Downloads

Software

  • QA-SRL to Open-IE – Open-IE benchmark and conversion from QA-SRL.
    More details
  • context2vec – sentential context embeddings using bidirectional LSTM
    More details
  • HypeNET- Integrated path-based and distributional method for hypernymy detection
    More details
  • LinKeR - Lexical Inference using Knowledge Resources
    More details
  • hyperwords – Hyperparameter-Enabled Word Representations
    More details
  • word2parvec – generate paraphrase vector representations for words in context
    More details
  • word2vecf – word2vec with Arbitrary Contexts
    More details
  • BIUTEE – Bar Ilan University Textual Entailment Engine
    BIUTEE is now maintained as part of the EXCITEMENT-Open-Platform.
  • TruthTeller – A Semantic Truth Value Annotator
    More details
  • The Syntax-based Rulebase for Textual Entailment
    More details
  • Java number normalizer (Beta)
    More details
  • M-PLM: a Probabilistic Lexical Model for Inference Ranking
    More details
  • PLIS: a Probabilistic Lexical Inference System
    More details

Knowledge Resources

  • Non-Restrictive NP Modification Corpus
    More details
  • Word Embeddings with Different Context Types
    More details
  • Word Embeddings for Lexical Substitution
    More details
  • Dependency-Based Word Embeddings
    More details
  • WikiRules!
    A rule-base of about 8 million lexical reference rules extracted from Wikipedia.
    More details
  • FRED
    FrameNet-derived entailment rule-base
    More details
  • ReutersBinaryDirt
    The result of applying BIU’s implementation of the DIRT learning algorithm over the Reuters corpus.
    More details
  • ReutersUnaryBInc
    The result of applying Szpektor and Dagan’s algorithm for learning entailment between unary templates using the BInc similarity function.
    More details
  • DIRECT – Directional Distributional Term-Similarity Resource
    The resource contains directional distributional term-similarity rules automatically extracted as described in (Kotlerman et.al., JNLE-DLS 2010). Most of the rules are lexical entailment rules, where the meaning of the rule’s left-hand-side implies the meaning of its right-hand-side.For instance: koala–>animal, bread–>food, imprisonment–>arrest,
    wedding–>marriage.
    More details
  • ReverbLocalGlobal
    Three resources of predicative entailment rules containing millions of rules in both shallow and syntactic formats learned over the REVERB data set using local and global algorithms as described in Chapter 5 of Jonathan Berant’s thesis.
    More details
  • Context Sensitive Predicate Inference Rules
    Context sensitive predicate inference rules as described in Melamud et al. (ACL 2013).
    More details

Datasets