Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Purely distributional meaning • High-dimensional • Meaning from dense linguistic inter-relationships • Meaning solely from ( n th-order) co-occurrence 8 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Purely distributional meaning • High-dimensional • Meaning from dense linguistic inter-relationships • Meaning solely from ( n th-order) co-occurrence • No grounding in physical or social contexts 8 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Purely distributional meaning • High-dimensional • Meaning from dense linguistic inter-relationships • Meaning solely from ( n th-order) co-occurrence • No grounding in physical or social contexts • Not symbolic 8 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Faruqui et al. (2015): Retrofitting to graphs � 2 + � 2 � � � � � � � q i − ˆ q i � q i − q j α i β ij i ∈ V ( i , j , r ) ∈ E Balances fidelity to the against looking more like original vector ˆ one’s graph neighbors. q i 1 Forces are balanced with α = 1 and β = Degree ( i ) 9 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Faruqui et al. (2015): Retrofitting to graphs � 2 + � 2 � � � � � � � q i − ˆ q i � q i − q j α i β ij i ∈ V ( i , j , r ) ∈ E Balances fidelity to the against looking more like original vector ˆ one’s graph neighbors. q i 1 Forces are balanced with α = 1 and β = Degree ( i ) 9 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier What retrofitting to WordNet might do 10 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier What retrofitting to WordNet might do • Cluster mammal with dog and puppy even though mammal has a different, unusual distribution. 10 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier What retrofitting to WordNet might do • Cluster mammal with dog and puppy even though mammal has a different, unusual distribution. • Avoid polarity mistakes like modeling superb and awful as similar (though beware antonym edges!). 10 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier What retrofitting to WordNet might do • Cluster mammal with dog and puppy even though mammal has a different, unusual distribution. • Avoid polarity mistakes like modeling superb and awful as similar (though beware antonym edges!). • Holistic consistency: Figure 3: Two-dimensional PCA projections of 100-dimensional SG vector pairs holding the “adjective to adverb” relation, before (left) and after (right) retrofitting. 10 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Concerns about identity retrofitting • No attention to edge Kingsfoil semantics; edges mean Treats ‘similar to’. Is • Presupposes a uniform Treats Black Athelas initial embedding Breath space. Uses Causes • No modeling of missing Aragorn Nazgûl edges. 11 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Hand-build functions from Mrkˇ si´ c et al. (2016) • AntonymRepel: � ReLU � 1 . 0 − d ( q i , q j ) � ( i,j ) ∈ A • SynonymAttract: � ReLU � d ( q i , q j ) − 0 � ( i,j ) ∈ S • VectorSpacePreservation: � � ReLU � d ( q i , q j ) − d ( ˆ q i , ˆ � q j ) i j ∈ N ( i ) 12 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Functional relations (Lengerich et al. 2018) Framework � 2 + � � � � � � � q i − ˆ β ijr f r ( q i , q j ) − α i q i β ijr f r ( q i , q j ) + λ ρ ( f r ) ( i , j , r ) ∈ E − r i ∈ V ( i , j , r ) ∈ E 13 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Functional relations (Lengerich et al. 2018) Framework � 2 + � � � � � � � q i − ˆ β ijr f r ( q i , q j ) − α i q i β ijr f r ( q i , q j ) + λ ρ ( f r ) ( i , j , r ) ∈ E − r i ∈ V ( i , j , r ) ∈ E Faruqui et al. � 2 � � f r ( q i , q j ) = � q i − q j with β ijr = 0 13 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Functional relations (Lengerich et al. 2018) Framework � 2 + � � � � � � � q i − ˆ β ijr f r ( q i , q j ) − α i q i β ijr f r ( q i , q j ) + λ ρ ( f r ) ( i , j , r ) ∈ E − r i ∈ V ( i , j , r ) ∈ E Linear � 2 � � f r ( q i , q j ) = � A r q j + b r − q i • ρ ( f r ) = � A r � 2 • We initialize A r = 1 and b r = 0 • Initialization can be different for different relations, e.g., A antonym = − 1 13 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Functional relations (Lengerich et al. 2018) Framework � 2 + � � � � � � � q i − ˆ β ijr f r ( q i , q j ) − α i q i β ijr f r ( q i , q j ) + λ ρ ( f r ) ( i , j , r ) ∈ E − r i ∈ V ( i , j , r ) ∈ E Simplest neural (akin to Sutskever et al. 2009) f r ( q i , q j ) = tanh ( q ⊤ i A r q j ) 13 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Functional relations (Lengerich et al. 2018) Framework � 2 + � � � � � � � q i − ˆ β ijr f r ( q i , q j ) − α i q i β ijr f r ( q i , q j ) + λ ρ ( f r ) ( i , j , r ) ∈ E − r i ∈ V ( i , j , r ) ∈ E Neural T ensor Network (akin to Socher et al. 2013) f r ( q i , q j ) = u r ⊤ tanh ( q ⊤ i A r q j ) where A r ∈ R d × d × k and ρ ( f r ) = � A r � 2 + � u r � 2 13 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Functional relations (Lengerich et al. 2018) Framework � 2 + � � � � � � � q i − ˆ β ijr f r ( q i , q j ) − α i q i β ijr f r ( q i , q j ) + λ ρ ( f r ) ( i , j , r ) ∈ E − r i ∈ V ( i , j , r ) ∈ E Your favorite graph embedding method Bordes et al. 2013; Wang et al. 2014; Lin et al. 2015; for overviews, see Nickel et al. 2011; Hamilton et al. 2017. 13 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier FrameNet evaluation ‘Inheritance’ ‘Using’ ‘Reframing’ ‘Subframe’ ‘Perspective On’ Model (2132/992) (1552/668) (544/312) (356/168) (336/148) None 87 . 58 88 . 59 85 . 60 91 . 24 89 . 59 Faruqui et al. 90 . 79 87 . 87 87 . 02 94 . 50 94 . 24 FR-Linear 92 . 04 89 . 37 94 . 65 92 . 92 94 . 73 FR-Neural 92 . 46 94 . 04 92 . 54 89 . 57 95 . 65 ‘Precedes’ ‘See Also’ ‘Causative Of’ ‘Inchoative Of’ Model (220/136) (268/76) (204/36) (60/16) None 87 . 30 85 . 11 86 . 11 82 . 50 Faruqui et al. 85 . 26 83 . 81 84 . 49 78 . 33 FR-Linear 87 . 00 91 . 93 92 . 09 82 . 50 FR-Neural 89 . 16 93 . 25 94 . 33 85 . 00 14 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier A drug–disease knowledge graph Faruqui et al. FR-Linear ‘Treats’ Model (9152/2490) None 72 . 02 ± 0 . 50 Faruqui et al. 72 . 93 ± 0 . 82 FR-Linear 84 . 22 ± 0 . 82 FR-Neural 73 . 52 ± 0 . 89 15 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Functional complexity in the lexicon Category Semantic type noun properties intransitive verbs properties transitive verbs entities to properties adjectives properties to properties prepositions entities to (properties to properties) determiner properties to sets of properties Vector space models tend to be monotyped, but see Clark et al. 2011. 16 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Some lexical generalizations 17 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Some lexical generalizations 1. Some transitive verbs entail the existence of their direct object ( see ) and some do not ( seek ). 17 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Some lexical generalizations 1. Some transitive verbs entail the existence of their direct object ( see ) and some do not ( seek ). 2. Across languages, verbs lexicalize manner or result, but not both (Rappaport Hovav & Levin 2010): ◮ Manner : nibble , scribble , sweep , flutter ◮ Result : clean , cover , empty , fill 17 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Some lexical generalizations 1. Some transitive verbs entail the existence of their direct object ( see ) and some do not ( seek ). 2. Across languages, verbs lexicalize manner or result, but not both (Rappaport Hovav & Levin 2010): ◮ Manner : nibble , scribble , sweep , flutter ◮ Result : clean , cover , empty , fill 3. Some adjectives predicate distributively across their arguments, others do not (Glass 2018): ◮ Box A and Box B are new. (entails both are new) ◮ Box A and Box B are heavy. (does not entail both are heavy) 17 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Some lexical generalizations 1. Some transitive verbs entail the existence of their direct object ( see ) and some do not ( seek ). 2. Across languages, verbs lexicalize manner or result, but not both (Rappaport Hovav & Levin 2010): ◮ Manner : nibble , scribble , sweep , flutter ◮ Result : clean , cover , empty , fill 3. Some adjectives predicate distributively across their arguments, others do not (Glass 2018): ◮ Box A and Box B are new. (entails both are new) ◮ Box A and Box B are heavy. (does not entail both are heavy) Can we develop deep learning systems that derive such generalizations? No training against them; that’s just restating them! 17 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Compositional semantics 18 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier A semanticist’s ideal Every student attended a lecture ∀ z (( student z ) → ( ∃ x ( lecture x ) ∧ ( attended x z ))) λ g ∀ z (( student z ) → ( g z )) λ y ( ∃ x ( lecture x ) ∧ ( attended x y )) λ f λ g ∀ z (( f z ) → ( g z )) student attended λ g ( ∃ x ( lecture x ) ∧ ( g x )) λ f λ g ( ∃ x ( f x ) ∧ ( g x )) lecture 19 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier A semanticist’s ideal Every student attended a lecture ∀ z (( student z ) → ( ∃ x ( lecture x ) ∧ ( attended x z ))) λ g ∀ z (( student z ) → ( g z )) λ y ( ∃ x ( lecture x ) ∧ ( attended x y )) λ f λ g ∀ z (( f z ) → ( g z )) student attended λ g ( ∃ x ( lecture x ) ∧ ( g x )) λ f λ g ( ∃ x ( f x ) ∧ ( g x )) lecture But is this really so ideal? 19 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Complete semantic representations? MacCartney & Manning (2009) The difficulty is plain: truly natural language is fiendishly complex. [. . . ] Consider for a moment the difficulty of fully and accurately translating 1. Every firm polled saw costs grow more than expected, even after adjusting for inflation. to a formal meaning representation. 20 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Sparse, fragmented feature representations S NP VP The NYT reported S NP VP the deal fell through source_NYT the 2 T embedded_implicit_neg NYT 1 T deal_neg report 1 1 length 7 vocab 6 . . . . . . 21 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier The answer from deep learning � � S 1 = f NP 1 , VP 1 � � � � NP 1 = f The , NYT VP 1 = f reported , S 2 NYT The The NYT � � reported S 2 = f NP 2 , VP 2 reported � � � � NP 2 = f the , deal VP 2 = f fell , through the deal fell through the deal fell through 22 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier The answer from deep learning � � h 1 = f h 0 , The � � The h 2 = f h 1 , NYT The � � NYT h 3 = f h 2 , reported NYT � � reported h 4 = f h 3 , the reported � � the h 5 = f h 4 , deal the � � deal h 6 = f h 5 , fell deal � � fell h 7 = f h 6 , through fell through through 23 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier The answer from deep learning h 0 h 1 h 2 h 3 h 4 h 5 h 6 h 7 The NYT reported the deal fell through The NYT reported the deal fell through 24 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier The answer from deep learning All our parses are wrong, but perhaps we can discover the right one(s). h 0 h 1 h 2 h 3 h 4 h 5 h 6 h 7 The NYT reported the deal fell through The NYT reported the deal fell through 25 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier A new perspective on compositionality 26 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier A new perspective on compositionality Compositionality The meaning of a complex phrase is a function of the meaning of its consituent phrases. 26 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier A new perspective on compositionality Compositionality The meaning of a complex phrase is a function of the meaning of its consituent phrases. Partee (1984): Context-dependence, Ambiguity, and Challenges to Local, Deterministic Compositionality 26 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier A new perspective on compositionality Compositionality The meaning of a complex phrase is a function of the meaning of its consituent phrases. Partee (1984): Context-dependence, Ambiguity, and Challenges to Local, Deterministic Compositionality NP 1 ; VP 1 W + b � � = tanh S 1 � � reported ; S 2 W + b = tanh NP 1 VP 1 reported S 2 26 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Compositional generalizations: monotonicity (Bowman 2017) 27 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Compositional generalizations: monotonicity Kim smoked. Kim didn’t smoke. ↑ ↓ Kim smoked cigars. Kim didn’t smoke cigars. (Bowman 2017) 27 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Compositional generalizations: monotonicity Kim smoked. Kim didn’t smoke. ↑ ↓ Kim smoked cigars. Kim didn’t smoke cigars. A student smoked. ր տ A Swedish student smoked. A student smoked cigars. (Bowman 2017) 27 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Compositional generalizations: monotonicity Kim smoked. Kim didn’t smoke. ↑ ↓ Kim smoked cigars. Kim didn’t smoke cigars. A student smoked. ր տ A Swedish student smoked. A student smoked cigars. No student smoked. ւ ց No Swedish student smoked. No student smoked cigars. (Bowman 2017) 27 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Compositional generalizations: monotonicity Kim smoked. Kim didn’t smoke. ↑ ↓ Kim smoked cigars. Kim didn’t smoke cigars. A student smoked. ր տ A Swedish student smoked. A student smoked cigars. No student smoked. ւ ց No Swedish student smoked. No student smoked cigars. Every student smoked. ւ տ Every Swedish student smoked. Every student smoked cigars. (Bowman 2017) 27 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Compositional generalizations: monotonicity Kim smoked. Kim didn’t smoke. ↑ ↓ Kim smoked cigars. Kim didn’t smoke cigars. A student smoked. ր տ A Swedish student smoked. A student smoked cigars. No student smoked. ւ ց No Swedish student smoked. No student smoked cigars. Every student smoked. ւ տ Every Swedish student smoked. Every student smoked cigars. Most students smoked. − տ Most Swedish students smoked. Most students smoked cigars. (Bowman 2017) 27 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Pragmatics 28 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 1. I am speaking. 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 1. I am speaking. 2. We won. [A team I’m on; a team I support; . . . ] 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 1. I am speaking. 2. We won. [A team I’m on; a team I support; . . . ] 3. I am here. [NAACL; New Orleans; planet earth; . . . ] 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 1. I am speaking. 2. We won. [A team I’m on; a team I support; . . . ] 3. I am here. [NAACL; New Orleans; planet earth; . . . ] 4. We are here. [pointing at a map] 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 1. I am speaking. 2. We won. [A team I’m on; a team I support; . . . ] 3. I am here. [NAACL; New Orleans; planet earth; . . . ] 4. We are here. [pointing at a map] 5. I’m not here now. [answering machine] 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 1. I am speaking. 2. We won. [A team I’m on; a team I support; . . . ] 3. I am here. [NAACL; New Orleans; planet earth; . . . ] 4. We are here. [pointing at a map] 5. I’m not here now. [answering machine] 6. We went to a local bar after the workshop. 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 1. I am speaking. 2. We won. [A team I’m on; a team I support; . . . ] 3. I am here. [NAACL; New Orleans; planet earth; . . . ] 4. We are here. [pointing at a map] 5. I’m not here now. [answering machine] 6. We went to a local bar after the workshop. 7. three days ago, tomorrow, now, . . . 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? a. Connecticut. 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? a. Connecticut. b. Stanford. 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? a. Connecticut. b. Stanford. c. The U.S. 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? a. Connecticut. b. Stanford. c. The U.S. d. Planet earth. 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? a. Connecticut. b. Stanford. c. The U.S. d. Planet earth. 9. If Kangaroos had no tails, they would fall over. 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? a. Connecticut. b. Stanford. c. The U.S. d. Planet earth. 9. If Kangaroos had no tails, they would fall over. a. True, 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? a. Connecticut. b. Stanford. c. The U.S. d. Planet earth. 9. If Kangaroos had no tails, they would fall over. a. True, b. as long as we don’t slip in the premise that they have jet packs. 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? a. Connecticut. b. Stanford. c. The U.S. d. Planet earth. 9. If Kangaroos had no tails, they would fall over. a. True, b. as long as we don’t slip in the premise that they have jet packs. 10. I didn’t see any. 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? a. Connecticut. b. Stanford. c. The U.S. d. Planet earth. 9. If Kangaroos had no tails, they would fall over. a. True, b. as long as we don’t slip in the premise that they have jet packs. 10. I didn’t see any. a. Are there typos in my slides? 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? a. Connecticut. b. Stanford. c. The U.S. d. Planet earth. 9. If Kangaroos had no tails, they would fall over. a. True, b. as long as we don’t slip in the premise that they have jet packs. 10. I didn’t see any. a. Are there typos in my slides? b. Are the cookies in cupboard. 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 8. Where are you from? a. Connecticut. b. Stanford. c. The U.S. d. Planet earth. 9. If Kangaroos had no tails, they would fall over. a. True, b. as long as we don’t slip in the premise that they have jet packs. 10. I didn’t see any. a. Are there typos in my slides? b. Are the cookies in cupboard. c. Are there bookstores downtown? 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 11. I think this is the way to the library. [politeness] 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 11. I think this is the way to the library. [politeness] 12. Would you mind if I stole your pen for a second? [politeness] 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 11. I think this is the way to the library. [politeness] 12. Would you mind if I stole your pen for a second? [politeness] 13. He’s not exactly a genius/idiot. [irony] 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 11. I think this is the way to the library. [politeness] 12. Would you mind if I stole your pen for a second? [politeness] 13. He’s not exactly a genius/idiot. [irony] 14. Great idea! [sarcasm(?)] 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 11. I think this is the way to the library. [politeness] 12. Would you mind if I stole your pen for a second? [politeness] 13. He’s not exactly a genius/idiot. [irony] 14. Great idea! [sarcasm(?)] 15. Any chance we can sort this out here, officer? [bribery(?)] 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Natural language is situated and social 11. I think this is the way to the library. [politeness] 12. Would you mind if I stole your pen for a second? [politeness] 13. He’s not exactly a genius/idiot. [irony] 14. Great idea! [sarcasm(?)] 15. Any chance we can sort this out here, officer? [bribery(?)] 16. It’d be a shame if something happened to your dog. [threat(?)] 29 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier SHRDLU (Winograd 1972) Fully grounded system that parses the user’s input, maps it to logical form, interprets that logical form in its world, tries to take appropriate action, and generates responses. One project did succeed. Terry Winograd’s program SHRDLU could use English intelligently, but there was a catch: the only subject you could discuss was a micro- world of simulated blocks. 30 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier SHRDLU (Winograd 1972) 1. Person: WHAT DOES THE BOX CONTAIN? 2. Computer: THE BLUE PYRAMID AND THE BLUE BLOCK. 3. Person: WHAT IS THE PYRAMID SUPPORTED BY? 4. Computer: THE BOX. 5. Person: HOW MANY BLOCKS ARE NOT IN THE BOX? 6. Computer: FOUR OF THEM. 7. Person: IS AT LEAST ONE OF THEM NARROWER THAN THE ONE WHICH I TOLD YOU TO PICK UP? 8. Computer: YES, THE RED CUBE. 9. Person: IS IT SUPPORTED? 10. Computer: YES, BY THE TABLE. 30 / 42
Signs of the apocalypse? Lexical semantics Compositional semantics Pragmatics The next frontier Example: sarcasm sarcastic? w h 0 h 1 h 2 h 3 Yeah great idea Yeah great idea 31 / 42
Recommend
More recommend
