How we BLESSed distributional semantic evaluation

By Marco Baroni, Alessandro Lenci. GEometrical Models of Natural Language Semantics, 2011

327 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/e58e0c28e1ea142b7ed6b0f88fc2b01b711abc3b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Analogy-based detection of morphological and semantic relations with word embeddings: what works and what doesn’t. (2016, methodology): “Some studies report effects of changing model parameters on 1For semantic relations there are also generic resources such as EVALution (Santus et al., 2015), and semantic similarity sets such as BLESS and WordSim353 (Baroni and Lenci, 2011), which are sometimes used as sources for compiling analogy ”
Multilingual Reliability and “Semantic” Structure of Continuous Word Spaces (2015, methodology): “BLESS is available only for English, but we included it in Sem-Para as it is a popular semantic benchmark.”
Context Mover's Distance & Barycenters: Optimal transport of contexts for building representations (2018, unknown): “In total, we evaluate our method on 10 standard datasets: BLESS (Baroni & Lenci, 2011), EVALution (Santus et al., 2015), (Benotto, 2015), (Weeds et al., 2014), BIBLESS (Kiela et al., 2015), (Baroni et al., 2012), (Kotlerman et al., 2010), (Levy et al., 2014), HypeNet-Test (Shwartz et al., 2016), and”
Entailment above the word level in distributional semantics (2012, methodology): “We then took the Cartesian product of these 256 adjectives with the 200 concrete nouns in the BLESS data set (Baroni and Lenci, 2011).”
Poincaré GloVe: Hyperbolic Word Embeddings (2018, methodology): “For hypernymy evaluation we use the Hyperlex (Vuli´c et al., 2017) and WBLess (subset of BLess) (Baroni & Lenci, 2011) datasets.”
Rare Words: A Major Problem for Contextualized Embeddings And How to Fix it by Attentive Mimicking (2019, background): “For example, Baroni and Lenci (2011) compile the BLESS dataset that covers ﬁve different semantic relations (e.g., hyponymy) from multiple sources.”
How Contextual are Contextualized Word Representations? Comparing the Geometry of BERT, ELMo, and GPT-2 Embeddings (2019, methodology): “These tasks cover semantic similarity, analogy solving, and concept categorization: SimLex999 (Hill et al., 2015), MEN (Bruni et al., 2014), WS353 (Finkelstein et al., 2002), RW (Luong et al., 2013), SemEval-2012 (Jurgens et al., 2012), Google analogy solving (Mikolov et al., 2013a) MSR analogy solv”
On the Systematicity of Probing Contextualized Word Representations: The Case of Hypernymy in BERT (2020, background): “…Baroni et al., 2012; Rimell, 2014; Roller et al., 2014; Weeds et al., 2014; Shwartz et al., 2015; Vuli´c and Mrkˇsi´c, 2018) and evaluating distributional semantic models on their ability to represent it (Baroni and Lenci, 2011; Santus et al., 2015, 2016; Necsulescu, 2011; Vyas and Carpuat, 2017).”
Evaluating word embedding models: methods and experimental results (2019, unknown): “ntactic result is much better than its semantic result. 3) Concept Categorization: Three datasets are used in concept categorization evaluation. They are: 1) the AP dataset [52], 2) the BLESS dataset [53] and 3) the BM dataset [54]. The AP dataset contains 402 words that are divided into 21 2https:/”
Distributional Models of Word Meaning (2018, methodology): “How we BLESSed distributional semantic evaluation.”
Learning to Compute Word Embeddings On the Fly (2017, methodology): “Right plot shows tSNE of word embeddings for 3 random categories from BLESS dataset (Baroni and Lenci, 2011) (denoted by color).”
Double-Hard Debias: Tailoring Word Embeddings for Gender Bias Mitigation (2020, methodology): “, 2008); the Battig 1969 set (Battig and Montague, 1969) and the BLESS dataset (Baroni and Lenci, 2011).”
Detecting Asymmetric Semantic Relations in Context: A Case-Study on Hypernymy Detection (2017, background): “For instance, Baroni and Lenci (2011) use cohyponyms and random words as negative examples.”
Human and Machine Judgements for Russian Semantic Relatedness (2016, methodology): “This dataset follows the structure of the BLESS dataset [25].”
pair2vec: Compositional Word-Pair Embeddings for Cross-Sentence Inference (2018, methodology): “These methods are evaluated on either intrinsic relation prediction tasks, such as BLESS (Ba-roni and Lenci, 2011) and CogALex (Santus et al., 2016), or knowledge-base population benchmarks, e.g. FB15 (Bordes et al., 2013).”
Do Supervised Distributional Methods Really Learn Lexical Inference Relations? (2015, unknown): “(Baroni and Lenci, 2011) a.”