Task stringclasses 9
values | Dataset stringclasses 35
values | subset stringclasses 1
value | Samples sequencelengths 0 0 | Instance dict | instruction stringclasses 9
values | response stringlengths 1 4.68k | option stringclasses 61
values | label_json stringlengths 14 7.45k | grouth_truth stringlengths 1 4.68k | instruction_v3 stringclasses 9
values | prompt listlengths 2 2 | answer stringlengths 14 7.45k |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : Eruptive epidermoid cysts , imiquimod )",
"id": "0",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relat... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : Eruptive epidermoid cysts , imiquimod ) | Option: adverse effect | {"relations": [{"head": "Eruptive epidermoid cysts", "relation": "adverse effect", "tail": "imiquimod"}]} | ( adverse effect : Eruptive epidermoid cysts , imiquimod ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "Eruptive epidermoid cysts", "relation": "adverse effect", "tail": "imiquimod"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : Bronchiolitis obliterans organising pneumonia , nitrofurantoin )",
"id": "1",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is foun... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : Bronchiolitis obliterans organising pneumonia , nitrofurantoin ) | Option: adverse effect | {"relations": [{"head": "Bronchiolitis obliterans organising pneumonia", "relation": "adverse effect", "tail": "nitrofurantoin"}]} | ( adverse effect : Bronchiolitis obliterans organising pneumonia , nitrofurantoin ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "Bronchiolitis obliterans organising pneumonia", "relation": "adverse effect", "tail": "nitrofurantoin"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : overanticoagulation , acenocoumarol )",
"id": "2",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relatio... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : overanticoagulation , acenocoumarol ) | Option: adverse effect | {"relations": [{"head": "overanticoagulation", "relation": "adverse effect", "tail": "acenocoumarol"}]} | ( adverse effect : overanticoagulation , acenocoumarol ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "overanticoagulation", "relation": "adverse effect", "tail": "acenocoumarol"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : angioedema , aspirin ) ; ( adverse effect : systemic anaphylactoid reaction , aspirin ) ; ( adverse effect : urticaria , aspirin )",
"id": "3",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relati... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : angioedema , aspirin ) ; ( adverse effect : systemic anaphylactoid reaction , aspirin ) ; ( adverse effect : urticaria , aspirin ) | Option: adverse effect | {"relations": [{"head": "angioedema", "relation": "adverse effect", "tail": "aspirin"}, {"head": "systemic anaphylactoid reaction", "relation": "adverse effect", "tail": "aspirin"}, {"head": "urticaria", "relation": "adverse effect", "tail": "aspirin"}]} | ( adverse effect : angioedema , aspirin ) ; ( adverse effect : systemic anaphylactoid reaction , aspirin ) ; ( adverse effect : urticaria , aspirin ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "angioedema", "relation": "adverse effect", "tail": "aspirin"}, {"head": "systemic anaphylactoid reaction", "relation": "adverse effect", "tail": "aspirin"}, {"head": "urticaria", "relation": "adverse effect", "tail": "aspirin"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : severe hypersensitivity , carboplatin )",
"id": "4",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relat... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : severe hypersensitivity , carboplatin ) | Option: adverse effect | {"relations": [{"head": "severe hypersensitivity", "relation": "adverse effect", "tail": "carboplatin"}]} | ( adverse effect : severe hypersensitivity , carboplatin ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "severe hypersensitivity", "relation": "adverse effect", "tail": "carboplatin"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : nephropathy , gold - salt )",
"id": "5",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : word1... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : nephropathy , gold - salt ) | Option: adverse effect | {"relations": [{"head": "nephropathy", "relation": "adverse effect", "tail": "gold - salt"}]} | ( adverse effect : nephropathy , gold - salt ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "nephropathy", "relation": "adverse effect", "tail": "gold - salt"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : neurologic events , Phenylpropanolamine ) ; ( adverse effect : neurologic events , PPA ) ; ( adverse effect : stroke , Phenylpropanolamine ) ; ( adverse effect : stroke , PPA )",
"id": "6",
"instruction": "Identify and extract pairs of words along with their semantic relation... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : neurologic events , Phenylpropanolamine ) ; ( adverse effect : neurologic events , PPA ) ; ( adverse effect : stroke , Phenylpropanolamine ) ; ( adverse effect : stroke , PPA ) | Option: adverse effect | {"relations": [{"head": "neurologic events", "relation": "adverse effect", "tail": "Phenylpropanolamine"}, {"head": "neurologic events", "relation": "adverse effect", "tail": "PPA"}, {"head": "stroke", "relation": "adverse effect", "tail": "Phenylpropanolamine"}, {"head": "stroke", "relation": "adverse effect", "tail":... | ( adverse effect : neurologic events , Phenylpropanolamine ) ; ( adverse effect : neurologic events , PPA ) ; ( adverse effect : stroke , Phenylpropanolamine ) ; ( adverse effect : stroke , PPA ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "neurologic events", "relation": "adverse effect", "tail": "Phenylpropanolamine"}, {"head": "neurologic events", "relation": "adverse effect", "tail": "PPA"}, {"head": "stroke", "relation": "adverse effect", "tail": "Phenylpropanolamine"}, {"head": "stroke", "relation": "adverse effect", "tail":... |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : eosinophilia , propylthiouracil ) ; ( adverse effect : granulocytopenia , propylthiouracil ) ; ( adverse effect : hepatitis , propylthiouracil ) ; ( adverse effect : lymphocyte sensitization , propylthiouracil ) ; ( adverse effect : skin reaction , propylthiouracil )",
"id": "7... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : eosinophilia , propylthiouracil ) ; ( adverse effect : granulocytopenia , propylthiouracil ) ; ( adverse effect : hepatitis , propylthiouracil ) ; ( adverse effect : lymphocyte sensitization , propylthiouracil ) ; ( adverse effect : skin reaction , propylthiouracil ) | Option: adverse effect | {"relations": [{"head": "eosinophilia", "relation": "adverse effect", "tail": "propylthiouracil"}, {"head": "granulocytopenia", "relation": "adverse effect", "tail": "propylthiouracil"}, {"head": "hepatitis", "relation": "adverse effect", "tail": "propylthiouracil"}, {"head": "lymphocyte sensitization", "relation": "ad... | ( adverse effect : eosinophilia , propylthiouracil ) ; ( adverse effect : granulocytopenia , propylthiouracil ) ; ( adverse effect : hepatitis , propylthiouracil ) ; ( adverse effect : lymphocyte sensitization , propylthiouracil ) ; ( adverse effect : skin reaction , propylthiouracil ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "eosinophilia", "relation": "adverse effect", "tail": "propylthiouracil"}, {"head": "granulocytopenia", "relation": "adverse effect", "tail": "propylthiouracil"}, {"head": "hepatitis", "relation": "adverse effect", "tail": "propylthiouracil"}, {"head": "lymphocyte sensitization", "relation": "ad... |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : hyperthyroidism , IFN - alpha )",
"id": "8",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : w... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : hyperthyroidism , IFN - alpha ) | Option: adverse effect | {"relations": [{"head": "hyperthyroidism", "relation": "adverse effect", "tail": "IFN - alpha"}]} | ( adverse effect : hyperthyroidism , IFN - alpha ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "hyperthyroidism", "relation": "adverse effect", "tail": "IFN - alpha"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : hypersensitivity , carboplatin )",
"id": "9",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : ... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : hypersensitivity , carboplatin ) | Option: adverse effect | {"relations": [{"head": "hypersensitivity", "relation": "adverse effect", "tail": "carboplatin"}]} | ( adverse effect : hypersensitivity , carboplatin ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "hypersensitivity", "relation": "adverse effect", "tail": "carboplatin"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : QT prolongation , pirmenol ) ; ( adverse effect : T wave inversion , pirmenol )",
"id": "10",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one rela... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : QT prolongation , pirmenol ) ; ( adverse effect : T wave inversion , pirmenol ) | Option: adverse effect | {"relations": [{"head": "QT prolongation", "relation": "adverse effect", "tail": "pirmenol"}, {"head": "T wave inversion", "relation": "adverse effect", "tail": "pirmenol"}]} | ( adverse effect : QT prolongation , pirmenol ) ; ( adverse effect : T wave inversion , pirmenol ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "QT prolongation", "relation": "adverse effect", "tail": "pirmenol"}, {"head": "T wave inversion", "relation": "adverse effect", "tail": "pirmenol"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : urinary calculus , Magnesium )",
"id": "11",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : w... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : urinary calculus , Magnesium ) | Option: adverse effect | {"relations": [{"head": "urinary calculus", "relation": "adverse effect", "tail": "Magnesium"}]} | ( adverse effect : urinary calculus , Magnesium ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "urinary calculus", "relation": "adverse effect", "tail": "Magnesium"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : tuberculosis , Efalizumab ) ; ( adverse effect : tuberculosis , Etanercept )",
"id": "12",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relatio... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : tuberculosis , Efalizumab ) ; ( adverse effect : tuberculosis , Etanercept ) | Option: adverse effect | {"relations": [{"head": "tuberculosis", "relation": "adverse effect", "tail": "Efalizumab"}, {"head": "tuberculosis", "relation": "adverse effect", "tail": "Etanercept"}]} | ( adverse effect : tuberculosis , Efalizumab ) ; ( adverse effect : tuberculosis , Etanercept ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "tuberculosis", "relation": "adverse effect", "tail": "Efalizumab"}, {"head": "tuberculosis", "relation": "adverse effect", "tail": "Etanercept"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : Hashimoto 's disease , IFN - alpha ) ; ( adverse effect : Hashimoto 's disease , interferon - alpha )",
"id": "13",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"No... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : Hashimoto 's disease , IFN - alpha ) ; ( adverse effect : Hashimoto 's disease , interferon - alpha ) | Option: adverse effect | {"relations": [{"head": "Hashimoto 's disease", "relation": "adverse effect", "tail": "IFN - alpha"}, {"head": "Hashimoto 's disease", "relation": "adverse effect", "tail": "interferon - alpha"}]} | ( adverse effect : Hashimoto 's disease , IFN - alpha ) ; ( adverse effect : Hashimoto 's disease , interferon - alpha ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "Hashimoto 's disease", "relation": "adverse effect", "tail": "IFN - alpha"}, {"head": "Hashimoto 's disease", "relation": "adverse effect", "tail": "interferon - alpha"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : Hepatic damage , danazol )",
"id": "14",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : word1... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : Hepatic damage , danazol ) | Option: adverse effect | {"relations": [{"head": "Hepatic damage", "relation": "adverse effect", "tail": "danazol"}]} | ( adverse effect : Hepatic damage , danazol ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "Hepatic damage", "relation": "adverse effect", "tail": "danazol"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : maculopathy , Niacin )",
"id": "15",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : word1 , w... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : maculopathy , Niacin ) | Option: adverse effect | {"relations": [{"head": "maculopathy", "relation": "adverse effect", "tail": "Niacin"}]} | ( adverse effect : maculopathy , Niacin ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "maculopathy", "relation": "adverse effect", "tail": "Niacin"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : Angioedema , budesonide ) ; ( adverse effect : dysphagia , budesonide )",
"id": "16",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : Angioedema , budesonide ) ; ( adverse effect : dysphagia , budesonide ) | Option: adverse effect | {"relations": [{"head": "Angioedema", "relation": "adverse effect", "tail": "budesonide"}, {"head": "dysphagia", "relation": "adverse effect", "tail": "budesonide"}]} | ( adverse effect : Angioedema , budesonide ) ; ( adverse effect : dysphagia , budesonide ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "Angioedema", "relation": "adverse effect", "tail": "budesonide"}, {"head": "dysphagia", "relation": "adverse effect", "tail": "budesonide"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : disfiguring facial edema , peginterferon alfa-2a ) ; ( adverse effect : disfiguring facial edema , ribavirin )",
"id": "17",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, re... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : disfiguring facial edema , peginterferon alfa-2a ) ; ( adverse effect : disfiguring facial edema , ribavirin ) | Option: adverse effect | {"relations": [{"head": "disfiguring facial edema", "relation": "adverse effect", "tail": "peginterferon alfa-2a"}, {"head": "disfiguring facial edema", "relation": "adverse effect", "tail": "ribavirin"}]} | ( adverse effect : disfiguring facial edema , peginterferon alfa-2a ) ; ( adverse effect : disfiguring facial edema , ribavirin ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "disfiguring facial edema", "relation": "adverse effect", "tail": "peginterferon alfa-2a"}, {"head": "disfiguring facial edema", "relation": "adverse effect", "tail": "ribavirin"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : pulmonary vasoconstriction , protamine ) ; ( adverse effect : right ventricular failure , protamine )",
"id": "18",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"No... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : pulmonary vasoconstriction , protamine ) ; ( adverse effect : right ventricular failure , protamine ) | Option: adverse effect | {"relations": [{"head": "pulmonary vasoconstriction", "relation": "adverse effect", "tail": "protamine"}, {"head": "right ventricular failure", "relation": "adverse effect", "tail": "protamine"}]} | ( adverse effect : pulmonary vasoconstriction , protamine ) ; ( adverse effect : right ventricular failure , protamine ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "pulmonary vasoconstriction", "relation": "adverse effect", "tail": "protamine"}, {"head": "right ventricular failure", "relation": "adverse effect", "tail": "protamine"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : Fatal pulmonary fibrosis , BCNU )",
"id": "19",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 ... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : Fatal pulmonary fibrosis , BCNU ) | Option: adverse effect | {"relations": [{"head": "Fatal pulmonary fibrosis", "relation": "adverse effect", "tail": "BCNU"}]} | ( adverse effect : Fatal pulmonary fibrosis , BCNU ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "Fatal pulmonary fibrosis", "relation": "adverse effect", "tail": "BCNU"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : Syncope , nitrate )",
"id": "20",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : word1 , word... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : Syncope , nitrate ) | Option: adverse effect | {"relations": [{"head": "Syncope", "relation": "adverse effect", "tail": "nitrate"}]} | ( adverse effect : Syncope , nitrate ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "Syncope", "relation": "adverse effect", "tail": "nitrate"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : morphea , bromocriptine )",
"id": "21",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : word1 ... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : morphea , bromocriptine ) | Option: adverse effect | {"relations": [{"head": "morphea", "relation": "adverse effect", "tail": "bromocriptine"}]} | ( adverse effect : morphea , bromocriptine ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "morphea", "relation": "adverse effect", "tail": "bromocriptine"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : cardiac arrhythmias , amphotericin B ) ; ( adverse effect : mortality , amphotericin B )",
"id": "22",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : cardiac arrhythmias , amphotericin B ) ; ( adverse effect : mortality , amphotericin B ) | Option: adverse effect | {"relations": [{"head": "cardiac arrhythmias", "relation": "adverse effect", "tail": "amphotericin B"}, {"head": "mortality", "relation": "adverse effect", "tail": "amphotericin B"}]} | ( adverse effect : cardiac arrhythmias , amphotericin B ) ; ( adverse effect : mortality , amphotericin B ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "cardiac arrhythmias", "relation": "adverse effect", "tail": "amphotericin B"}, {"head": "mortality", "relation": "adverse effect", "tail": "amphotericin B"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : severe hypoglycaemia , mefloquine )",
"id": "23",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : severe hypoglycaemia , mefloquine ) | Option: adverse effect | {"relations": [{"head": "severe hypoglycaemia", "relation": "adverse effect", "tail": "mefloquine"}]} | ( adverse effect : severe hypoglycaemia , mefloquine ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "severe hypoglycaemia", "relation": "adverse effect", "tail": "mefloquine"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : primary and secondary perforation of the bladder , epirubicin )",
"id": "24",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is foun... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : primary and secondary perforation of the bladder , epirubicin ) | Option: adverse effect | {"relations": [{"head": "primary and secondary perforation of the bladder", "relation": "adverse effect", "tail": "epirubicin"}]} | ( adverse effect : primary and secondary perforation of the bladder , epirubicin ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "primary and secondary perforation of the bladder", "relation": "adverse effect", "tail": "epirubicin"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : nephropathy , Gold )",
"id": "25",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : word1 , wor... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : nephropathy , Gold ) | Option: adverse effect | {"relations": [{"head": "nephropathy", "relation": "adverse effect", "tail": "Gold"}]} | ( adverse effect : nephropathy , Gold ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "nephropathy", "relation": "adverse effect", "tail": "Gold"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : amnesia , Triazolam ) ; ( adverse effect : nocturnal bingeing , Triazolam )",
"id": "26",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relation... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : amnesia , Triazolam ) ; ( adverse effect : nocturnal bingeing , Triazolam ) | Option: adverse effect | {"relations": [{"head": "amnesia", "relation": "adverse effect", "tail": "Triazolam"}, {"head": "nocturnal bingeing", "relation": "adverse effect", "tail": "Triazolam"}]} | ( adverse effect : amnesia , Triazolam ) ; ( adverse effect : nocturnal bingeing , Triazolam ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "amnesia", "relation": "adverse effect", "tail": "Triazolam"}, {"head": "nocturnal bingeing", "relation": "adverse effect", "tail": "Triazolam"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : epiglottitis , prochlorperazine ) ; ( adverse effect : oro - pharyngeal dystonic reaction , prochlorperazine )",
"id": "27",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, re... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : epiglottitis , prochlorperazine ) ; ( adverse effect : oro - pharyngeal dystonic reaction , prochlorperazine ) | Option: adverse effect | {"relations": [{"head": "epiglottitis", "relation": "adverse effect", "tail": "prochlorperazine"}, {"head": "oro - pharyngeal dystonic reaction", "relation": "adverse effect", "tail": "prochlorperazine"}]} | ( adverse effect : epiglottitis , prochlorperazine ) ; ( adverse effect : oro - pharyngeal dystonic reaction , prochlorperazine ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "epiglottitis", "relation": "adverse effect", "tail": "prochlorperazine"}, {"head": "oro - pharyngeal dystonic reaction", "relation": "adverse effect", "tail": "prochlorperazine"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : proarrhythmic , amiodarone )",
"id": "28",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : wor... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : proarrhythmic , amiodarone ) | Option: adverse effect | {"relations": [{"head": "proarrhythmic", "relation": "adverse effect", "tail": "amiodarone"}]} | ( adverse effect : proarrhythmic , amiodarone ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "proarrhythmic", "relation": "adverse effect", "tail": "amiodarone"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : abdominal right lower quadrant pain , cyclophosphamide ) ; ( adverse effect : cecal lymphoma , cyclophosphamide )",
"id": "29",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found,... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : abdominal right lower quadrant pain , cyclophosphamide ) ; ( adverse effect : cecal lymphoma , cyclophosphamide ) | Option: adverse effect | {"relations": [{"head": "abdominal right lower quadrant pain", "relation": "adverse effect", "tail": "cyclophosphamide"}, {"head": "cecal lymphoma", "relation": "adverse effect", "tail": "cyclophosphamide"}]} | ( adverse effect : abdominal right lower quadrant pain , cyclophosphamide ) ; ( adverse effect : cecal lymphoma , cyclophosphamide ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "abdominal right lower quadrant pain", "relation": "adverse effect", "tail": "cyclophosphamide"}, {"head": "cecal lymphoma", "relation": "adverse effect", "tail": "cyclophosphamide"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : cutaneous and hematologic toxicity , IL-2 )",
"id": "30",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( ... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : cutaneous and hematologic toxicity , IL-2 ) | Option: adverse effect | {"relations": [{"head": "cutaneous and hematologic toxicity", "relation": "adverse effect", "tail": "IL-2"}]} | ( adverse effect : cutaneous and hematologic toxicity , IL-2 ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "cutaneous and hematologic toxicity", "relation": "adverse effect", "tail": "IL-2"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : lupus erythematosus , Sulfasalazine )",
"id": "31",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relati... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : lupus erythematosus , Sulfasalazine ) | Option: adverse effect | {"relations": [{"head": "lupus erythematosus", "relation": "adverse effect", "tail": "Sulfasalazine"}]} | ( adverse effect : lupus erythematosus , Sulfasalazine ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "lupus erythematosus", "relation": "adverse effect", "tail": "Sulfasalazine"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : Nephrotic syndrome , interferon - beta-1b )",
"id": "32",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( ... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : Nephrotic syndrome , interferon - beta-1b ) | Option: adverse effect | {"relations": [{"head": "Nephrotic syndrome", "relation": "adverse effect", "tail": "interferon - beta-1b"}]} | ( adverse effect : Nephrotic syndrome , interferon - beta-1b ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "Nephrotic syndrome", "relation": "adverse effect", "tail": "interferon - beta-1b"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : symptomatic methemoglobinemia , Depakote ) ; ( adverse effect : symptomatic methemoglobinemia , divalproex sodium )",
"id": "33",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is foun... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : symptomatic methemoglobinemia , Depakote ) ; ( adverse effect : symptomatic methemoglobinemia , divalproex sodium ) | Option: adverse effect | {"relations": [{"head": "symptomatic methemoglobinemia", "relation": "adverse effect", "tail": "Depakote"}, {"head": "symptomatic methemoglobinemia", "relation": "adverse effect", "tail": "divalproex sodium"}]} | ( adverse effect : symptomatic methemoglobinemia , Depakote ) ; ( adverse effect : symptomatic methemoglobinemia , divalproex sodium ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "symptomatic methemoglobinemia", "relation": "adverse effect", "tail": "Depakote"}, {"head": "symptomatic methemoglobinemia", "relation": "adverse effect", "tail": "divalproex sodium"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : dedifferentiation of the adipocytes , insulin ) ; ( adverse effect : hyperproduction of TNF - alpha , insulin )",
"id": "34",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, r... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : dedifferentiation of the adipocytes , insulin ) ; ( adverse effect : hyperproduction of TNF - alpha , insulin ) | Option: adverse effect | {"relations": [{"head": "dedifferentiation of the adipocytes", "relation": "adverse effect", "tail": "insulin"}, {"head": "hyperproduction of TNF - alpha", "relation": "adverse effect", "tail": "insulin"}]} | ( adverse effect : dedifferentiation of the adipocytes , insulin ) ; ( adverse effect : hyperproduction of TNF - alpha , insulin ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "dedifferentiation of the adipocytes", "relation": "adverse effect", "tail": "insulin"}, {"head": "hyperproduction of TNF - alpha", "relation": "adverse effect", "tail": "insulin"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : Renal injury , anastrozole )",
"id": "35",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : wor... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : Renal injury , anastrozole ) | Option: adverse effect | {"relations": [{"head": "Renal injury", "relation": "adverse effect", "tail": "anastrozole"}]} | ( adverse effect : Renal injury , anastrozole ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "Renal injury", "relation": "adverse effect", "tail": "anastrozole"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : reversible granulocytopenia , procainamide )",
"id": "36",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"(... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : reversible granulocytopenia , procainamide ) | Option: adverse effect | {"relations": [{"head": "reversible granulocytopenia", "relation": "adverse effect", "tail": "procainamide"}]} | ( adverse effect : reversible granulocytopenia , procainamide ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "reversible granulocytopenia", "relation": "adverse effect", "tail": "procainamide"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : ataxia , carbamazepine ) ; ( adverse effect : ataxia , clarithromycin ) ; ( adverse effect : ataxia , clarithromycin ) ; ( adverse effect : dizziness , carbamazepine ) ; ( adverse effect : dizziness , clarithromycin ) ; ( adverse effect : dizziness , clarithromycin ) ; ( adverse ... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : ataxia , carbamazepine ) ; ( adverse effect : ataxia , clarithromycin ) ; ( adverse effect : ataxia , clarithromycin ) ; ( adverse effect : dizziness , carbamazepine ) ; ( adverse effect : dizziness , clarithromycin ) ; ( adverse effect : dizziness , clarithromycin ) ; ( adverse effect : drowsiness ,... | Option: adverse effect | {"relations": [{"head": "ataxia", "relation": "adverse effect", "tail": "carbamazepine"}, {"head": "ataxia", "relation": "adverse effect", "tail": "clarithromycin"}, {"head": "ataxia", "relation": "adverse effect", "tail": "clarithromycin"}, {"head": "dizziness", "relation": "adverse effect", "tail": "carbamazepine"}, ... | ( adverse effect : ataxia , carbamazepine ) ; ( adverse effect : ataxia , clarithromycin ) ; ( adverse effect : ataxia , clarithromycin ) ; ( adverse effect : dizziness , carbamazepine ) ; ( adverse effect : dizziness , clarithromycin ) ; ( adverse effect : dizziness , clarithromycin ) ; ( adverse effect : drowsiness ,... | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "ataxia", "relation": "adverse effect", "tail": "carbamazepine"}, {"head": "ataxia", "relation": "adverse effect", "tail": "clarithromycin"}, {"head": "ataxia", "relation": "adverse effect", "tail": "clarithromycin"}, {"head": "dizziness", "relation": "adverse effect", "tail": "carbamazepine"}, ... |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : nephrotoxicity , cyclosporine )",
"id": "38",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : ... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : nephrotoxicity , cyclosporine ) | Option: adverse effect | {"relations": [{"head": "nephrotoxicity", "relation": "adverse effect", "tail": "cyclosporine"}]} | ( adverse effect : nephrotoxicity , cyclosporine ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "nephrotoxicity", "relation": "adverse effect", "tail": "cyclosporine"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : acute hemorrhagic gastritis , AZ ) ; ( adverse effect : AZ intoxication , AZ )",
"id": "39",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relat... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : acute hemorrhagic gastritis , AZ ) ; ( adverse effect : AZ intoxication , AZ ) | Option: adverse effect | {"relations": [{"head": "acute hemorrhagic gastritis", "relation": "adverse effect", "tail": "AZ"}, {"head": "AZ intoxication", "relation": "adverse effect", "tail": "AZ"}]} | ( adverse effect : acute hemorrhagic gastritis , AZ ) ; ( adverse effect : AZ intoxication , AZ ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "acute hemorrhagic gastritis", "relation": "adverse effect", "tail": "AZ"}, {"head": "AZ intoxication", "relation": "adverse effect", "tail": "AZ"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : life - threatening complications , 5-FU )",
"id": "40",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( re... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : life - threatening complications , 5-FU ) | Option: adverse effect | {"relations": [{"head": "life - threatening complications", "relation": "adverse effect", "tail": "5-FU"}]} | ( adverse effect : life - threatening complications , 5-FU ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "life - threatening complications", "relation": "adverse effect", "tail": "5-FU"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : neurologic disease , penicillamine ) ; ( adverse effect : neurologic worsening , penicillamine )",
"id": "41",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\".... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : neurologic disease , penicillamine ) ; ( adverse effect : neurologic worsening , penicillamine ) | Option: adverse effect | {"relations": [{"head": "neurologic disease", "relation": "adverse effect", "tail": "penicillamine"}, {"head": "neurologic worsening", "relation": "adverse effect", "tail": "penicillamine"}]} | ( adverse effect : neurologic disease , penicillamine ) ; ( adverse effect : neurologic worsening , penicillamine ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "neurologic disease", "relation": "adverse effect", "tail": "penicillamine"}, {"head": "neurologic worsening", "relation": "adverse effect", "tail": "penicillamine"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : Vitamin K deficiency , Cephalosporins )",
"id": "42",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( rela... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : Vitamin K deficiency , Cephalosporins ) | Option: adverse effect | {"relations": [{"head": "Vitamin K deficiency", "relation": "adverse effect", "tail": "Cephalosporins"}]} | ( adverse effect : Vitamin K deficiency , Cephalosporins ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "Vitamin K deficiency", "relation": "adverse effect", "tail": "Cephalosporins"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : Multiple seizures , bupropion )",
"id": "43",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : ... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : Multiple seizures , bupropion ) | Option: adverse effect | {"relations": [{"head": "Multiple seizures", "relation": "adverse effect", "tail": "bupropion"}]} | ( adverse effect : Multiple seizures , bupropion ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "Multiple seizures", "relation": "adverse effect", "tail": "bupropion"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : histological abnormalities of the glomerulus , IFN - beta )",
"id": "44",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, f... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : histological abnormalities of the glomerulus , IFN - beta ) | Option: adverse effect | {"relations": [{"head": "histological abnormalities of the glomerulus", "relation": "adverse effect", "tail": "IFN - beta"}]} | ( adverse effect : histological abnormalities of the glomerulus , IFN - beta ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "histological abnormalities of the glomerulus", "relation": "adverse effect", "tail": "IFN - beta"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : hyperandrogenism , sodium valproate ) ; ( adverse effect : menstrual disorders , sodium valproate ) ; ( adverse effect : polycystic ovaries , sodium valproate ) ; ( adverse effect : Reproductive endocrine disorders , sodium valproate )",
"id": "45",
"instruction": "Identify a... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : hyperandrogenism , sodium valproate ) ; ( adverse effect : menstrual disorders , sodium valproate ) ; ( adverse effect : polycystic ovaries , sodium valproate ) ; ( adverse effect : Reproductive endocrine disorders , sodium valproate ) | Option: adverse effect | {"relations": [{"head": "hyperandrogenism", "relation": "adverse effect", "tail": "sodium valproate"}, {"head": "menstrual disorders", "relation": "adverse effect", "tail": "sodium valproate"}, {"head": "polycystic ovaries", "relation": "adverse effect", "tail": "sodium valproate"}, {"head": "Reproductive endocrine dis... | ( adverse effect : hyperandrogenism , sodium valproate ) ; ( adverse effect : menstrual disorders , sodium valproate ) ; ( adverse effect : polycystic ovaries , sodium valproate ) ; ( adverse effect : Reproductive endocrine disorders , sodium valproate ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "hyperandrogenism", "relation": "adverse effect", "tail": "sodium valproate"}, {"head": "menstrual disorders", "relation": "adverse effect", "tail": "sodium valproate"}, {"head": "polycystic ovaries", "relation": "adverse effect", "tail": "sodium valproate"}, {"head": "Reproductive endocrine dis... |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : neuropathic side effects , thalidomide )",
"id": "46",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( rel... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : neuropathic side effects , thalidomide ) | Option: adverse effect | {"relations": [{"head": "neuropathic side effects", "relation": "adverse effect", "tail": "thalidomide"}]} | ( adverse effect : neuropathic side effects , thalidomide ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "neuropathic side effects", "relation": "adverse effect", "tail": "thalidomide"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : sub - acute neurotoxicity , methotrexate )",
"id": "47",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( r... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : sub - acute neurotoxicity , methotrexate ) | Option: adverse effect | {"relations": [{"head": "sub - acute neurotoxicity", "relation": "adverse effect", "tail": "methotrexate"}]} | ( adverse effect : sub - acute neurotoxicity , methotrexate ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "sub - acute neurotoxicity", "relation": "adverse effect", "tail": "methotrexate"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : type 1 diabetes mellitus , recombinant alpha-2b peginterferon ) ; ( adverse effect : type 1 diabetes mellitus , ribavirin )",
"id": "48",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : type 1 diabetes mellitus , recombinant alpha-2b peginterferon ) ; ( adverse effect : type 1 diabetes mellitus , ribavirin ) | Option: adverse effect | {"relations": [{"head": "type 1 diabetes mellitus", "relation": "adverse effect", "tail": "recombinant alpha-2b peginterferon"}, {"head": "type 1 diabetes mellitus", "relation": "adverse effect", "tail": "ribavirin"}]} | ( adverse effect : type 1 diabetes mellitus , recombinant alpha-2b peginterferon ) ; ( adverse effect : type 1 diabetes mellitus , ribavirin ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "type 1 diabetes mellitus", "relation": "adverse effect", "tail": "recombinant alpha-2b peginterferon"}, {"head": "type 1 diabetes mellitus", "relation": "adverse effect", "tail": "ribavirin"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : anaphylactoid reaction , Gelofusine )",
"id": "49",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relati... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : anaphylactoid reaction , Gelofusine ) | Option: adverse effect | {"relations": [{"head": "anaphylactoid reaction", "relation": "adverse effect", "tail": "Gelofusine"}]} | ( adverse effect : anaphylactoid reaction , Gelofusine ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "anaphylactoid reaction", "relation": "adverse effect", "tail": "Gelofusine"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : nephrosis , interferon )",
"id": "50",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : word1 ,... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : nephrosis , interferon ) | Option: adverse effect | {"relations": [{"head": "nephrosis", "relation": "adverse effect", "tail": "interferon"}]} | ( adverse effect : nephrosis , interferon ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "nephrosis", "relation": "adverse effect", "tail": "interferon"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : CDI , lithium ) ; ( adverse effect : central diabetes insipidus , lithium )",
"id": "51",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relation... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : CDI , lithium ) ; ( adverse effect : central diabetes insipidus , lithium ) | Option: adverse effect | {"relations": [{"head": "CDI", "relation": "adverse effect", "tail": "lithium"}, {"head": "central diabetes insipidus", "relation": "adverse effect", "tail": "lithium"}]} | ( adverse effect : CDI , lithium ) ; ( adverse effect : central diabetes insipidus , lithium ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "CDI", "relation": "adverse effect", "tail": "lithium"}, {"head": "central diabetes insipidus", "relation": "adverse effect", "tail": "lithium"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : central nervous system depression , brimonidine )",
"id": "52",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it a... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : central nervous system depression , brimonidine ) | Option: adverse effect | {"relations": [{"head": "central nervous system depression", "relation": "adverse effect", "tail": "brimonidine"}]} | ( adverse effect : central nervous system depression , brimonidine ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "central nervous system depression", "relation": "adverse effect", "tail": "brimonidine"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : ATIN , pranlukast ) ; ( adverse effect : ATIN , pranlukast ) ; ( adverse effect : hypersensitivity reaction , pranlukast ) ; ( adverse effect : hypersensitivity reaction , pranlukast )",
"id": "53",
"instruction": "Identify and extract pairs of words along with their semantic... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : ATIN , pranlukast ) ; ( adverse effect : ATIN , pranlukast ) ; ( adverse effect : hypersensitivity reaction , pranlukast ) ; ( adverse effect : hypersensitivity reaction , pranlukast ) | Option: adverse effect | {"relations": [{"head": "ATIN", "relation": "adverse effect", "tail": "pranlukast"}, {"head": "ATIN", "relation": "adverse effect", "tail": "pranlukast"}, {"head": "hypersensitivity reaction", "relation": "adverse effect", "tail": "pranlukast"}, {"head": "hypersensitivity reaction", "relation": "adverse effect", "tail"... | ( adverse effect : ATIN , pranlukast ) ; ( adverse effect : ATIN , pranlukast ) ; ( adverse effect : hypersensitivity reaction , pranlukast ) ; ( adverse effect : hypersensitivity reaction , pranlukast ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "ATIN", "relation": "adverse effect", "tail": "pranlukast"}, {"head": "ATIN", "relation": "adverse effect", "tail": "pranlukast"}, {"head": "hypersensitivity reaction", "relation": "adverse effect", "tail": "pranlukast"}, {"head": "hypersensitivity reaction", "relation": "adverse effect", "tail"... |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : bilateral ptosis , vincristine )",
"id": "54",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 :... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : bilateral ptosis , vincristine ) | Option: adverse effect | {"relations": [{"head": "bilateral ptosis", "relation": "adverse effect", "tail": "vincristine"}]} | ( adverse effect : bilateral ptosis , vincristine ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "bilateral ptosis", "relation": "adverse effect", "tail": "vincristine"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : polyserositis , Clozapine )",
"id": "55",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relation1 : word... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : polyserositis , Clozapine ) | Option: adverse effect | {"relations": [{"head": "polyserositis", "relation": "adverse effect", "tail": "Clozapine"}]} | ( adverse effect : polyserositis , Clozapine ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "polyserositis", "relation": "adverse effect", "tail": "Clozapine"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : flare - up , goserelin acetate ) ; ( adverse effect : severe dyspnea , goserelin acetate ) ; ( adverse effect : worsening pleuritis carcinomatosa , goserelin acetate )",
"id": "56",
"instruction": "Identify and extract pairs of words along with their semantic relationship des... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : flare - up , goserelin acetate ) ; ( adverse effect : severe dyspnea , goserelin acetate ) ; ( adverse effect : worsening pleuritis carcinomatosa , goserelin acetate ) | Option: adverse effect | {"relations": [{"head": "flare - up", "relation": "adverse effect", "tail": "goserelin acetate"}, {"head": "severe dyspnea", "relation": "adverse effect", "tail": "goserelin acetate"}, {"head": "worsening pleuritis carcinomatosa", "relation": "adverse effect", "tail": "goserelin acetate"}]} | ( adverse effect : flare - up , goserelin acetate ) ; ( adverse effect : severe dyspnea , goserelin acetate ) ; ( adverse effect : worsening pleuritis carcinomatosa , goserelin acetate ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "flare - up", "relation": "adverse effect", "tail": "goserelin acetate"}, {"head": "severe dyspnea", "relation": "adverse effect", "tail": "goserelin acetate"}, {"head": "worsening pleuritis carcinomatosa", "relation": "adverse effect", "tail": "goserelin acetate"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : neutropenia , olanzapine ) ; ( adverse effect : neutropenia , thiazide )",
"id": "57",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationshi... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : neutropenia , olanzapine ) ; ( adverse effect : neutropenia , thiazide ) | Option: adverse effect | {"relations": [{"head": "neutropenia", "relation": "adverse effect", "tail": "olanzapine"}, {"head": "neutropenia", "relation": "adverse effect", "tail": "thiazide"}]} | ( adverse effect : neutropenia , olanzapine ) ; ( adverse effect : neutropenia , thiazide ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "neutropenia", "relation": "adverse effect", "tail": "olanzapine"}, {"head": "neutropenia", "relation": "adverse effect", "tail": "thiazide"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : squamous cell carcinoma , psoralens )",
"id": "58",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relati... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : squamous cell carcinoma , psoralens ) | Option: adverse effect | {"relations": [{"head": "squamous cell carcinoma", "relation": "adverse effect", "tail": "psoralens"}]} | ( adverse effect : squamous cell carcinoma , psoralens ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "squamous cell carcinoma", "relation": "adverse effect", "tail": "psoralens"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : acute tubulointerstitial nephritis , Flurbiprofen )",
"id": "59",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : acute tubulointerstitial nephritis , Flurbiprofen ) | Option: adverse effect | {"relations": [{"head": "acute tubulointerstitial nephritis", "relation": "adverse effect", "tail": "Flurbiprofen"}]} | ( adverse effect : acute tubulointerstitial nephritis , Flurbiprofen ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "acute tubulointerstitial nephritis", "relation": "adverse effect", "tail": "Flurbiprofen"}]} |
RE | ADE_corpus_sample_15000 | train | [] | {
"ground_truth": "( adverse effect : anaphylactoid reaction , zomepirac )",
"id": "60",
"instruction": "Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return \"None\". If only one relationship is found, format it as \"( relatio... | Identify and extract pairs of words along with their semantic relationship described in the text. If no valid relationship is found, return "None". If only one relationship is found, format it as "( relation1 : word1 , word2 )". For multiple relationships, separate them with semicolons using the format: "( relation1 : ... | ( adverse effect : anaphylactoid reaction , zomepirac ) | Option: adverse effect | {"relations": [{"head": "anaphylactoid reaction", "relation": "adverse effect", "tail": "zomepirac"}]} | ( adverse effect : anaphylactoid reaction , zomepirac ) | Extract all relationships between entities from the given text and return them in JSON format with the key "relations". The output should be a JSON string containing a list of objects, each with "head", "relation", and "tail". | [
{
"content": "You are an AI assistant specialized in RE tasks for the ADE_corpus_sample_15000 dataset. \nYour goal is to follow the instructions carefully and generate accurate outputs in JSON format.\nThe available options for this task are: adverse effect.",
"role": "system"
},
{
"content": "### I... | {"relations": [{"head": "anaphylactoid reaction", "relation": "adverse effect", "tail": "zomepirac"}]} |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 3