Discourse Structure
parsing
Coherence Relations
• Take the following example
• Jane took a train from Paris to Istanbul. She likes spinach.
• Jane took a train from Paris to Istanbul. She had to attend a conference.
• The reason in the second example is more coherent ,the reader can form a connection
between the two sentences, in which the second sentence provides a potential REASON for
the first sentences.
• This link is harder to form for example 1.These connections between text spans in a
discourse can be specified as a set of coherence relations.
• There are two commonly used models of coherence relations and associated corpora:
• Rhetorical Structure Theory (RST),
Rhetorical Structure Theory
• The most commonly used model of discourse organization is Rhetorical Structure
Theory (RST) (Mann and Thompson, 1987).
• In RST relations are defined between two spans of text, generally a nucleus and a
satellite.
• The nucleus is the unit that satellite is more central to the writer’s purpose and that
is interpretable independently;
• the satellite is less central and generally is only interpretable with respect to the
nucleus. Some symmetric relations, however, hold between two nuclei.
• Below are a few examples of RST coherence relations, with definitions adapted from
the RST Treebank Manual
• Reason: The nucleus is an action carried out by an animate agent and the
satellite is the reason for the nucleus.
• [NUC Jane took a train from Paris to Istanbul.] [SAT She had to attend a
conference.]
• Elaboration: The satellite gives additional information or detail about the
situation presented in the nucleus.
• [NUC Dorothy was from Kansas.] [SAT She lived in the midst of the great Kansas
prairies.]
• Evidence: The satellite gives additional information or detail about the situation
presented in the nucleus. The information is presented with the goal of convince
the reader to accept the information presented in the nucleus.
• [NUC Kevin must be here.] [SAT His car is parked outside.]
• Attribution: The satellite gives the source of attribution for an instance of
reported speech in the nucleus.
• [SAT Analysts estimated] [NUC that sales at U.S. stores declined in the quarter,
too]
• List: In this multinuclear relation, a series of nuclei is given, without contrast
or explicit comparison:
• [NUC Billy Bones was the mate; ] [NUC Long John, he was quartermaster]
Graphical representation of RST
relations
• RST relations are traditionally represented graphically;
• the asymmetric Nucleus Satellite relation is represented with an
arrow from the satellite to the nucleus:
Complete Discourse example
• We can also consider the coherence of a larger text by the hierarchical
structure between coherence relations.
• Figure below shows the rhetorical structure of a paragraph from Marcu (2000a)
for the text in from the Scientific American magazine
Example Paragraph:
With its distant orbit–50 percent farther from the sun than Earth–and slim atmospheric blanket, Mars experiences frigid
weather conditions. Surface temperatures typically average about -60 degrees Celsius (-76 degrees Fahrenheit) at the
equator and can dip to -123 degrees C near the poles. Only the midday sun at tropical latitudes is warm enough to thaw
ice on occasion, but any liquid water formed in this way would evaporate almost instantly because of the low
atmospheric pressure.
• The leaves in the Fig. above tree correspond to text spans of a sentence, clause
or EDU phrase that are called elementary discourse units or EDUs in RST; these
units can also be referred to as discourse segments.
• Because these units may correspond to arbitrary spans of text, determining the
boundaries of an EDU is an important task for extracting coherence relations.
• Roughly speaking, one can think of discourse segments as being analogous to
constituents in sentence syntax,
• There are corpora for many discourse coherence models; the RST Discourse
TreeBank (Carlson et al., 2001) is the largest available discourse corpus.
• It consists of 385 English language documents selected from the Penn
Treebank, with full RST parses for each one, using a large set of 78 distinct
relations, grouped into 16 classes.
• RST treebanks exist also for Spanish, German, Basque, Dutch and Brazilian
Portuguese
Penn Discourse TreeBank (PDTB
• The Penn Discourse TreeBank (PDTB) is a second commonly used
dataset that embodies another model of coherence relations
• PDTB labeling is lexically grounded.
• Instead of asking annotators to directly tag the coherence relation
between text spans, they were given a list of discourse connectives,
words that signal discourse relations, like because, discourse
connectives although, when, since, or as a result.
• In a part of a text where these words marked a coherence relation between two
text spans, the connective and the spans were then annotated, as shown below
• Jewelry displays in department stores were often cluttered and uninspired. And
the merchandise was, well, fake. As a result, marketers of faux gems steadily
lost space in department stores to more fashionable rivals—cosmetics makers.
• In July, the Environmental Protection Agency imposed a gradual ban on virtually
all uses of asbestos. (implicit=as a result) By 1997, almost all remaining uses of
cancer-causing asbestos will be outlawed.
• where the phrase as a result signals a causal relationship between what PDTB
calls Arg1 (the first two sentences, here in italics) and Arg2 (the third sentence,
here in bold).
• Not all coherence relations are marked by an explicit discourse connective,
and so the PDTB also annotates pairs of neighboring sentences with no
explicit signal, like example 2 .
• The annotator first chooses the word or phrase that could have been its signal
(in this case as a result), and then labels its sense.
• For example for the ambiguous discourse connective since annotators marked
whether it is using a CAUSAL or a TEMPORAL sense.
The four high-level semantic
distinctions in the PDTB sense
hierarchy
Discourse Structure Parsing
• Given a sequence of sentences, how can we automatically determine
the coherence relations between them?
• This task is often called discourse parsing (even though discourse
parsing for PDTB we are only assigning labels to leaf spans and not
building a full parse tree as we do for RST)
EDU segmentation for RST parsing
• RST parsing is generally done in two stages. The first stage, EDU
segmentation, extracts the start and end of each EDU. The output of
this stage would be a labeling like the following: (22.16) [Mr. Rambo
says]e1 [that a 3.2-acre property]e2 [overlooking the San Fernando
Valley]e3 [is priced at $4 million]e4 [because the late actor Erroll
Flynn once lived there.]e5
RST parsing Tools for building
RST
• coherence structure for a discourse have long been based on syntactic
parsing algorithms like shift-reduce parsing (Marcu, 1999).
• Many modern RST parsers are build on the neural syntactic parsers
which we have seen in previous Chapters
• using representation learning to build representations for each span,
and training a parser to choose the correct shift and reduce actions
based on the gold parses in the training set.
• The parser state consists of a stack and a queue, and produces this
structure by taking a series of actions on the states. Actions include:
• shift: pushes the first EDU in the queue onto the stack creating a single-
node subtree.
• reduce(l,d): merges the top two subtrees on the stack, where l is the
coherence relation label, and d is the nuclearity direction, d ∈ {NN,NS,SN}.
• As well as the pop root operation, to remove the final tree from the stack
Example RST discourse tree
showing four EDUs:
• Fig. below shows the actions the parser takes to build the structure in
Fig. above
Centering Theory
• At any point in the discourse, one of the entities in the discourse
model is salient (being “centered” on)
• Discourses in which adjacent sentences continue to maintain the
same salient entity are more coherent than those which shift back and
forth between multiple entities
Centering Theory: Intuition
• The following two texts from Grosz et al. (1995) which have exactly the same
propositional content but different saliences, can help in understanding the main
Centering intuition.
• a. John went to his favorite music store to buy a piano.
• b. He had frequented the store for many years.
• c. He was excited that he could finally buy a piano.
• d. He arrived just as the store was closing for the day.
• a. John went to his favorite music store to buy a piano.
• b. It was a store John had frequented for many years.
• c. He was excited that he could finally buy a piano.
• d. It was closing just as John arrived.
Centering Theory: Intuition
• While these two texts differ only in how the two entities (John and
the store) are realized in the sentences, the discourse in First example
is intuitively more coherent than the one in second example .
How does Centering Theory
realize this intuition?
