Datasets:

havens2

/

naacl2022

like 0

[ "6", "Conclusion" ]

[ 0, 0 ]

[ "The", "above", "three", "types", "of", "errors", "show", "that", "selecting", "complete,", "accurate,", "and", "context-sensitive", "for", "more", "effective", "KGknowledge", "is", "vital", "augmented", "models." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Some", "questions", "may", "require", "reasoning", "based", "on", "events,", "but", "the", "knowledge", "in", "ConceptNet", "is", "more", "based", "on", "entities", "and", "attributes." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "The", "model", "is", "difficult", "to", "understand", "the", "scene", "described", "by", "the", "question." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Incomprehensible", "questions", "(23/100)", "This", "type", "of", "error", "often", "occurs", "when", "the", "question", "is", "particularly", "long,", "involving", "various", "events", "and", "changes", "in", "the", "characters’", "emotions." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "The", "model", "may", "choose", "bed", "because", "the", "bed", "appears", "more", "frequently", "in", "the", "pre-trained", "corpus." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "“cat”", "may", "be", "at", "location", "“bed”", "or", "“comfortable", "chair”,", "and", "the", "knowledge", "provided", "by", "ConceptNet", "is", "also", "the", "same." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "For", "example,", "“human”", "and" ]

[ 0, 0, 0, 0 ]

[ "Indistinguishable", "knowledge", "(25/100)", "Several", "choices", "of", "the", "question", "may", "be", "correct,", "difficult", "to", "distinguish,", "and", "which", "one", "is", "correct", "may", "vary", "from", "person", "to", "person." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "However,", "ConceptNet", "does", "not", "cover", "such", "knowledge", "or", "not", "is", "retrieved." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "For", "example,", "although", "“eating_dinner”", "will", "cause", "“sleepiness”", "or", "“indigestion”,", "knowledge", "such", "as", "“lactose", "intolerance", "causes", "indigestion”", "is", "essential", "to", "answer", "the", "question", "(Wikipedia:", "Lactose", "intolerance", "is", "a", "common", "condition", "caused", "by", "a", "decreased", "ability", "to", "digest", "lactose,", "a", "sugar", "found", "in", "dairy", "products.)." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Although", "we", "can", "retrieve", "many", "nodes", "related", "to", "questions", "and", "choices", "from", "ConceptNet,", "due", "to", "the", "incompleteness", "of", "the", "knowledge", "graph,", "there", "may", "be", "missing", "essential", "evidence", "nodes", "in", "the", "reasoning", "paths", "to", "answer", "the", "question." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "There", "are", "three", "main", "types", "of", "errors,", "and", "we", "show", "some", "examples", "in", "the", "Appendix", "C.", "Miss", "important", "evidence", "(39/100)" ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "In", "order", "to", "understand", "why", "our", "model", "fails", "in", "some", "cases,", "we", "randomly", "select", "100", "error", "cases", "and", "group", "them", "into", "several", "categories." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "5.5", "Error", "Analysis" ]

[ 0, 0, 0 ]

[ "These", "two", "paths", "describe", "two", "possible", "scenarios", "that", "support", "answering", "the", "question." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "The", "question", "and", "answer", "choices", "corresponding", "to", "this", "case", "are:", "\"What", "do", "people", "typically", "do", "while", "playing", "guitar?" ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "From", "(a)", "to", "(b),", "although", "the", "nodes", "wood", "and", "burn", "bridge", "the", "reasoning", "gap", "between", "question", "entity", "and", "answer", "entity,", "their" ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "The", "flow", "from", "(a)", "to", "(b)", "to", "(c)", "represents", "the", "recursive", "pruning", "of", "the", "subgraph", "according", "to", "the", "LM-to-KG", "attention", "weight", "at", "each", "GNN", "update", "layer." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Figure", "4", "shows", "an", "example", "from", "CommonsenseQA", "where", "our", "model", "correctly", "answers", "the", "question", "and", "finally", "retains", "reasonable", "reasoning", "paths", "by", "pruning", "the", "subgraph." ]

[ 0, 0, 0, 0, 0, 0, 13, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "We", "aim", "to", "interpret", "JointLK’s", "reasoning", "process", "by", "analyzing", "the", "pruning", "of", "the", "knowledge", "subgraph." ]

[ 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "5.4", "Interpretability:", "A", "Case", "Study" ]

[ 0, 0, 0, 0, 0 ]

[ "Compared", "with", "QA-GNN,", "JointLK", "has", "a", "bigger", "boost", "on", "questions", "with", "more", "entities", "(", "2.01%)", "than", "those", "with", "fewer", "entities", "↑", "(", "0.96%),", "suggesting", "that", "our", "model", "can", "reduce", "↑", "the", "reasoning", "difficulty", "of", "complex", "questions", "because", "it", "can", "remove", "irrelevant", "nodes", "in", "reasoning." ]

[ 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 9, 0, 0, 0, 0, 0, 0, 0, 9, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "According", "to", "statistics", "(see", "Appendix", "A),", "questions", "contain", "an", "average", "of", "7", "entities,", "so", "we", "divide", "the", "question", "into", "two", "categories:", "containing", "fewer", "entities", "(", "7)", "and", "more", "entities(>7)." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Questions", "with", "fewer/more", "entities", "When", "the", "question", "contains", "many", "entities,", "the", "size", "and", "noise", "of", "the", "retrieved", "KG", "may", "limit", "the", "model’s", "performance", "because", "the", "model", "needs", "to", "understand", "the", "complex", "relationship", "between", "entities." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "The", "fine-grained", "joint", "inference", "of", "LM", "and", "GNN", "allows", "the", "model", "to", "pay", "attention", "to", "the", "semantic", "nuances", "of", "language", "expressions." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "JointLK", "exhibits", "a", "big", "boost", "(", "3.00%)", "over", "QA↑", "GNN,", "suggesting", "its", "strength", "in", "negation", "reasoning." ]

[ 1, 0, 0, 0, 0, 0, 9, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "To", "investigate", "the", "reasoning", "ability", "of", "the", "model", "on", "negation,", "we", "retrieved", "133", "questions", "with", "negation", "terms", "(e.g.,", "no,", "not,", "nothing,", "never,", "unlikely,", "don’t,", "doesn’t,", "didn’t,", "can’t,", "couldn’t)", "from", "the", "CommonsenseQA", "IHdev", "set." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 13, 0, 0 ]

[ "Questions", "with", "negation", "Large", "LMs", "do", "well", "due", "to", "memorizing", "subject", "and", "filler", "co-occurrences", "but", "are", "easily", "distracted", "by", "elements", "like", "negation", "(Zagoury", "et", "al.,", "2021)." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "We", "compare", "our", "model", "with", "the", "prior", "best", "LM+KG", "model,", "QA-GNN", "in", "Table", "6." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0 ]

[ "more", "entities." ]

[ 0, 0 ]

[ "The", "questions", "are", "retrieved", "from", "the", "CommonsenseQA", "IHdev", "set." ]

[ 0, 0, 0, 0, 0, 0, 13, 0, 0 ]

[ "Considering", "the", "overall", "performance", "improvement", "of", "our", "model", "on", "these", "two", "datasets,", "we", "analyze", "whether", "the", "improvement", "is", "reflected", "in", "questions", "that", "require", "more", "complex", "reasoning,", "such", "as", "questions", "with", "negation", "and", "complex", "questions", "with" ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "5.3", "Quantitative", "Analysis" ]

[ 0, 0, 0 ]

[ "As", "shown", "in", "Figure", "3", "(b),", "when", "the", "number", "of", "JointLK", "layers", "N", "=", "5,", "K", "=", "0.92", "(about", "66%", "of", "the", "original", "nodes", "remain", "in", "the", "last", "layer)", "works", "the", "best", "on", "the", "CommonsenseQA", "dev", "set." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 3, 3, 0, 5, 3, 0, 5, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 13, 0, 0 ]

[ "Experiments", "show", "that", "if", "the", "retention", "ratio", "is", "too", "high,", "there", "may", "be", "almost", "no", "pruning", "effect", "(for", "example,", "K=0.98,", "90%", "of", "the", "nodes", "are", "retained", "in", "the", "last", "layer);", "otherwise,", "useful", "nodes", "may", "be", "deleted." ]

[ 0, 0, 0, 0, 0, 3, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Since", "it", "is", "recursively", "pruning", "in", "each", "stacked", "layer", "of", "JointLK,", "the", "percentage", "of", "graph", "nodes", "that", "the", "model", "ultimately", "retains", "is", "also", "related", "to", "the", "number", "of", "layers", "of", "JointLK,", "that", "is,", "KN", ",", "where", "N", "=", "5." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 3, 0, 5 ]

[ "Impact", "of", "the", "Retention", "Ratio", "in", "Pruning", "The", "retention", "ratio", "K", "is", "a", "hyperparameter", "of", "the", "dynamic", "pruning", "module." ]

[ 0, 0, 0, 3, 4, 0, 0, 0, 3, 4, 3, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "As", "the", "number", "of", "layers", "increases,", "the", "model", "changes", "from", "underfitting", "to", "overfitting." ]

[ 0, 0, 3, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "However,", "performance", "begins", "to", "drop", "when", "N", ">", "5." ]

[ 0, 0, 0, 0, 0, 0, 3, 0, 5 ]

[ "The", "increase", "of", "layers", "continues", "to", "bring", "benefits", "until", "layers", "N", "=", "5." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 0, 5 ]

[ "We", "investigate", "the", "impact", "of", "the", "number", "of", "JointLK", "layers", "(shown", "in", "Figure", "3", "(a))." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "results", "have", "a", "significant", "drop:", "77.88%", "76.61%,", "suggesting", "that", "the", "joint", "reasoning", "between", "LM", "and", "KG", "is", "critical." ]

[ 0, 0, 0, 0, 0, 9, 9, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "We", "report", "the", "IHdev", "accuracy", "on", "CommonsenseQA." ]

[ 0, 0, 0, 0, 7, 0, 13 ]

[ "Especially,", "when", "we", "disable", "the", "joint", "reasoning", "module,", "the", "corresponding", "dynamic", "pruning", "module", "will", "also", "be", "removed,", "because", "the", "latter", "depends", "on", "the", "attention", "value", "in", "the", "former." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Disabling", "the", "dynamic", "pruning", "module", "results", "in", "0.5%", "drop", "in", "performance,", "showing", "that", "some", "nodes", "in", "subgraph", "are", "not", "conducive", "to", "reasoning." ]

[ 0, 0, 0, 0, 0, 0, 0, 9, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Impact", "of", "JointLK", "components", "We", "assess", "the", "impact", "of", "the", "joint", "reasoning", "module", "(§", "3.4)", "and", "the", "dynamic", "pruning", "module", "(§", "3.5),", "shown", "in", "Table", "5." ]

[ 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "We", "show", "the", "accuracy", "of", "JointLK", "on", "the", "CommonsenseQA", "IHdev", "set." ]

[ 0, 0, 0, 7, 0, 1, 0, 0, 13, 0, 0 ]

[ "We", "further", "conduct", "in-depth", "analyses", "to", "investigate", "the", "effectiveness", "of", "different", "components", "in", "our", "model." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "5.2", "Ablation", "Studies" ]

[ 0, 0, 0 ]

[ "In", "particular,", "we", "do", "not", "compare", "with", "the", "higher", "ranking", "models", "on", "the", "leaderboard,", "such", "as", "unified", "QA", "(Khashabi", "et", "al.,", "2020),", "Albert", "+", "DESC-KCR", "(Xu", "et", "al.,", "2021),", "because", "they", "either", "use", "a", "stronger", "text", "encoder", "or", "use", "additional", "data", "resources,", "while", "our", "model", "focuses", "on", "improving", "the", "joint", "reasoning", "between", "LM", "and", "KG." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "The", "previous", "top", "2", "systems,", "UnifiedQA", "(11B", "params)", "and", "T5", "(3B", "params)", "are", "30x", "and", "8x", "larger", "than", "our", "model." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "All", "listed", "methods", "use", "the", "provided", "science", "facts", "as", "an", "additional", "input", "to", "the", "language", "context." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Methods", "with", "AristoRoBERTa", "use", "the", "textual", "evidence", "by", "Clark", "et", "al.", "(2020)", "as", "an", "additional", "input", "to", "the", "QA", "context." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Compared", "with", "the", "previous", "best", "model", "MHGRN", "and", "QA-GNN,", "the", "boost", "over", "them", "suggests", "the", "effectiveness", "of", "our", "proposed", "joint", "reasoning", "between", "LM", "and", "KG", "and", "the", "dy" ]

[ 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Additionally,", "we", "also", "submit", "our", "best", "model", "to", "the", "leaderboards,", "and", "our", "JointLK", "(with", "the", "text", "encoder", "being", "RoBERTa-large)", "ranks", "first", "among", "comparable", "approaches." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "On", "OpenbookQA,", "our", "model’s", "test", "performance", "improves", "by", "6.52%", "over", "fine-tuned", "AristoRoBERTa,", "and", "2.15%", "over", "QA-GNN." ]

[ 0, 13, 0, 0, 0, 0, 0, 0, 9, 0, 0, 1, 0, 9, 0, 1 ]

[ "On", "CommonsenseQA,", "our", "model’s", "test", "performance", "improves", "by", "5.74%", "over", "fine-tuned", "LMs", "and", "1.02%", "over", "the", "prior", "best", "LM+KG", "model,", "QA-GNN." ]

[ 0, 13, 0, 0, 0, 0, 0, 0, 9, 0, 0, 0, 0, 9, 0, 0, 0, 0, 0, 0, 1 ]

[ "We", "can", "observe", "that", "JointLK", "performs", "best", "among", "all", "fine-tuned", "LMs", "and", "existing", "LM+KG", "models." ]

[ 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "The", "results", "on", "OpenBookQA", "test", "dataset", "and", "leaderboard", "are", "shown", "in", "Table", "3", "and", "Table", "4." ]

[ 0, 0, 0, 13, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "The", "results", "on", "CommonsenseQA", "in-house", "split", "dataset", "and", "official", "test", "dataset", "are", "shown", "in", "Table", "1", "and", "Table", "2." ]

[ 0, 0, 0, 13, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "5.1", "Main", "Results" ]

[ 0, 0, 0 ]

[ "5", "Results", "and", "Analysis" ]

[ 0, 0, 0, 0 ]

[ "Our", "model", "has", "achieved", "state-of-the-art", "under", "the", "setting", "of", "RoBERTa-large." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Methods" ]

[ 0 ]

[ "We", "follow", "the", "data", "division", "method", "of", "Lin", "et", "al.", "(2019)", "and", "report", "the", "in-house", "Dev", "(IHdev)", "and", "Test", "(IHtest)", "accuracy(mean", "and", "standard", "deviation", "of", "four", "runs)." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Methods" ]

[ 0 ]

[ "To", "be", "fair,", "we", "use", "the", "same", "LM", "for", "all", "comparison", "methods." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "(1),", "(2)", "and", "(3)", "are", "the", "relational", "perception", "GNNs", "for", "KGs,", "and", "(4),", "(5)", "and", "(6)", "are", "further", "model", "paths", "in", "KGs." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "GconAttn", "(Wang", "et", "al.,", "2019),", "(4)KagNet", "(Lin", "et", "al.,", "2019)", "and", "(5)MHGRN", "(Feng", "et", "al.,", "2020),", "(6)", "QA-GNN", "(Yasunaga", "et", "al.,", "2021)." ]

[ 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0 ]

[ "For", "LM+KG", "methods,", "they", "share", "a", "similar", "high-level", "framework", "with", "our", "methods,", "that", "is,", "LM", "is", "used", "as", "a", "text", "encoder,", "GNN", "or", "RN", "is", "used", "as", "a", "KG", "encoder,", "but", "the", "way", "of", "using", "knowledge", "or", "reasoning", "is", "different:", "(1)", "Relationship", "network", "(RN)", "(Santoro", "et", "al.,", "2017),", "(2)", "RGCN", "(Schlichtkrull", "et", "al.,", "2018),", "(3)" ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0 ]

[ "To", "investigate", "the", "role", "of", "KGs,", "we", "compare", "with", "the", "benchmark", "model", "RoBERTa-large", "(Liu", "et", "al.,", "2019)", "for", "CommonsenseQA,", "and", "compare", "with", "RoBERTa-large", "and", "AristoRoBERTa", "(Clark", "et", "al.,", "2020)", "for", "OpenBookQA." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 13, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 13 ]

[ "Although", "text", "corpus", "can", "provide", "complementary", "knowledge", "except", "for", "knowledge", "graphs,", "our", "model", "focuses", "on", "improving", "the", "use", "of", "KG", "and", "the", "joint", "reasoning", "between", "LM", "and", "KG,", "so", "we", "choose", "LM", "and", "LM+KG", "as", "the", "comparison", "methods." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0 ]

[ "4.3", "Compared", "Method" ]

[ 0, 0, 0 ]

[ "We", "use", "separate", "learning", "rates", "for", "the", "LM", "encoder", "and", "the", "graph", "encoder." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "We", "set", "the", "dimension", "(D", "=", "200)", "and", "number", "of", "layers", "(N", "=", "5)", "of", "our", "GNN", "module,", "with", "dropout", "rate", "0.2", "applied", "to", "each", "layer", "(Srivastava", "et", "al.,", "2014)." ]

[ 0, 0, 0, 3, 3, 0, 5, 0, 3, 4, 4, 3, 0, 5, 0, 0, 0, 0, 0, 3, 4, 5, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "In", "training,", "we", "set", "the", "maximum", "input", "sequence", "length", "to", "text", "encoders", "to", "100,", "batch", "size", "to", "128,", "and", "perform", "early", "stopping." ]

[ 0, 0, 0, 0, 0, 0, 3, 4, 4, 0, 0, 0, 0, 5, 3, 4, 0, 5, 0, 0, 0, 0 ]

[ "We", "use", "cross-entropy", "loss", "and", "RAdam", "optimizer", "(Liu", "et", "al.,", "2020)." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "Given", "each", "query,", "we", "follow", "the", "preprocessing", "steps", "described", "in", "Feng", "et", "al.", "(2020)", "to", "retrieve", "the", "subgraph", "from", "ConceptNet,", "and", "the", "max", "hop", "size", "is", "3", "(see", "Appendix", "A", "for", "the", "detail)." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 4, 4, 0, 5, 0, 0, 0, 0, 0, 0 ]

[ "Following", "previous", "work", "(Yasunaga", "et", "al.,", "2021),", "we", "use", "ConceptNet", "(Speer", "et", "al.,", "2017),", "a", "commonsense", "knowledge", "graph,", "as", "our", "structured", "knowledge", "source", "for", "both", "of", "the", "above", "tasks." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "4.2", "Implementation", "Details" ]

[ 0, 0, 0 ]

[ "We", "use", "the", "official", "data", "split." ]

[ 0, 0, 0, 0, 0, 0 ]

[ "It", "contains", "5,957", "questions", "along", "with", "an", "open", "book", "of", "scientific", "facts." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "question", "answering", "dataset", "that", "requires", "reasoning", "with", "elementary", "science", "knowledge." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "OpenBookQA", "is", "a", "4-way", "multiple", "choice" ]

[ 13, 0, 0, 0, 0, 0 ]

[ "(2019),", "and", "report", "the", "accuracy", "of", "our", "final", "system", "on", "the", "official", "test", "set." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "We", "experiment", "and", "report", "the", "accuracy", "on", "the", "in-house", "dev", "(IHdev)", "and", "test", "(IHtest)", "splits", "used", "by", "Lin", "et", "al." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "CommonsenseQA", "is", "a", "5-way", "multiple-choice", "question", "answering", "dataset", "that", "requires", "commonsense", "for", "reasoning", "and", "contains", "12,102", "questions." ]

[ 13, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "(Talmor", "et", "al.,", "2019)", "and", "OpenBookQA", "(Mihaylov", "et", "al.,", "2018)." ]

[ 0, 0, 0, 0, 0, 13, 0, 0, 0, 0 ]

[ "We", "evaluate", "our", "model", "on", "two", "typical", "commonsense", "question", "answering", "datasets", "CommonsenseQA" ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 13 ]

[ "4.1", "Datasets" ]

[ 0, 0 ]

[ "4", "Experimental", "Setup" ]

[ 0, 0, 0 ]

[ "We", "get", "the", "final", "probability", "by", "normalize", "all", "question-choice", "pairs", "with", "softmax." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "where", "s", "is", "the", "mean", "pooling", "of", "QN", ",", "and", "g", "is", "the", "attention-based", "pooling", "of", "XN", "." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "We", "compute", "the", "score", "of", "a", "being", "the", "correct", "answer", "as:" ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "After", "N", "layers", "of", "iteration,", "we", "finally", "obtain", "the", "query", "representation", "QN", "that", "fuses", "knowledge", "information", "and", "the", "graph", "representation", "XN", "that", "fuses", "question", "information." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "3.6", "Answer", "Prediction" ]

[ 0, 0, 0 ]

[ "Next,", "the", "subgraph", "is", "formed", "by", "pooling", "out", "the", "less", "essential", "entity", "nodes", "as:" ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "where", "top-rank", "is", "a", "function", "that", "returns", "the", "index", "of", "top", "·idx", "is", "an", "indexing", "operation,", "and", "Zmask", "is", "corresponding", "attention", "mask." ]

[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]

[ "K" ]

[ 0 ]