Semantic Role Labelling

Question 1

What is a semantic role?

Answer 1

It is a role that arguments of a predicate take in an event

Question 2

What are some semantic role datasets?

Answer 2

PropBank and FrameNet

Question 3

What is semantic role labelling?

Answer 3

It is the task of assigning roles to spans of text in sentences

Question 4

What is a thematic role?

Answer 4

It captures the semantic commonality around participants of event. There are many roles, typically 12

Question 5

What are some problems with thematic roles?

Answer 5

There are lots of different themes/events that lead to a lot of thematic roles

Question 6

What has been developed instead of thematic roles?

Answer 6

Generalised semantic roles

Question 7

What are the options for generalised semantic roles?

Answer 7

Defining them heuristically (PropBank)

Defining roles specific to groups of verbs or nouns (FrameNet)

Question 8

In PropBank, what are semantic roles tied to?

Answer 8

They are tied to specific verb sense predicates

Question 9

In PropBank, what does each verb have?

Answer 9

Each verb has N arguments:

arg0 is usually the PROTO-AGENT, arg1 is usually the PROTO-PATIENT

Question 10

Explain what is shown in the image

Answer 10

In the image, it shows the first word sense for the words agree and fall and their arguments. For agree, argument 0, the PROTO-AGENT, is the agreer, while argument 1, the PROTO-PATIENT, is the proposition, and argument 2 is the other entity agreeing

Question 11

What else can argument have in PropBank?

Answer 11

They can have argument modifiers, known as ArgM

Question 12

What is shown in the image?

Answer 12

These are different types of argument modifiers in PropBank

Question 13

How is FrameNet different to PropBank?

Answer 13

It associates semantic roles with frames rather than verb predicates

Question 14

What is a frame?

Answer 14

A frame is a background knowledge structure that unites a set of words

Question 15

What are frame elements?

Answer 15

They are a set of semantic roles and the predicates that use these roles

Question 16

Explain what is shown in the image

Answer 16

This is a frame and its frame element. The frame is change_position_on_a_scale. The description contains some frame elements that are within the brackets. It gives a description of some components that relate to the concept and each of these components are a frame element

Question 17

What are core and non core roles in FrameNet?

Answer 17

Core roles are frame specific, while non-core roles are some more general properties that can occur across a number of different frames, such as time, location etc

Question 18

What is feature based SRL?

Answer 18

It is one approach to SRL. The training set is PropBank or FrameNet, the input is a set of sentences from these training sets. A node is a node of the parse tree. We use a supervised classifier such as CRF to predict SRL for a node. The input for the classifier, X is a feature template for a specific (node, predicate) in the parse tree. Y is a SRL tag.

Question 19

What is feature-based SRL typically used with?

Answer 19

They are used with a pipeline where you prune the parse tree, identify nodes that are/are not arguments, then classify using a multi-class classifier for nodes

Question 20

What is global consistency?

Answer 20

Semantic roles of constituents are not independent, so per node local role classification is not optimal. Global consistency step is added as a second pass

Question 21

What are some feature templates for SRL?

Answer 21

Governing Predicate, phrase type of constituent, Headword of constituent + POS, Path in parse tree from constituent to predicate, Voice of clause, Position of constituent, subcategorisation of the predicate, NE type, first and last word

Question 22

What is Neural SRL?

Answer 22

It is a neural method to perform semantic role labelling

Question 23

What is a benefit of the Neural SRL model?

Answer 23

It does not require the parse tree as input, only a sentence and instead the parse tree can be embedded within the layers of the NN

Question 24

What is the input for SRL?

Answer 24

A sentence with some separators and an identifier for the predicate

Question 25

What is the output for Neural SRL?

Answer 25

The BIO tags for SRL

Question 26

How does the Neural SRL model work?

Answer 26

Inputs go through embedding layer. We include a flag for the word that is a predicate. This enters the biLSTM layer, as output we pass to a MLP with softmax to classify each of the labels.

Question 27

In Neural SRL, how does the BiLSTM encoder work?

Answer 27

Concatenate the BERT embedding and the predicate flag embedding

Can use transformer instead of BiLSTM

Question 28

In Neural SRL, how does the MLP + softmax decoder work?

Answer 28

Concatenate the hidden layer output for each word and hidden layer output for the predicate.

Can use CRF instead of MLP