Presentation is loading. Please wait.

Presentation is loading. Please wait.

Packed Computation of Exact Meaning Representations

Similar presentations


Presentation on theme: "Packed Computation of Exact Meaning Representations"— Presentation transcript:

1 Packed Computation of Exact Meaning Representations
Iddo Lev Department of Computer Science Stanford University

2 Iddo Lev, Packed Computation of Exact Meaning Representations
Outline  Motivation  Motivation  From Syntax to Semantics  Packed Computation  Conclusion April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

3 Natural Language Understanding
How can we improve accuracy? Let’s take it for a moment to the extreme Exact NLU applications April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

4 Example: Logic Puzzles
Six sculptures—C, D, E, F, G, and H—are to be exhibited in rooms 1, 2, and 3 of an art gallery. Sculptures C and E may not be exhibited in the same room. Sculptures D and G must be exhibited in the same room. If sculptures E and F are exhibited in the same room, no other sculpture may be exhibited in that room. At least one sculpture must be exhibited in each room, and no more than three sculptures may be exhibited in any room. 1. If sculpture D is exhibited in room 3 and sculptures E and F are exhibited in room 1, which of the following may be true? (A) Sculpture C is exhibited in room 1. (B) No more than 2 sculptures are exhibited in room 3. (C) Sculptures F and H are exhibited in the same room. (D) Three sculptures are exhibited in room 2. (E) Sculpture G is exhibited in room 2. April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

5 Example: Logic Puzzles
If sculptures E and F are exhibited in the same room, no other sculpture may be exhibited in that room. x.[(room(x)  exhibited-in(E,x)  exhibited-in(F,x))  ¬y.sculpture(y)  y  E  y  F  exhibited-in(y,x)] exact meaning representation: April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

6 Iddo Lev, Packed Computation of Exact Meaning Representations
Example: MSCS Degree Requirements A candidate is required to complete a program of 45 units. At least 36 of these must be graded units, passed with an average 3.0 (B) grade point average (GPA) or higher. The 45 units may include no more than 21 units of courses from those listed below in Requirements 1 and 2. … Has Patrick Davis completed the program? Can/must Patrick Davis take CS287? Similar to logic puzzles: General constraints + specific situation April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

7 Iddo Lev, Packed Computation of Exact Meaning Representations
Exact NLU More examples Word problems Logic puzzles Math, physics, chemistry questions Simple regulation texts, controlled language NL interfaces to databases Like SQL, but looks like NL In these tasks “Almost correct” (“only slightly wrong”) is not good enough Simple approximations won’t do E.g. syntactic matching between text and questions Because answer does not appear explicitly in the text Need exact calculation of NL meaning representations Answer needs to be inferred from the text Need to carefully combine information/meaning throughout the text April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

8 Iddo Lev, Packed Computation of Exact Meaning Representations
Structural Semantics Need to rely on high-quality meaning representations and linguistic knowledge In particular, structural semantics Meaning of functional words Logical structure of sentences Essential for exact NLU tasks Could also improve precision of other NLP tasks T: Michael Melvill guided a tiny rocket-ship more than 100 kilometers above the Earth. H: A rocket-ship was guided more than 80 kilometers above the Earth. Follows H: A rocket-ship was guided more than 120 kilometers above the Earth. Does not follow Relatively small size of knowledge Functional: #functional words  400 #grammar rules  400 Lexical: #verb frames  45,000 #nouns > 100,000 April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

9 Iddo Lev, Packed Computation of Exact Meaning Representations
My Dissertation How to map syntactic analysis to meaning representations How to compute all meaning representations efficiently Linguistic analysis of advanced NL constructions using the above framework anaphora (interaction with truth conditions) comparatives reciprocals (each other, one another) same/different How to translate meaning representations to inference representations (FOL) Focus of this talk April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

10 Structural Semantics Challenges
When analyzing one sentence: (1) Bills 2 and 6 are paid on the same day as each other. it might seem enough to use: x.day(x)paid-on(bill2,x)paid-on(bill6,x) But this is not enough when we consider other sentences: (2) John, Mary, and Frank like each other. each_other({john,mary,frank}, xy.like(x,y)) Goal Uniformity: one analysis of “each other” for both (1) and (2). Should interact correctly with “the same” in (1). Solution should also be consistent with “different”, “similar”: Men and women have a different sense of humor (than each other). April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

11 Iddo Lev, Packed Computation of Exact Meaning Representations
Outline  Motivation  From Syntax to Semantics  Packed Computation  Conclusion April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

12 From Syntax to Semantics
How do we get from one parse tree to a semantic representation? Classic Method (Montague): one-to-one correspondence: assign a lambda-term to each syntactic node S x. [dog(x)  bark(x)] λR. x. [dog(x)  R(x)] NP VP | V barks λz. bark(z) Det every λP.λR. x. [P(x)  R(x)] Noun dog λy.dog(y) April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

13 Problem 1: Floating Operators
NP VP V NP PP NP Frank introduced Rachel to Patrick. introduce-to(frank, rachel, patrick) April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

14 Problem 1: Floating Operators
NP VP V NP PP NP N’ RC VP Det N VP NP Frank introduced Rachel to every person who visited me that summer. every(λx.person(x)visit(x,me), λx.introduce-to(frank, rachel, x)) every(P,Q) x. [P(x)  Q(x)] April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

15 Problem 1: Floating Operators
A brave sailor walked by. a(λx.[sailor(x)brave(x)], λx.walk-by(x)) S NP N’ N VP Adj An occasional sailor walked by. occasionally(a(λx.sailor(x), λx.walk-by(x))) S NP N’ N VP Adj April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

16 Problem 2: More Than One Meaning
“In this country, a woman gives birth every 15 minutes. Our job is to find that woman, and stop her.” -- Groucho Marx every 15 minutes a woman gives birth a woman every 15 minutes gives birth You may not smoke. You may not succeed. All these books are not interesting. All that glitters is not gold. April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

17 Iddo Lev, Packed Computation of Exact Meaning Representations
Glue Semantics Glue Semantics: A flexible framework for mapping syntax to semantics Pieces of syntax correspond to pieces of semantics Pieces of semantics combine with each other according to constraints Like jigsaw puzzle, but possibly with more than one solution Not a simple one-to-one mapping References Dalrymple et al. Semantics and Syntax in Lexical Functional Grammar Mary Dalrymple. Lexical Functional Grammar Asudeh, Crouch, Dalrymple. The syntax-semantics interface. 2002 April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

18 Iddo Lev, Packed Computation of Exact Meaning Representations
Glue Semantics (simplified example) S NP VP NP Name V Name statements John saw Mary mary xy.see(x,y) john April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

19 Iddo Lev, Packed Computation of Exact Meaning Representations
Glue Semantics (simplified example) S a b NP VP NP c Name V Name statements John saw Mary mary : c xy.see(x,y) : b  c  a john : b prover derivation b b  c  a c c  a a john : xy.saw(x,y) : y.saw(john,y) : mary : saw(john,mary) : gain: order of combination does not have to follow tree hierarchy April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

20 Problem 1: Floating Operators
NP VP NP VP λx.[sailor(x)brave(x)] λQλR.occasionally[Q(λx.sailor(x),R)] N’ N’ Adj N Adj N A brave sailor walked by. An occas. sailor walked by. λPλR.a(P,R) λx.sailor(x) λx.walk-by(x) λPλR.a(P,R) λx.sailor(x) λx.walk-by(x) λPλx.[P(x)brave(x)] λPλQλR.occasionally[Q(P,R)] a(λx.[sailor(x)brave(x)], λx.walk-by(x)) occasionally[a(λx.sailor(x), λx.walk-by(x))] April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

21 Iddo Lev, Packed Computation of Exact Meaning Representations
Glue Semantics S a Flexible handling of floating operators. b NP VP N’ c λPλR.a(P,R) : c  (b  a)  a Adj N An occas. sailor walked by. λx.sailor(x) : c λx.walk-by(x) : b  a occasionally[a(λx.sailor(x), λx.walk-by(x))] c c  (b  a)  a b  a (b  a)  a a a  a a λS.occasionally[S] : a  a April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

22 Iddo Lev, Packed Computation of Exact Meaning Representations
Glue Semantics (simplified example) Can yield more than one meaning. A woman gives birth every 15 minutes. “gives birth” G : a “a woman” A : a  a “every 15 minutes” E : a  a two possible derivations: G : a A : a  a A(G) : a E : a  a E(A(G)) : a G : a E : a  a E(S) : a A : a  a A(E(S)) : a April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

23 Glue Semantics Shared labels constrain how statements combine
“Resource Sensitive”: Use each statement exactly once Inference rules: Application :A :AB ():B Abstraction [x:A] :B x.:AB Linear Logic (implicative fragment) In Glue Semantics, can impose further constraints on combinations. April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

24 Iddo Lev, Packed Computation of Exact Meaning Representations
Outline  Motivation  From Syntax to Semantics  Packed Computation  Conclusion April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

25 Iddo Lev, Packed Computation of Exact Meaning Representations
Ambiguity Flying planes can be dangerous. Therefore, only licensed pilots are allowed to do it. Flying planes can be dangerous. Therefore, some people are afraid to ride in them. We cannot always disambiguate the sentence just by looking at the sentence itself. We sometimes need to take the larger context and information into account. April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

26 Iddo Lev, Packed Computation of Exact Meaning Representations
Ambiguity Alternatives multiply across layers… Text Morphology Syntax Semantics KR Reasoning … so we can’t keep all the alternatives separately April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

27 Iddo Lev, Packed Computation of Exact Meaning Representations
Early Pruning Select most likely analysis at each level X Oops: Strong constraints may reject the so-far-best (and only) option Statistics X X X Text Morphology Syntax Semantics KR Reasoning Locally less likely option but globally correct April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

28 Packing The sheep liked the fish. More than one sheep?
More than one fish? The sheep-sg liked the fish-sg. The sheep-pl liked the fish-sg. The sheep-sg liked the fish-pl. The sheep-pl liked the fish-pl. Options multiplied out The sheep liked the fish sg pl Options packed Packed representation: Encodes all analyses without loss of information Common items represented and computed just once April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

29 Iddo Lev, Packed Computation of Exact Meaning Representations
Packing Calculate compactly all analyses at each stage Push ambiguities through the stages Possibly, filter and keep only N-best at each stage in a packed form (not only 1-best) This approach is being pursued in the XLE system at PARC (and Powerset Inc.) (Maxwell & Kaplan ’89,93,95) April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

30 Packing In Syntax: Chart Parser
A chart parser for a context-free grammar can compute an exponential number of parse trees in O(n3) time by representing and computing them compactly. Instead of separately: we have: April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

31 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Structures XLE manages natural language ambiguity by packing similar structures and managing them under a free-choice space C-structure forest Packed F-structure Choice Space: true  A1  A2 A1  A2  false April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

32 Currently in XLE : : Text morph. C-str F-str MR KR answer
semantic rewrite rules FST morph. glue spec. glue prover parser unpack F-str1 Glue1 MR1 pack : : C-str F-str MR KR C-F F-strn Gluen MRn answer = packed = packed calculation + possibly filter N-best April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

33 Iddo Lev, Packed Computation of Exact Meaning Representations
The Goal Text FST morph. parser Glue statements C-str F-str MR KR glue spec. glue prover C-F answer = packed = packed calculation + possibly filter N-best April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

34 Goal: Packed Meaning Representation
Bill saw the girl with the telescope. a:1 e. see(e)  agent(e,bill)  theme(e,the(x.girl(x))  with(e,the(y.tele(y))) a:2 e. see(e)  agent(e,bill)  theme(e, the(x. girl(x)  with(x,the(y.tele(y))) ) e. see(e)  agent(e,bill)  ● girl(x) theme(e, the(x. ● )) ●● with(●,the(y.tele(y))) x e packed meaning representation April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

35 Iddo Lev, Packed Computation of Exact Meaning Representations
Glue Specification Glue specification – connecting syntactic and semantic pieces NTYPE(f, NAME), PRED(f, p)  p : f NTYPE(f, COMMON), PRED(f, p)  λx.p(x) : fv  fr F-Structure glue statements john : a λx.cake(x) : bv  br “John ate the cake.” April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

36 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Glue Input Packed F-structure NTYPE(f, NAME), PRED(f, p)  p : f e Glue specification {1} e.see(e) : aveart {2} P.e.P(e) : (aveart)at {3} bill : be {4} xPe.P(e)agent(e,x) : be(aveart)(aveart) {5} P.the(P) : (gvegrt)ge {6} x.girl(x) : gvegrt {7} xPe.P(e)theme(e,x) : ge(aveart)(aveart) {8} P.the(P) : (hvehrt)he {9} x.tele(x) : hvehrt {10} A1: yPe.P(e)with(e,y) : he(aveart)(aveart) {11} A2: yPx.P(x)with(x,y) : he(gvegrt)(gvegrt) “Bill saw the girl with the telescope.” This combines Glue Semantics + packing at the input level April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

37 Non-packed Prover (Hepple’96)
Rule: Input:  : A | S  : AB | S2 () : B | S1  S2 f : c c d q : c r : c provided S1  S2 =  Chart: meaning category span f c c d {1} q c {2} r f(q) cd {3} {1,2} f(r) {1,3} f(q,r) d {1,2,3} f(r,q) cannot combine: {2}{1,2}   complete derivation (all indices were used) Output: April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

38 Iddo Lev, Packed Computation of Exact Meaning Representations
Syntactic Ambiguity “Time flies like an arrow. Fruit-flies like a banana.” -- Groucho Marx April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

39 Naive Packed Algorithm
A1: [[John]a thinks that [[time]d flies [like [an arrow]c]]b]g A2: [[John]a thinks that [[time flies]f like [an arrow]c]b]g The chart algorithm will discover one history for [[time]d flies [like [an arrow]c]]b under A1 It may then continue under A1 with “John thinks that” It will later discover a history for [[time flies]f like [an arrow]c]b under A2 So it will have to redo the work for “John thinks that” under A2 April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

40 Iddo Lev, Packed Computation of Exact Meaning Representations
Non-Packed Prover Forget about meaning terms for now (can reconstruct them after the derivation finishes) Combine histories according to topological order of category graph {2} {3} {1} {4} {5} {6} category graph a ab ac b c bcd cd d df f mean. category span q ab {1} p a {2} r s ac {3} {4} t bcd {5} u df {6} {1,2} {1,3} {2,4} {3,4} {1,2,5} {1,3,5} t(q(p),s(r)) t(q(r),s(p)) {1,2,3,4,5} {1,2,3,4,5,6} April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

41 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation (Simplified example) A1: [[John]a thinks that [[time]d flies [like [an arrow]c]]b]g A2: [[John]a thinks that [[time flies]f like [an arrow]c]b]g premises choice john a {1} 1 think abg {2} anarrow c {3} time d {4} A1 fly deb {5} like ce {6} timeflies f {7} A2 fcb {8} April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

42 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation (Simplified example) A1: [[John]a thinks that [[time]d flies [like [an arrow]c]]b]g A2: [[John]a thinks that [[time flies]f like [an arrow]c]b]g premises choice john a {jn} 1 think abg {th} anarrow c {ar} time d {t} A1 fly deb {f} like ce {k1} timeflies f {tf} A2 fcb {k2} April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

43 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation  Imagine how each derivation works separately; then figure out how to pack. premises choice john a {jn} 1 think abg {th} anarrow c {ar} time d {t} A1 fly deb {f} like ce {k1} timeflies f {tf} A2 fcb {lk2} {f} {t} {k1} {ar} {k1,ar} {th} {jn} {t,f} {t,f,k1,ar} {jn,th} {jn,th,t,f,k1,ar} Category graph April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

44 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation  Imagine how each derivation works separately; then figure out how to pack. {tf} {k2} premises choice john a {jn} 1 think abg {th} anarrow c {ar} time d {t} A1 fly deb {f} like ce {lk1} timeflies f {tf} A2 fcb {k2} {ar} {tf,k2} {tf,k2,ar} {th} {jn} {jn,th} {jn,th,tf,k2,ar} Category graph April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

45 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation  Imagine how each derivation works separately; then figure out how to pack. {tf} {k2} premises choice john a {j} 1 think abg {th} anarrow c {a} time d {t} A1 fly deb {f} like ce {k1} timeflies f {tf} A2 fcb {k2} {f} {t} {k1} {ar} {tf,k2} {k1,ar} {th} {jn} {t,f} A2:{tf,k2,ar} A1:{t, f,k1,ar} {jn,th} 1:{ar} A1:{t, f,k1} A2:{tf,k2} {jn,th,t,f,k1,ar} {jn,th,tf,k2,ar} 1:{jn,th,ar} A1:{t, f,k1} A2:{tf,k2} April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

46 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation only possible in A1 history under A1 under A2 packed span h1 {ar} 1:{ar} h2 {k1,ar} A1:{k1,ar} h3 {t,f} A1:{t,f} h4 {t,f,k1,ar} A1:{t,f,k1,ar} h5 {tf,k2} A2:{tf,k2} h6 {tf,k2,ar} A2:{tf,k2,ar} h7 1:{ar} A1:{t,f,k1} A2:{tf,k2} h8 {jn,th} 1:{jn,th} h9 {jn,th,t,f,k1,ar} {jn,thtf,k2,ar} 1:{jn,th,ar} A1:{t,f,k1} A2:{tf,k2} packed common part {tf} {k2} {f} {t} {k1} {ar} {tf,k2} {k1,ar} {t,f} A2:{tf,k2,ar} {jn} {th} A1:{t, f,k1,ar} {jn,th} 1:{ar} A1:{t, f,k1} A2:{tf,k2} 1:{jn,th,ar} A1:{t, f,k1} A2:{tf,k2} April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

47 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation Two histories with categories A and AB can be combined: original algorithm: if their spans are disjoint packed algorithm: can combine them in all contexts in which their spans are disjoint original combination: provided S1  S2 =  and S = S1  S2 A | S AB | S2 B | S packed combination: provided C1  C2  0 and combinable(PS1, PS2, C) and PS = union(C, PS1, PS2) C1 | A | PS C2 | AB | PS2 C | B | PS April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

48 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation Two histories with categories A and AB can be combined: original algorithm: if their spans are disjoint packed algorithm: can combine them in all contexts in which their spans are disjoint packed combination: provided C1  C2  0 and combinable(PS1, PS2, C) and PS = union(C, PS1, PS2) C1 | A | PS C2 | AB | PS2 C | B | PS combinable: 1:{3,4} 1:{5,6,7} combinable: 1:{3},A1:{6,7} 1:{4,5},A2:{6,8} combinable: A1:{6},A2:{7}  A1:{6},A2:{8} non-combinable: 1:{4},A1:{6} 1:{5,6},A2:{4} (6 is in A1 in both, 4 is in A2 in both)  1:{3,4,5,6,7}  1:{3,4,5,6},A1:{7},A2:{8}  A2:{7,8} A1:{4,6} A2:{4} A1:{5,6} A2:{4,5,6} April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

49 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation Two histories with the same category can be packed: original algorithm: if their spans are identical packed algorithm: if their spans are identical in the shared contexts can pack: 1:{3,4,5} 1:{3,4,5} can pack: A1:{1},A2:{2} A2:{2},A3:{3} can pack: A1:{t,f,k1,ar} A2:{tf,k2,ar} cannot pack: 1:{5},A1:{6} 1:{5},A2:{7} ({5,6}{5} in A1 , {5}{5,7} in A2))  1:{3,4,5}  A1:{1},A2:{2},A3:{3}  1:{ar}, A1:{t,f,k1}, A2:{tf,k2} A1:{5,6} A2:{5} A1:{5} A2:{5,7} A2:{tf,k2,ar} A1:{t, f,k1,ar} 1:{ar} A1:{t, f,k1} A2:{tf,k2} {ar} April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

50 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation Reconstruction of packed meaning representation: history packed span meaning h1 1:{ar} l1 : anarrow h2 A1:{k1,ar} like(l1) h3 A1:{t,f} fly(time) h4 A1:{t,f,k1,ar} fly(time,like(l1)) h5 A2:{tf,k2} like(timeflies) h6 A2:{tf,k2,ar} like(timeflies,l1) h7 1:{ar} A1:{t,f,k1} A2:{tf,k2} A1:fly(time,like(l1)) A2:like(timeflies,l1) h8 1:{jn,th} think(john) h9 1:{jn,th,ar} A1:{t,f,k1} A2:{tf,k2} think(john, ●) anarrow fly(time,like(●)) like(timeflies,●) A1 A2 category graph packed meaning representation April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

51 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation What if the category graph has cycles? Calculate strongly connected components (SCCs) and the induced directed-acyclic graph (DAG) (+ topological sort) In each SCC, run basic algorithm to find all possibilities If SCC is simple (X, XX) then optimize: use as much material as possible before moving out of the cycle {2} {3} {4} {1} XX X {1} {1,2} {1,3} {1,4} {1,2,3} {1,2,4} {1,3,4} {1,2,3,4} category graph April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

52 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation [girl [with the telescope]]A2 A2:{wt} 1:{grl} XX X 1:{grl} A1:{grl} A2:{grl,wt} 1:{grl}, A2:{wt} category graph April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

53 Iddo Lev, Packed Computation of Exact Meaning Representations
Packed Derivation Category graph {1} e.see(e) : aveart {2} P.e.P(e) : (aveart)at {3} bill : be {4} xPe.P(e)agent(e,x) : be(aveart)(aveart) {5} P.the(P) : (gvegrt)ge {6} x.girl(x) : gvegrt {7} xPe.P(e)theme(e,x) : ge(aveart)(aveart) {8} P.the(P) : (hvehrt)he {9} x.tele(x) : hvehrt {10} A1: yPe.P(e)with(e,y) : he(aveart)(aveart) {11} A2: yPx.P(x)with(x,y) : he(gvegrt)(gvegrt) 1:{8} 1:{9} (hvehrt)he hvehrt A2:{11} he(gvegrt)(gvegrt) he 1:{8,9} A2:{8,9,11} (gvegrt)(gvegrt) (gvegrt)ge gvegrt 1:{6},A2:{8,9,11} packing in a cycle A1:{10} ge geaetaet 1:{3} heaetaet beaetaet be Need to calculate strongly-connected components before topological sort. A1:{8,9,10} 1:{5,6,7}, A2:{8,9,11} 1:{3,4} aetaet 1:{2} aetat aet 1:{1,3,4,5,6,7,8,9}, A1:{10},A2:{11} at 1:{1,2,3,4,5,6,7,8,9}, A1:{10},A2:{11} April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

54 Iddo Lev, Packed Computation of Exact Meaning Representations
Outline  Motivation  From Syntax to Semantics  Packed Computation  Conclusion April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

55 Iddo Lev, Packed Computation of Exact Meaning Representations
My Dissertation How to map syntactic analysis to meaning representations How to compute all meaning representations efficiently Linguistic analysis of advanced NL constructions using the above framework anaphora (interaction with truth conditions) comparatives reciprocals (each other, one another) same/different How to translate meaning representations to inference representations (FOL) Focus of this talk April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

56 Iddo Lev, Packed Computation of Exact Meaning Representations
Summary Mapping syntax to exact meaning representations using Glue Semantics More powerful than traditional approach Easier for users, more principled than semantic rewrite rules Covered advanced NL constructions Computing all meaning representations efficiently Input: packed syntactic analysis Output: packed meaning representation  Pushing packed ambiguities through the semantics stage April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

57 Iddo Lev, Packed Computation of Exact Meaning Representations
Future Work Researchers can use this work as a basis Use this in applications Logic puzzles, word problems, NLIDB, regulation texts Extend this approach to additional NL constructions (requires some linguistic research) Extend idea of packing to anaphora/plurality and back-end inference stages Some initial work on packed reasoning at PARC Extend statistical disambiguation to packed semantic structures April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations

58 Iddo Lev, Packed Computation of Exact Meaning Representations
Thanks Stanley Peters Dick Crouch Chris Manning Mike Genesereth Johan van Benthem NLTT group at PARC Ivan Sag Bill MacCartney, Mihaela Enachescu, Powerset Inc. Beth Nowadnick April 17, 2007 Iddo Lev, Packed Computation of Exact Meaning Representations


Download ppt "Packed Computation of Exact Meaning Representations"

Similar presentations


Ads by Google