketen", "im", "Mittelmeer", "abgefeuert" ]
[ 0, 0, 1, 0.2, 0, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
5-forward
[ "Russia", "says", "ballistic", "missiles", "fired", "in", "Mediterranean" ]
[ "Russland", "entdeckt", "nach", "eigenen", "Angaben", "vom", "Mittelmeer", "aus", "abgefeuerte", "Raketen" ]
[ 0, 0, 0.2, 0.2, 0, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
5-backward
[ "Pro-Russian", "protesters", "storm", "new", "police", "station", "in", "east", "Ukraine" ]
[ "Pro", "-", "russische", "Demonstranten", "stürmen", "Polizeistation", "in", "der", "Ukraine", ",", "während", "Kiew", "die", "U", ".", "N", ".", "wegen", "der", "Unruhen", "anruft" ]
[ 0, 0, 0, 0.2, 0.2, 0.2, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
6-forward
[ "Pro-Russian", "Protesters", "Storm", "Police", "Station", "in", "Ukraine", "as", "Kiev", "Urges", "U.N.", "over", "unrest" ]
[ "Pro", "-", "russische", "Demonstranten", "stürmen", "neue", "Polizeistation", "in", "der", "Ostukraine" ]
[ 0, 0, 0, 0.2, 0.2, 0.2, 0.2, 1, 1, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
6-backward
[ "Cyclone", "leaves", "trail", "of", "destruction", "in", "India" ]
[ "Evakuierung", "des", "Zyklons", "Phailin", "hielt", "indische", "Dorfbewohner", "am", "Leben" ]
[ 0.4, 1, 1, 0.4, 0.4, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
7-forward
[ "Cyclone", "Phailin", "evacuation", "kept", "Indian", "villagers", "alive" ]
[ "Wirbelsturm", "hinterlässt", "eine", "Spur", "der", "Verwüstung", "in", "Indien" ]
[ 0.4, 0.4, 0.4, 1, 0.4, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
7-backward
[ "4", "foreigners", "among", "9", "killed", "in", "Kabul", "hotel", "attack" ]
[ "Ausländer", "unter", "den", "Toten", "des", "Hotelanschlags", "in", "Kabul" ]
[ 0.2, 0.2, 0.2, 0.2, 0.2, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
8-forward
[ "Foreigners", "among", "dead", "in", "Kabul", "hotel", "attack" ]
[ "4", "Ausländer", "unter", "9", "Toten", "bei", "Anschlag", "auf", "Hotel", "in", "Kabul" ]
[ 0.2, 0.2, 0.2, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
8-backward
[ "Many", "dead", "as", "asylum", "boat", "sinks", "off", "Australia" ]
[ "Mandela", "verbringt", "dritten", "Tag", "im", "Krankenhaus" ]
[ 1, 1, 1, 1, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
9-forward
[ "Mandela", "spends", "third", "day", "in", "hospital" ]
[ "Viele", "Tote", "beim", "Untergang", "eines", "Asylboots", "vor", "Australien" ]
[ 1, 1, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
9-backward
[ "Israeli", "calls", "to", "execute", "Palestinian", "captives", "instead", "of", "releasing", "them" ]
[ "Israel", "lässt", "palästinensische", "Gefangene", "frei" ]
[ 0.2, 0.2, 1, 1, 0.2, 0.2, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1 ]
en
de
headlines
10-forward
[ "Israel", "allows", "Palestinian", "prisoner", "release" ]
[ "Israel", "fordert", ",", "palästinensische", "Gefangene", "hinzurichten", ",", "anstatt", "sie", "freizulassen" ]
[ 0.2, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
10-backward
[ "DPRK", "leader", "calls", "for", "better", "ties", "with", "S.Korea" ]
[ "Nordkorea", "fordert", "Ausländer", "auf", ",", "Südkorea", "zu", "verlassen" ]
[ 0.4, 0.4, 1, 1, 1, 1, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
11-forward
[ "North", "Korea", "urges", "foreigners", "to", "leave", "South", "Korea" ]
[ "DVRK", "-", "Führer", "ruft", "zu", "besseren", "Beziehungen", "zu", "Südkorea", "auf" ]
[ 0.4, 0.4, 1, 1, 1, 1, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
11-backward
[ "World", "'s", "oldest", "person", ",", "and", "oldest", "man", "ever", ",", "dies", "at", "116" ]
[ "Japan", ":", "Der", "älteste", "Mann", "der", "Welt", "stirbt", "im", "Alter", "von", "116", "Jahren" ]
[ 0, 0.2, 0.2, 0.2, -1, 0.2, 0.2, 0.2, 0, -1, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
12-forward
[ "Japan", ":", "World", "'s", "Oldest", "Man", "Ever", "Dies", "Aged", "116" ]
[ "Der", "älteste", "Mensch", "der", "Welt", "und", "der", "älteste", "Mann", "der", "Welt", "stirbt", "mit", "116", "Jahren" ]
[ 1, -1, 0, 0.2, 0.2, 0.2, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
12-backward
[ "World", "powers", "and", "Iran", "to", "resume", "expert", "nuclear", "talks", "on", "Dec.", "30" ]
[ "Iran", "-", "Atomexperten", "und", "Weltmächte", "wollen", "nächste", "Woche", "ihre", "Gespräche", "wieder", "aufnehmen" ]
[ 0.2, 0.2, 0.2, 0.2, 0, 0, 0.2, 0.2, 0.2, 0.4, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
13-forward
[ "Iran", "nuclear", "experts", ",", "world", "powers", "to", "resume", "talks", "next", "week" ]
[ "Weltmächte", "und", "Iran", "nehmen", "am", "30", ".", "Dezember", "Expertengespräche", "über", "Atomfragen", "wieder", "auf" ]
[ 0.2, 0.2, 0.2, -1, 0.2, 0.2, 0, 0, 0.2, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
13-backward
[ "Biden", "warns", "'", "it's", "time", "for", "Russia", "to", "stop", "talking", "and", "start", "acting", "'", "in", "Ukraine" ]
[ "Biden", ":", "Russland", "muss", "aufhören", "zu", "reden", "und", "anfangen", "zu", "handeln", ",", "um", "die", "Spannungen", "in", "der", "Ukraine", "abzubauen" ]
[ 0, 1, -1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, -1, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
14-forward
[ "Biden", ":", "Time", "for", "Russia", "to", "'", "stop", "talking", "and", "start", "acting", "'", "to", "reduce", "tension", "in", "Ukraine" ]
[ "Biden", "warnt", ":", "\"", "Es", "ist", "Zeit", ",", "dass", "Russland", "in", "der", "Ukraine", "aufhört", "zu", "reden", "und", "zu", "handeln", "\"." ]
[ 0, -1, 0, 0, 0, 0, -1, 0, 0, 0, 0, 0, -1, 1, 1, 1, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
14-backward
[ "India", "Ink", ":", "Image", "of", "the", "Day", ":", "February", "25" ]
[ "Tusche", ":", "Bild", "des", "Tages", ":", "21", ".", "April" ]
[ 0, 0, -1, 0, 0, 0, 0, -1, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
15-forward
[ "India", "Ink", ":", "Image", "of", "the", "Day", ":", "April", "21" ]
[ "Tusche", ":", "Bild", "des", "Tages", ":", "25", ".", "Februar" ]
[ 0, 0, -1, 0, 0, 0, 0, -1, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
15-backward
[ "Philippines", "defers", "comment", "on", "Taiwan", "fisher", "attack" ]
[ "Der", "Anschlag", "auf", "das", "Einkaufszentrum", "in", "Kenia", "zeigt", ",", "dass", "Bin", "Laden", "weiterhin", "Einfluss", "hat" ]
[ 1, 1, 1, 0.8, 0.8, 0.8, 0.8 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
16-forward
[ "Bin", "Laden", "'s", "continuing", "sway", "shown", "in", "Kenya", "mall", "attack" ]
[ "Philippinen", "verschieben", "Kommentar", "zu", "Taiwan", "-", "Fischerangriff" ]
[ 1, 1, 1, 1, 1, 1, 0.8, 0.8, 0.8, 0.8 ]
[ -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
16-backward
[ "New", "Zealand", "set", "to", "legalise", "gay", "marriage" ]
[ "Neuseeland", "stimmt", "für", "die", "Legalisierung", "der", "gleichgeschlechtlichen", "Ehe" ]
[ 0, 0, 0.2, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
17-forward
[ "New", "Zealand", "votes", "to", "legalise", "same-sex", "marriage" ]
[ "Neuseeland", "will", "die", "Homo", "-", "Ehe", "legalisieren" ]
[ 0, 0, 0.2, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
17-backward
[ "Israel", "'s", "ex-PM", "Ariel", "Sharon", "dies" ]
[ "Ehemaliger", "israelischer", "Premierminister", "Scharon", "stirbt" ]
[ 0, 0, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1 ]
en
de
headlines
18-forward
[ "Former", "Israeli", "Prime", "Minister", "Sharon", "dies" ]
[ "Israels", "Ex", "-", "Premierminister", "Ariel", "Sharon", "gestorben" ]
[ 0, 0, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
18-backward
[ "Peace", "talks", "with", "Palestinians", "would", "take", "months", ":", "Israel", "official" ]
[ "3jähriges", "palästinensisches", "Mädchen", "bei", "israelischer", "Razzia", "getötet" ]
[ 0.4, 0.4, 0.6, 0.6, 1, 1, 1, -1, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
19-forward
[ "3yo", "Palestinian", "girl", "killed", "in", "Israeli", "raid" ]
[ "Friedensgespräche", "mit", "den", "Palästinensern", "würden", "Monate", "dauern", ":", "Israelischer", "Beamter" ]
[ 1, 0.6, 0.6, 0.4, 0.4, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
19-backward
[ "Ronaldo", "'s", "hat", "trick", "sends", "Portugal", "to", "World", "Cup" ]
[ "Ronaldos", "Hattrick", "schießt", "Portugal", "zur", "Weltmeisterschaft" ]
[ 0, 0, 0, 0, 0.2, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
20-forward
[ "Ronaldo", "hat-trick", "fires", "Portugal", "to", "World", "Cup" ]
[ "Ronaldos", "Hattrick", "schickt", "Portugal", "zur", "Weltmeisterschaft" ]
[ 0, 0, 0.2, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
20-backward
[ "At", "Least", "22", "Dead", "After", "China", "Earthquake" ]
[ "Mindestens", "89", "Tote", "bei", "Erdbeben", "in", "China" ]
[ 0, 0, 0.4, 0.4, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
21-forward
[ "At", "least", "89", "dead", "in", "China", "earthquakes" ]
[ "Mindestens", "22", "Tote", "nach", "Erdbeben", "in", "China" ]
[ 0, 0, 0.4, 0.4, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
21-backward
[ "Canadian", "freight", "train", "blast", "death", "toll", "hits", "three", ",", "to", "rise" ]
[ "Kanadische", "Regierung", "'", "Überwachung", "der", "Situation", "bei", "der", "Zugexplosion" ]
[ 0.4, 0.4, 0.4, 0.4, 0.4, 0.4, 0.6, 1, -1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
22-forward
[ "Canadian", "government", "'", "monitoring", "'", "train", "blast", "situation" ]
[ "Explosion", "eines", "kanadischen", "Güterzuges", "fordert", "drei", "Todesopfer", ",", "Tendenz", "steigend" ]
[ 1, 1, -1, 0.6, -1, 0.4, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
22-backward
[ "Five", "children", "buried", "alive", "in", "Bohol", "landslide" ]
[ "Neun", "Kinder", "sterben", "bei", "Angriff", "in", "Afghanistan" ]
[ 0.4, 0.4, 0.2, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
23-forward
[ "Nine", "children", "die", "in", "attack", "in", "Afghanistan", "attack" ]
[ "Fünf", "Kinder", "bei", "Erdrutsch", "auf", "Bohol", "lebendig", "begraben" ]
[ 0.4, 0.4, 0.2, 1, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
23-backward
[ "Rosberg", "emulates", "father", "with", "Monaco", "win" ]
[ "FORMEL", "1", ":", "Rosberg", "bleibt", "trotz", "Monaco", "-", "Sieg", "bescheiden" ]
[ 0, 1, 1, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
24-forward
[ "FORMULA", "1", ":", "Rosberg", "stays", "modest", "despite", "Monaco", "win" ]
[ "Rosberg", "eifert", "mit", "Monaco", "-", "Sieg", "seinem", "Vater", "nach" ]
[ 1, 1, -1, 0, 1, 1, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
24-backward
[ "Suicide", "truck", "bomber", "kills", "11", "in", "northern", "Iraq" ]
[ "Selbstmord", "-", "Autobombe", "tötet", "12", "Menschen", "im", "Nordirak" ]
[ 0.2, 0.2, 0.2, 0, 0.2, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
25-forward
[ "Suicide", "car", "bomb", "kills", "12", "in", "northern", "Iraq" ]
[ "Selbstmordattentäter", "tötet", "11", "Menschen", "im", "Nordirak" ]
[ 0.2, 0.2, 0.2, 0, 0.2, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
25-backward
[ "Tropical", "Storm", "Barry", "forms", "off", "Mexico", "'s", "coast", ",", "bringing", "heavy", "rain", "as", "it", "heads", "toward", "Veracruz" ]
[ "Tropensturm", "Barry", "schwächt", "sich", "ab", "und", "lässt", "Regen", "über", "Mexiko", "nieder" ]
[ 0, 0, 0, 0.6, 0, 0, 1, 1, -1, 0.2, 0.2, 0.2, 1, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
26-forward
[ "Tropical", "Storm", "Barry", "weakens", ",", "dumps", "rain", "on", "Mexico" ]
[ "Der", "Tropensturm", "Barry", "bildet", "sich", "vor", "der", "Küste", "Mexikos", "und", "zieht", "mit", "heftigen", "Regenfällen", "auf", "Veracruz", "zu" ]
[ 0, 0, 0, 0.6, -1, 0.2, 0.2, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
26-backward
[ "Syrian", "rebels", "target", "military", "airports" ]
[ "Syrische", "Truppen", "stellen", "sich", "heftigen", "Angriffen", "von", "Rebellen", "auf", "Militärflughafen" ]
[ 0.2, 0.2, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
27-forward
[ "Syrian", "troops", "confront", "fierce", "attacks", "by", "rebels", "in", "military", "airport" ]
[ "Syrische", "Rebellen", "greifen", "Militärflughäfen", "an" ]
[ 1, 1, 0.2, 1, 1, 0.2, 0.2, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1 ]
en
de
headlines
27-backward
[ "6.6-magnitude", "quake", "hits", "South", "Atlantic", "Ocean", "--", "USGS" ]
[ "Beben", "der", "Stärke", "5", ",", "1", "im", "Süden", "des", "Iran", "--", "USGS" ]
[ 0.2, 0.2, 0, 1, 1, 1, 1, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
28-forward
[ "Magnitude-5.1", "quake", "hits", "southern", "Iran", "--", "USGS" ]
[ "Beben", "der", "Stärke", "6", ",", "6", "im", "Südatlantik", "--", "USGS" ]
[ 0.2, 0.2, 0, 1, 1, 1, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
28-backward
[ "Poverty", "stats", "doubt", "Modi", "'s", "development", "claims" ]
[ "GM", "nimmt", "ab", "1", ".", "August", "Entschädigungsanträge", "für", "Rückrufaktion", "entgegen" ]
[ 1, 1, 1, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
29-forward
[ "GM", "to", "start", "taking", "recall", "compensation", "claims", "Aug", "1" ]
[ "Armutsstatistiken", "lassen", "Zweifel", "an", "Modis", "Entwicklungsversprechen", "aufkommen" ]
[ 1, 1, 1, 1, 1, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
29-backward
[ "North", "Korea", "blocks", "access", "to", "joint", "industrial", "zone" ]
[ "Nordkorea", "blockiert", "Zugang", "zu", "wichtiger", "Industriezone" ]
[ 0, 0, 0, 0, 0.2, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
30-forward
[ "North", "Korea", "Blocks", "Access", "to", "Key", "Industrial", "Zone" ]
[ "Nordkorea", "blockiert", "Zugang", "zur", "gemeinsamen", "Industriezone" ]
[ 0, 0, 0, 0, 0.2, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
30-backward
[ "Pakistan", "'s", "Musharraf", "charged", "over", "Bhutto", "murder" ]
[ "Pakistans", "Musharraf", "wegen", "Bhuttos", "Tod", "angeklagt" ]
[ 0, 0, 0, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
31-forward
[ "Pakistan", "'s", "Musharraf", "Charged", "In", "Bhutto", "Death" ]
[ "Pakistans", "Musharraf", "wegen", "Mordes", "an", "Bhutto", "angeklagt" ]
[ 0, 0, 0, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
31-backward
[ "Ukrainian", "protesters", "back", "in", "streets", "for", "anti-government", "rally" ]
[ "Ukrainische", "Demonstranten", "stürzen", "Lenin", "-", "Statue", "in", "Kiew" ]
[ 0, 0, 1, 0.6, 0.6, 0.6, 0.6, 0.6 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
32-forward
[ "Ukraine", "protesters", "topple", "Lenin", "statue", "in", "Kiev" ]
[ "Ukrainische", "Demonstranten", "gehen", "wieder", "auf", "die", "Straße", ",", "um", "gegen", "die", "Regierung", "zu", "demonstrieren" ]
[ 0, 0, 0.6, 1, 1, 0.6, 0.6 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
32-backward
[ "Jamaican", "connection", "in", "Liverpool", "'s", "win" ]
[ "Japan", "-", "Wahl", ":", "Shinzo", "Abe", "feiert", "Erdrutschsieg" ]
[ 1, 1, 1, 1, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
33-forward
[ "Japan", "Election", ":", "Shinzo", "Abe", "Hails", "Landslide", "Win" ]
[ "Jamaikanische", "Verbindung", "bei", "Liverpools", "Sieg" ]
[ 1, 1, -1, 1, 1, 1, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1 ]
en
de
headlines
33-backward
[ "Rescuers", "scour", "debris", "after", "India", "buildings", "collapse" ]
[ "Ausgewählte", "Bilder", "vom", "Gebäudeeinsturz", "in", "Bangladesch" ]
[ 1, 1, 1, 0.4, 0.4, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
34-forward
[ "Select", "images", "from", "Bangladesh", "building", "collapse" ]
[ "Rettungskräfte", "durchkämmen", "Trümmer", "nach", "Gebäudeeinsturz", "in", "Indien" ]
[ 1, 1, 0.4, 0.4, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
34-backward
[ "Rock", "bands", "'", "plane", "makes", "emergency", "landing", "in", "Ill." ]
[ "Rockbands", "'", "Flugzeug", "macht", "Notlandung", "in", "Ill", "." ]
[ 0, 0, -1, 0, 0, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
35-forward
[ "Rock", "Bands", "'", "Plane", "Makes", "Emergency", "Landing", "in", "Ill." ]
[ "Flugzeug", "der", "Rockband", "muss", "in", "Illinois", "notlanden", "." ]
[ 0, 0, -1, 0, 0, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
35-backward
[ "Boston", "bombing", "suspect", "remains", "in", "hospital" ]
[ "Verdächtiger", "des", "Bombenanschlags", "von", "Boston", "im", "Krankenhausbett", "angeklagt" ]
[ 0, 0, 0, 1, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
36-forward
[ "Boston", "bombing", "suspect", "charged", "in", "hospital", "bed" ]
[ "Verdächtiger", "des", "Bombenanschlags", "von", "Boston", "bleibt", "im", "Krankenhaus" ]
[ 0, 0, 0, 1, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
36-backward
[ "PPP", "'s", "Shehla", "Raza", "elected", "Deputy", "Speaker", "of", "SA" ]
[ "Shehla", "Raza", "wird", "als", "stellvertretende", "Sprecherin", "der", "Sindh", "-", "Versammlung", "vereidigt" ]
[ 1, 0, 0, 0, 0.2, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
37-forward
[ "Shehla", "Raza", "takes", "oath", "as", "Sindh", "Assembly", "deputy", "speaker" ]
[ "Shehla", "Raza", "von", "der", "PPP", "zum", "stellvertretenden", "Sprecher", "der", "SA", "gewählt" ]
[ 0, 0, 0.2, 0.2, 0, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
37-backward
[ "11:42", "Churchill", "'s", "last", "child", "dies", "aged", "91" ]
[ "Churchills", "jüngstes", "Kind", "Lady", "Soames", "stirbt", "im", "Alter", "von", "91", "Jahren" ]
[ 1, 0, 0.2, 0.2, 0.2, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
38-forward
[ "Churchill", "'s", "youngest", "child", "Lady", "Soames", "dies", "aged", "91" ]
[ "11", ":", "42", "Das", "letzte", "Kind", "von", "Churchill", "stirbt", "im", "Alter", "von", "91", "Jahren" ]
[ 0, 0.2, 0.2, 0.2, 0.2, 0.2, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
38-backward
[ "N.", "Korea", "test-fires", "30", "short-range", "missiles", "into", "sea", ":", "Yonhap" ]
[ "Iran", "testet", "nach", "eigenen", "Angaben", "neue", "Land", "-", "See", "-", "Rakete", "im", "Golf" ]
[ 0.4, 0.4, 0, 0.6, 0.6, 0.6, 0.4, 0.4, -1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
39-forward
[ "Iran", "says", "it", "test-fires", "new", "land-to-sea", "missile", "in", "Gulf" ]
[ "Nordkorea", "feuert", "testweise", "30", "Kurzstreckenraketen", "ins", "Meer", ":", "Yonhap" ]
[ 0.4, 1, 1, 0, 0.6, 0.6, 0.6, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
39-backward
[ "Chinese", "shares", "close", "higher", "Tuesday" ]
[ "Chinesische", "Aktien", "schließen", "am", "Mittwoch", "niedriger" ]
[ 0, 0, 0, 1, 0.4 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
40-forward
[ "Chinese", "shares", "close", "lower", "on", "Wednesday" ]
[ "Chinesische", "Aktien", "schließen", "am", "Dienstag", "höher" ]
[ 0, 0, 0, 1, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
40-backward
[ "North", "Korea", "shuns", "offer", "of", "talks" ]
[ "Nordkorea", "schießt", "2", "Raketen", "ab" ]
[ 0, 0, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1 ]
en
de
headlines
41-forward
[ "North", "Korea", "shoots", "2", "rockets" ]
[ "Nordkorea", "lehnt", "Gesprächsangebot", "ab" ]
[ 0, 0, 1, 1, 1 ]
[ -1, -1, -1, -1 ]
en
de
headlines
41-backward
[ "In", "surprise", "visit", "to", "Afghanistan", ",", "Obama", "tells", "troops", "'", "you", "inspire", "me", "'" ]
[ "Obamas", "Überraschungsbesuch", "in", "Afghanistan" ]
[ 0, 0, 0, 0, 0, -1, 0, 1, 1, -1, 1, 1, 1, -1 ]
[ -1, -1, -1, -1 ]
en
de
headlines
42-forward
[ "Obama", "Makes", "Surprise", "Visit", "To", "Afghanistan" ]
[ "Überraschungsbesuch", "in", "Afghanistan", ":", "Obama", "sagt", "den", "Truppen", ":", "\"", "Ihr", "inspiriert", "mich", "\"." ]
[ 0, 1, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
42-backward
[ "US", "awaits", "high", "court", "judgment", "on", "gay", "marriage" ]
[ "Oberstes", "Gericht", "von", "Ark", ".", "setzt", "Homo", "-", "Ehe", "erneut", "aus" ]
[ 1, 1, 0.4, 0.4, 0.4, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
43-forward
[ "Ark.", "high", "court", "again", "suspends", "gay", "marriage" ]
[ "USA", "warten", "auf", "höchstrichterliches", "Urteil", "zur", "Homo", "-", "Ehe" ]
[ 0.4, 0.4, 0.4, 1, 1, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
43-backward
[ "Sisi", "sworn", "in", "as", "Egypt", "'s", "president", ",", "cool", "reception", "from", "West" ]
[ "Ägyptens", "Sisi", "als", "Präsident", "vereidigt" ]
[ 0, 0, 0, 0, 0, 0, 0, -1, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1 ]
en
de
headlines
44-forward
[ "Egypt", "'s", "Sisi", "sworn", "in", "as", "president" ]
[ "Sisi", "als", "Ägyptens", "Präsident", "vereidigt", ",", "kühler", "Empfang", "im", "Westen" ]
[ 0, 0, 0, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
44-backward
[ "China", "to", "investigate", "EU", "wine", "exports" ]
[ "Christie", "-", "Untersuchung", "heizt", "sich", "auf" ]
[ 1, 1, 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
45-forward
[ "Christie", "investigation", "heats", "up" ]
[ "China", "will", "EU", "-", "Weinexporte", "untersuchen" ]
[ 1, 1, 1, 1 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
45-backward
[ "Eurozone", "'s", "unemployment", "climbs", "to", "12.1%", ",", "a", "fresh", "record", "high" ]
[ "Eurozone", ":", "Rekordarbeitslosigkeit", "von", "12", ",", "2", "%", "im", "September" ]
[ 0, 0, 0, 0.4, 0.2, 0.2, -1, 0.2, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
46-forward
[ "Eurozone", "unemployment", "hit", "record", "12.2%", "in", "September" ]
[ "Arbeitslosenquote", "in", "der", "Eurozone", "steigt", "auf", "12", ",", "1", "%", "und", "erreicht", "damit", "einen", "neuen", "Rekordwert" ]
[ 0, 0, 0.4, 0.2, 0.2, 1, 1 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
46-backward
[ "South", "Korean", "President", "Sorry", "For", "Ferry", "Response" ]
[ "Südkoreanischer", "Präsident", "entschuldigt", "sich", "für", "Fährunglück" ]
[ 0, 0, 0, 0, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1 ]
en
de
headlines
47-forward
[ "S.", "Korean", "president", "'", "sorry", "'", "for", "ferry", "disaster" ]
[ "Südkoreanischer", "Präsident", "entschuldigt", "sich", "für", "Reaktion", "auf", "Fähre" ]
[ 0, 0, 0, -1, 0, -1, 0, 0, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
47-backward
[ "Pro-Russian", "rebels", "shoot", "down", "Ukrainian", "plane", ",", "49", "killed" ]
[ "Ukraine", ":", "Rebellen", "schießen", "Flugzeug", "mit", "49", "Passagieren", "an", "Bord", "ab" ]
[ 0.4, 0.4, 0, 0, 0.2, 0.2, -1, 0, 0.6 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
48-forward
[ "Ukraine", "Rebels", "shoot", "down", "plane", "carrying", "49" ]
[ "Pro", "-", "russische", "Rebellen", "schießen", "ukrainisches", "Flugzeug", "ab", ",", "49", "Tote" ]
[ 0.4, 0.4, 0, 0, 0.2, 0.6, 0 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
48-backward
[ "Saudi", "pushes", "UN", "resolution", "on", "Syria", "abuse" ]
[ "Cameron", "drängt", "auf", "U", ".", "N", ".-", "Resolution", "zu", "syrischen", "Chemiewaffen" ]
[ 0.4, 0, 0, 0, 0.4, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
49-forward
[ "Cameron", "pushes", "U.N.", "resolution", "on", "Syrian", "chemical", "arms" ]
[ "Saudi", "-", "Arabien", "drängt", "auf", "UN", "-", "Resolution", "zum", "Missbrauch", "in", "Syrien" ]
[ 0.4, 0, 0, 0, 0.4, 0.4, 0.4, 0.4 ]
[ -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
49-backward
[ "At", "least", "45", "killed", "in", "Spain", "train", "crash" ]
[ "Update", "2", ":", "78", "Tote", "bei", "spanischem", "Zugunglück" ]
[ 1, 1, 0.6, 0, 0.2, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
50-forward
[ "Update", "2", ":", "78", "killed", "in", "Spanish", "train", "disaster" ]
[ "Mindestens", "45", "Tote", "bei", "Zugunglück", "in", "Spanien" ]
[ 1, 1, -1, 0.6, 0, 0.2, 0.2, 0.2, 0.2 ]
[ -1, -1, -1, -1, -1, -1, -1 ]
en
de
headlines
50-backward
End of preview. Expand in Data Studio

Training and test data for the task of Recognizing Semantic Differences (RSD).

See the paper for details on how the dataset was created, and see our code at https://github.com/ZurichNLP/recognizing-semantic-differences for an example of how to use the data for evaluation.

The data are derived from the SemEval-2016 Task 2 for Interpretable Semantic Textual Similarity organized by Agirre et al. (2016). The original URLs of the data are:

The translations into non-English languages have been created using machine translation (DeepL).

Citation 

@inproceedings{vamvas-sennrich-2023-rsd,
      title={Towards Unsupervised Recognition of Token-level Semantic Differences in Related Documents},
      author={Jannis Vamvas and Rico Sennrich},
      month = dec,
      year = "2023",
      booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
      address = "Singapore",
      publisher = "Association for Computational Linguistics",
}
Downloads last month
131
Size of downloaded dataset files:
2.92 MB
Size of the auto-converted Parquet files:
2.92 MB
Number of rows:
33,840