Models of Discourse Structure II Discourse & Dialogue CMSC October 14, 2004

Invitation Workshop on –Learning and Multimodal Communication Thursday/Friday 9:30-5pm Research Institutes: RI 480 (Access Grid) Lunch included, registration free –

Next Week Leading discussion: –Automatic extraction of discourse structure 1 speech primary (G&S) 1 speech+text (LM/HMM) 1 text in VSM framework 1 text in MaxEnt 1 text RST Volunteers?

Roadmap Models of Discourse Structure –Attention & Intentions (Grosz & Sidner 86) –Rhetorical Structure Theory (Mann & Thompson 87) Contrasts, Constraints & Conclusions

Questions What types of discourse is the model best suited to? Claim to capture? How could one extract the structure? Manually? Automatically? –What features, cues would you need? What applications are the models best suited to? Recognition? Generation? Synthesis? Summarization?

Attention, Intentions and the Structure of Discourse Grosz&Sidner (1986) Goals: –Integrate approaches for focus (reference res.), plan/task structure, discourse structure, goals Three part model: –Linguistic structure (utterances) –Attentional structure (focus, reference) –Intentional structure (plans, purposes)

Linguistic Structure Utterances group into discourse segments –Hierarchical, not necessarily contiguous –Not strictly decompositional 2-way interactions –Utterances define structure; Cue phrases mark segment boundaries –But, okay, fine, incidentally –Structure guides interpretation –Reference

Intentional Structure Discourse & participants: overall purpose –Discourse segments have purposes (DP/DSP) Contribute to overall Main DP/DSP intended to be recognized

Intentional Structure: Relations Two relations between purposes –Dominance DSP1 dominates DSP2 if doing DSP2 contributes to achieving DSP1 –Satisfaction-Precedence DSP1 must be satisfied before DSP2 Purposes: –Intend that someone know something, do something, believe something, etc –Open-ended

Attentional State Captures focus of attention in discourse –Incremental –Focus Spaces Include entities salient/evoked in discourse Include a current DSP Stack-structured: –higher->more salient, lower still accessible –Push:segment contributes to previous DSP –Pop: segment to contributes to more dominant DSP »Tied to intentional structure

Attentional State cntd. Focusing structure depends on the intentional structure: the relationships between DSPs determine pushes and pops from the stack Focusing structure coordinates the linguistic and intentional structures during processing Like the other 2 structures, focusing structure evolves as discourse proceeds

Discourse examples Essay Task-oriented dialog –Intentional structure is neither identical nor isomorphic to the general plan

The "movies" are so attractive to the great American public, especially to young people, that it is time to take careful thought about their effect on mind and morals. Ought any parent to permit his children to attend a moving picture show often or without being quite certain of the show he permits them to see? No one can deny, of course, that great educational and ethical gains may be made through the movies because of their astonishing vividness. But the important fact to be determined is the total result of continuous and indiscriminate attendance on shows of this kind. Can it other than harmful? In the first place the character of the plays is seldom of the best. One has only to read the ever-present "movie" billboard to see how cheap, melodramatic and vulgar most of the photoplays are. Even the best plays, moreover, are bound to be exciting and over-emotional. Without spoken words, facial expression and gesture must carry the meaning: but only strong emotion or buffoonery can be represented through facial expression and gesture. The more reasonable and quiet aspects of life are necessarily neglected. How can our young people drink in through their eyes a continuous spectacle of intense and strained activity and feeling without harmful effects? Parents and teachers will do well to guard the young against overindulgence in the taste for the "movie"

H:1. First you have to remove the flywheel. R:2. How do I remove the flywheel? H:3. First, loosen the screw, then pull it off. R:4. OK.5. The tool I have is awkward. Is there another tool that I could use instead? H:6. Show me the tool you are using. R:7. OK. H:8. Are you sure you are using the right size key? R:9. I’ll try some others. 10. I found an angle I can get at it. 11. The screw is loose, but I’m having trouble getting the flywheel off. H:12. Use the wheelpuller. Do you know how to use it ? R:13. No. H:14. Do you know what it looks like? R:15. Yes. H:16. Show it to me please. R:17. OK. H:18. Good. Loosen the screw in the center and place the jaws around the hub of the flywheel, then tighten the screw onto the center of the shaft. The flywheel should slide off.

Processing issues Intention recognition –What info can be used to recognize an intention –At what point does this info become available Overall processing module has to be able to operate on partial information It must allow for incrementally constraining the range of possibilities on the basis of new info that becomes available as the segment progresses

Info constraining DSP: –Specific linguistic markers –Utterance-level intentions –General knowledge about actions and objects in the domain of discourse Applications of the theory: –Interruptions Weak – not linked to immediate DSP Strong - not linked to any DSP –Cue words

Interruption John came by and left the groceries Stop that you kids And I put them away after he left John, groceries DSP1 kids DSP2 John, groceries DSP1

Conclusions Generalizes approaches to task-oriented dialogue –Goal: Domain-independence –Broad, general, abstract model Accounts for interesting phenomena –Interruptions, returns, cue phrases

More conclusions Asks more questions than it answers. How do we implement these aspects of dialog? –Is it remotely feasible????

Rhetorical Structure Theory Mann & Thompson (1987) Goal: Identify hierarchical structure of text –Cover wide range of TEXT types Language contrasts –Relational propositions (intentions) Derives from functional relations b/t clauses

Components of RST Relations: –Hold b/t two text spans, nucleus and satellite Constraints on each, between Effect: why the author wrote this Schemas: –Grammar of legal relations between text spans –Define possible RST text structures Most common: N + S, others involve two or more nuclei Schema applications: Structures: –Using clause units, complete, connected, unique, adjacent

RST Relations Core of RST –RST analysis requires building tree of relations –Circumstance, Solutionhood, Elaboration. Background, Enablement, Motivation, Evidence, Justify, Vol. Cause, Non-Vol. Cause, Vol. Result, Non-Vol. Result, Purpose, Antithesis, Concession, Condition, Otherwise, Interpretation, Evaluation, Restatement, Summary, Sequence, Contrast

Nuclearity Many relations between pairs asymmetrical –One is incomprehensible without other –One is more substitutable, more important to W Deletion of all nuclei creates gibberish –Deletion of all satellites is just terse, rough Demonstrates role in coherence

RST Relations Evidence –Effect: Evidence (Satellite) increases R’s belief in Nucleus The program really works. (N) I entered all my info and it matched my results. (S) Justify –Effect: Justify (Satellite) increases R’s willingness to accepts W’s authority to say Nucleus The next music day is September 1.(N) I’ll post more details shortly. (S) 12 Evidence

RST Relations Concession: –Effect: By acknowledging incompatibility between N and S, increase Rs positive regard of N Often signaled by “although” –Dioxin: Concerns about its health effects may be misplaced.(N1) Although it is toxic to certain animals (S), evidence is lacking that it has any long-tern effect on human beings.(N2) Elaboration: –Effect: By adding detail, S increases Rs belief in N

RST-relation example (1) 1. Heavy rain and thunderstorms in North Spain and on the Balearic Islands. 2. In other parts of Spain, still hot, dry weather with temperatures up to 35 degrees Celcius. CONTRAST Symmetric (multiple nuclei) Relation:

RST-relation example (2) 2. In Cadiz, the thermometer might rise as high as 40 degrees. 1. In other parts of Spain, still hot, dry weather with temperatures up to 35 degrees Celcius. ELABORATION Asymmetric (nucleus-satellite) Relation:

1) What if you're having to clean floppy drive heads too often? 2) Ask for SYNCOM diskettes, with burnished Ectype coating and dust- absorbing jacket liners. 3) As you floppy drive writes or reads, 4) a Syncom diskette is working four ways to keep loose particles and dust from causing soft errors, dropouts. 5) Cleaning agents on the burnished surface of the Ectype coating actually remove build-up from the head, 6) while lubricating it at the same time. 7) A carbon additive drains away static electricity 8) before it can attract dust or lint. 9) Strong binders hold the signal-carrying oxides tightly within the coating. 10) And the non-woven jacket liner, 11) more than just wiping the surface, provides thousands of tiny pockets to keep what it collects. 12) To see which Syncom diskette will replace the ones you're using now, 13) send for our free "Flexi-finder" selection guide and the name of the supplier nearest you.

Issues Goal: Single tree-shaped analysis of all text –Difficult to achieve Significant ambiguity Significant disagreement among labelers Relation recognition is difficult –Some clear “signals”, I.e. although –Not mandatory, only 25%

Contrasts ?

G&S –Participants Mono, dial, multi-party –Relations 2: Dom, SP –Speech issues Address some RST –Participants Monologue –Relations Myriad –Speech issues Largely ignored

Similarities ?

Structure –Predominantly hierarchical, tree-structured –Some notion of core intentions/topics Intentions of speaker and hearer –Co-constraint of discourse structure and ling form

Questions Which would be better for generation? –For recognition?