THE SOURCE OF NONFINITE TEMPORAL INTERPRETATION ELLISE MOON AND - - PowerPoint PPT Presentation

▶
the source of nonfinite temporal interpretation
SMART_READER_LITE
LIVE PREVIEW

THE SOURCE OF NONFINITE TEMPORAL INTERPRETATION ELLISE MOON AND - - PowerPoint PPT Presentation

Dec 19, 2023 180 likes •1.75k views

THE SOURCE OF NONFINITE TEMPORAL INTERPRETATION ELLISE MOON AND AARON STEVEN WHITE, UNIVERSITY OF ROCHESTER CENTRAL QUESTION Which aspects of semantic interpretation are due to predicates' denotations and which are due to the denotations of their

slide-1
SLIDE 1

THE SOURCE OF NONFINITE TEMPORAL INTERPRETATION

ELLISE MOON AND AARON STEVEN WHITE, UNIVERSITY OF ROCHESTER

slide-2
SLIDE 2

CENTRAL QUESTION

Which aspects of semantic interpretation are due to predicates' denotations and which are due to the denotations of their arguments?

2

slide-3
SLIDE 3

CENTRAL QUESTION

Which aspects of semantic interpretation are due to predicates' denotations and which are due to the denotations of their arguments? Focus: temporal interpretation in English nonfinite embedded clauses.

(Stowell, 1982; Landau, 2001; Wurmbrand, 2001, 2014; Grano, 2012, 2017; Pearson, 2016) 2

slide-4
SLIDE 4

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 1. Jo wanted to leave.

3

slide-5
SLIDE 5

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 1. Jo wanted to leave.

3

leaving

past future

slide-6
SLIDE 6

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 1. Jo wanted to leave.

3

leaving wanting

past future

slide-7
SLIDE 7

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 1. Jo wanted to leave.

3

leaving wanting

past future

  • 2. Jo regretted leaving.
slide-8
SLIDE 8

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 1. Jo wanted to leave.

3

leaving wanting regretting

past future

  • 2. Jo regretted leaving.
slide-9
SLIDE 9

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 1. Jo wanted to leave.

3

leaving wanting regretting

past future

  • 2. Jo regretted leaving.
slide-10
SLIDE 10

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 3. Jo remembered leaving.

4

slide-11
SLIDE 11

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 3. Jo remembered leaving.

4

leaving

past future

slide-12
SLIDE 12

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 3. Jo remembered leaving.

4

leaving remembering

past future

slide-13
SLIDE 13

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 3. Jo remembered leaving.

4

leaving remembering

past future

  • 4. Jo remembered to leave.
slide-14
SLIDE 14

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 3. Jo remembered leaving.

4

leaving remembering

past future

  • 4. Jo remembered to leave.
slide-15
SLIDE 15

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 3. Jo remembered leaving.

4

leaving remembering remembering

past future

  • 4. Jo remembered to leave.
slide-16
SLIDE 16

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 3. Jo remembered leaving.

4

leaving remembering remembering

past future

  • 4. Jo remembered to leave.
slide-17
SLIDE 17

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 4. Jo remembered to leave.

5

slide-18
SLIDE 18

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 4. Jo remembered to leave.

5

leaving

past future

slide-19
SLIDE 19

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 4. Jo remembered to leave.

5

leaving

past future

slide-20
SLIDE 20

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 4. Jo remembered to leave.

5

leaving remembering

past future

slide-21
SLIDE 21

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 4. Jo remembered to leave.

5

leaving remembering

past future

  • 5. Jo claimed to leave.
slide-22
SLIDE 22

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 4. Jo remembered to leave.

5

leaving remembering

past future

  • 5. Jo claimed to leave.
slide-23
SLIDE 23

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 4. Jo remembered to leave.

5

leaving remembering

past future

  • 5. Jo claimed to leave.
slide-24
SLIDE 24

CLAUSE EMBEDDING AND TEMPORAL ORIENTATION

  • 4. Jo remembered to leave.

5

leaving remembering claiming

past future

  • 5. Jo claimed to leave.
slide-25
SLIDE 25

QUESTION

What is the source of this temporal orientation?

6

slide-26
SLIDE 26

CHALLENGE

Are predicates like remember and claim just idiosyncratic?

7

slide-27
SLIDE 27

CHALLENGE

Are predicates like remember and claim just idiosyncratic?

7

Bird’s-eye view of temporal orientation across the lexicon

REQUIRES

slide-28
SLIDE 28

APPROACH

▪

Collect a lexicon-scale dataset of clause-embedding verbs with different possible embedded structures

8

slide-29
SLIDE 29

APPROACH

▪

Collect a lexicon-scale dataset of clause-embedding verbs with different possible embedded structures

▪

Formalize possible theoretical frameworks as parameters in a computational model and test on data

8

slide-30
SLIDE 30

TALK OUTLINE ▪ Introduction

9

slide-31
SLIDE 31

TALK OUTLINE ▪ Introduction ▪ Three Hypotheses

9

slide-32
SLIDE 32

TALK OUTLINE ▪ Introduction ▪ Three Hypotheses ▪ Data Collection

9

slide-33
SLIDE 33

TALK OUTLINE ▪ Introduction ▪ Three Hypotheses ▪ Data Collection ▪ Model Design

9

slide-34
SLIDE 34

TALK OUTLINE ▪ Introduction ▪ Three Hypotheses ▪ Data Collection ▪ Model Design ▪ Analysis and Results

9

slide-35
SLIDE 35

HYPOTHESES

  • 1. Lexical: T

emporal orientation is due to the predicate

10 (Pearson, 2016)

slide-36
SLIDE 36

Jo regretted leaving regret ⤳ t(regret) < t(leave)

HYPOTHESES

  • 1. Lexical: T

emporal orientation is due to the predicate

10 (Pearson, 2016)

slide-37
SLIDE 37
  • 2. Structural: T

emporal orientation is due to the structure of the argument selected by the predicate

HYPOTHESES

11 (Stowell, 1982; Landau, 2001; Wurmbrand, 2001, 2014; Grano, 2012)

slide-38
SLIDE 38

Jo regretted leaving ⤳ t(regret) < t(leave)

  • 2. Structural: T

emporal orientation is due to the structure of the argument selected by the predicate

HYPOTHESES

11

VP leave -ing

(Stowell, 1982; Landau, 2001; Wurmbrand, 2001, 2014; Grano, 2012)

slide-39
SLIDE 39

HYPOTHESES

12

  • 3. Mixed: temporal orientation depends on both the

predicate and argument type.

slide-40
SLIDE 40

HYPOTHESES

Jo remembered leaving. remember ⤳ t(remember) < t(leave) ⤳ t(leave) < t(remember)

12

VP leave -ing

  • 3. Mixed: temporal orientation depends on both the

predicate and argument type.

slide-41
SLIDE 41

TALK OUTLINE ▪ Introduction ▪ Three Hypotheses

13

slide-42
SLIDE 42

TALK OUTLINE ▪ Introduction ▪ Three Hypotheses ▪ Data Collection

13

slide-43
SLIDE 43

GOAL

A way to capture temporal orientation across different possible verb/structure pairings

14

slide-44
SLIDE 44

GOAL

A way to capture temporal orientation across different possible verb/structure pairings

14

REQUIRES

slide-45
SLIDE 45

GOAL

A way to capture temporal orientation across different possible verb/structure pairings A bleaching method for acceptability judgements, following White and Rawlins 2016

14

REQUIRES

slide-46
SLIDE 46

DATA COLLECTION

Jo wanted to leave in the future. *Jo will want to leave in the past.

15

slide-47
SLIDE 47

DATA COLLECTION

Jo wanted to leave in the future. *Jo will want to leave in the past.

15

temporal adverb phrase

slide-48
SLIDE 48

DATA COLLECTION

Jo wanted to leave in the future. *Jo will want to leave in the past.

15

tense manipulation temporal adverb phrase

slide-49
SLIDE 49

DATA COLLECTION

Jo wanted to leave in the future. *Jo will want to leave in the past.

15

future-oriented tense manipulation temporal adverb phrase past-oriented

slide-50
SLIDE 50

DATA COLLECTION

NP __ doing something Someone regretted doing something.

16

slide-51
SLIDE 51

DATA COLLECTION

NP __ to do something Someone wanted to do something.

17

slide-52
SLIDE 52

DATA COLLECTION

NP __ to have something Someone loved to have something.

18

slide-53
SLIDE 53

DATA COLLECTION

NP was __ to do something Someone was told to do something.

(Pesetsky 1991, Moulton 2009) 19

slide-54
SLIDE 54

DATA COLLECTION

NP was __ to have something Someone was believed to have something.

20

slide-55
SLIDE 55

21

slide-56
SLIDE 56

21

slide-57
SLIDE 57

DATA COLLECTION

▪

2208 verb/complement pairs in 2 orientations

22

slide-58
SLIDE 58

DATA COLLECTION

▪

2208 verb/complement pairs in 2 orientations

▪

Semantically bleached 3rd person singular subject

22

slide-59
SLIDE 59

DATA COLLECTION

▪

2208 verb/complement pairs in 2 orientations

▪

Semantically bleached 3rd person singular subject

▪

Lists of 48 sentences, with even distribution of

  • rientations and randomized item order

22

slide-60
SLIDE 60

DATA COLLECTION

▪

2208 verb/complement pairs in 2 orientations

▪

Semantically bleached 3rd person singular subject

▪

Lists of 48 sentences, with even distribution of

  • rientations and randomized item order

▪

10 acceptability judgements per sentence from 869 annotators on Mechanical Turk

22

slide-61
SLIDE 61

DATA COLLECTION

23

slide-62
SLIDE 62

DATA COLLECTION

23

verb

slide-63
SLIDE 63

DATA COLLECTION

23

verb complement

slide-64
SLIDE 64

DATA COLLECTION

23

verb complement

slide-65
SLIDE 65

DATA COLLECTION

23

verb complement future-oriented

slide-66
SLIDE 66

DATA COLLECTION

23

verb complement future-oriented

slide-67
SLIDE 67

DATA COLLECTION

24

slide-68
SLIDE 68

DATA COLLECTION

24

verb

slide-69
SLIDE 69

DATA COLLECTION

24

verb complement

slide-70
SLIDE 70

DATA COLLECTION

24

verb complement

slide-71
SLIDE 71

DATA COLLECTION

24

verb complement past-oriented

slide-72
SLIDE 72

DATA COLLECTION

24

verb complement past-oriented

slide-73
SLIDE 73

25

slide-74
SLIDE 74

25

I. predicates which permit both

  • rientations
slide-75
SLIDE 75

25

II. future-oriented predicates I. predicates which permit both

  • rientations
slide-76
SLIDE 76

25

II. future-oriented predicates III. simultaneous predicates I. predicates which permit both

  • rientations

(Wurmbrand 2014)

slide-77
SLIDE 77

25

II. future-oriented predicates III. simultaneous predicates I. predicates which permit both

  • rientations

IV. past-oriented predicates

(Wurmbrand 2014)

slide-78
SLIDE 78

26

slide-79
SLIDE 79

26

slide-80
SLIDE 80

26

slide-81
SLIDE 81

26

slide-82
SLIDE 82

26

slide-83
SLIDE 83

26

slide-84
SLIDE 84

26

slide-85
SLIDE 85

TALK OUTLINE ▪ Introduction ▪ Three Hypotheses ▪ Data Collection

27

slide-86
SLIDE 86

TALK OUTLINE ▪ Introduction ▪ Three Hypotheses ▪ Data Collection ▪ Model Design

27

slide-87
SLIDE 87

GOAL

✓ A way to capture temporal orientation across different

possible verb/structure pairings

28

slide-88
SLIDE 88

GOAL

✓ A way to capture temporal orientation across different

possible verb/structure pairings

▪

A way to model our hypotheses relative to this data

(White and Rawlins 2016) 28

slide-89
SLIDE 89

29

Predicate

slide-90
SLIDE 90

29

Predicate Semantic Type

select

slide-91
SLIDE 91

29

Predicate Syntactic Structures Semantic Type

select project

slide-92
SLIDE 92

29

Predicate Syntactic Structures Semantic Type

select project (White & Rawlins 2016)

slide-93
SLIDE 93

29

Predicate Type Predicate Syntactic Structures Semantic Type

select project has (White & Rawlins 2016) (An & White 2020)

slide-94
SLIDE 94

29

Predicate Type Predicate Syntactic Structures Semantic Type Temporal Orientation

select project has (White & Rawlins 2016) (An & White 2020)

slide-95
SLIDE 95

29

Predicate Type Predicate Syntactic Structures Semantic Type Temporal Orientation

select project has (White & Rawlins 2016) (An & White 2020)

slide-96
SLIDE 96

29

Predicate Type Predicate Syntactic Structures Semantic Type Temporal Orientation

select project has (White & Rawlins 2016) (An & White 2020)

slide-97
SLIDE 97

29

Predicate Type Predicate Syntactic Structures Semantic Type Temporal Orientation

select project has (White & Rawlins 2016) (An & White 2020)

slide-98
SLIDE 98

30

slide-99
SLIDE 99

30

slide-100
SLIDE 100

30

slide-101
SLIDE 101

30

Lexical Only

slide-102
SLIDE 102

30

Lexical Only Structural Only

slide-103
SLIDE 103

30

Lexical Only Structural Only

slide-104
SLIDE 104

31

slide-105
SLIDE 105

31

slide-106
SLIDE 106

31

slide-107
SLIDE 107

31

slide-108
SLIDE 108

31

slide-109
SLIDE 109

32

Verb Complement Future Acc. Past Acc.

abhor NP Ved VPing

  • 0.503955

0.413169 abhor NP was Ved to VP[+eventive] 0.134924

  • 1.559801

absolve NP Ved to VP[+eventive] 0.948428

  • 2.079783

accept NP Ved VPing 4.774069 1.883071 accept NP Ved to VP[-eventive] 2.434219

  • 1.854628

accept NP was Ved to VP[+eventive] 2.946932

  • 2.002958

acclaim NP Ved VPing

  • 2.137957

0.221483 acclaim NP Ved to VP[+eventive]

  • 2.549958
  • 0.554269

acclaim NP was Ved to VP[-eventive] 1.382240

  • 0.742686

add NP Ved VPing 3.664288

  • 3.777042

add NP Ved to VP[+eventive] 0.503324

  • 0.172519

add NP was Ved to VP[+eventive] 1.878762

  • 2.685818

address NP Ved VPing 1.876711 3.596447 address NP was Ved to VP[+eventive] 0.928784

  • 1.928204

admire NP Ved VPing

  • 0.070897
  • 0.475992

admit NP Ved VPing

  • 0.690028

4.566390 admit NP Ved to VP[+eventive]

  • 3.257618

0.955866 admit NP Ved to VP[-eventive] 0.373650

  • 2.930481

admit NP was Ved to VP[+eventive]

  • 1.103509

1.371476 admit NP was Ved to VP[-eventive] 0.318550 1.463886

slide-110
SLIDE 110

32

Verb Complement Future Acc. Past Acc.

abhor NP Ved VPing

  • 0.503955

0.413169 abhor NP was Ved to VP[+eventive] 0.134924

  • 1.559801

absolve NP Ved to VP[+eventive] 0.948428

  • 2.079783

accept NP Ved VPing 4.774069 1.883071 accept NP Ved to VP[-eventive] 2.434219

  • 1.854628

accept NP was Ved to VP[+eventive] 2.946932

  • 2.002958

acclaim NP Ved VPing

  • 2.137957

0.221483 acclaim NP Ved to VP[+eventive]

  • 2.549958
  • 0.554269

acclaim NP was Ved to VP[-eventive] 1.382240

  • 0.742686

add NP Ved VPing 3.664288

  • 3.777042

add NP Ved to VP[+eventive] 0.503324

  • 0.172519

add NP was Ved to VP[+eventive] 1.878762

  • 2.685818

address NP Ved VPing 1.876711 3.596447 address NP was Ved to VP[+eventive] 0.928784

  • 1.928204

admire NP Ved VPing

  • 0.070897
  • 0.475992

admit NP Ved VPing

  • 0.690028

4.566390 admit NP Ved to VP[+eventive]

  • 3.257618

0.955866 admit NP Ved to VP[-eventive] 0.373650

  • 2.930481

admit NP was Ved to VP[+eventive]

  • 1.103509

1.371476 admit NP was Ved to VP[-eventive] 0.318550 1.463886

slide-111
SLIDE 111

32

Verb Complement Future Acc. Past Acc.

abhor NP Ved VPing

  • 0.503955

0.413169 abhor NP was Ved to VP[+eventive] 0.134924

  • 1.559801

absolve NP Ved to VP[+eventive] 0.948428

  • 2.079783

accept NP Ved VPing 4.774069 1.883071 accept NP Ved to VP[-eventive] 2.434219

  • 1.854628

accept NP was Ved to VP[+eventive] 2.946932

  • 2.002958

acclaim NP Ved VPing

  • 2.137957

0.221483 acclaim NP Ved to VP[+eventive]

  • 2.549958
  • 0.554269

acclaim NP was Ved to VP[-eventive] 1.382240

  • 0.742686

add NP Ved VPing 3.664288

  • 3.777042

add NP Ved to VP[+eventive] 0.503324

  • 0.172519

add NP was Ved to VP[+eventive] 1.878762

  • 2.685818

address NP Ved VPing 1.876711 3.596447 address NP was Ved to VP[+eventive] 0.928784

  • 1.928204

admire NP Ved VPing

  • 0.070897
  • 0.475992

admit NP Ved VPing

  • 0.690028

4.566390 admit NP Ved to VP[+eventive]

  • 3.257618

0.955866 admit NP Ved to VP[-eventive] 0.373650

  • 2.930481

admit NP was Ved to VP[+eventive]

  • 1.103509

1.371476 admit NP was Ved to VP[-eventive] 0.318550 1.463886

Train model Train model Train model Train model Train model Test on held-out data

slide-112
SLIDE 112

32

Verb Complement Future Acc. Past Acc.

abhor NP Ved VPing

  • 0.503955

0.413169 abhor NP was Ved to VP[+eventive] 0.134924

  • 1.559801

absolve NP Ved to VP[+eventive] 0.948428

  • 2.079783

accept NP Ved VPing 4.774069 1.883071 accept NP Ved to VP[-eventive] 2.434219

  • 1.854628

accept NP was Ved to VP[+eventive] 2.946932

  • 2.002958

acclaim NP Ved VPing

  • 2.137957

0.221483 acclaim NP Ved to VP[+eventive]

  • 2.549958
  • 0.554269

acclaim NP was Ved to VP[-eventive] 1.382240

  • 0.742686

add NP Ved VPing 3.664288

  • 3.777042

add NP Ved to VP[+eventive] 0.503324

  • 0.172519

add NP was Ved to VP[+eventive] 1.878762

  • 2.685818

address NP Ved VPing 1.876711 3.596447 address NP was Ved to VP[+eventive] 0.928784

  • 1.928204

admire NP Ved VPing

  • 0.070897
  • 0.475992

admit NP Ved VPing

  • 0.690028

4.566390 admit NP Ved to VP[+eventive]

  • 3.257618

0.955866 admit NP Ved to VP[-eventive] 0.373650

  • 2.930481

admit NP was Ved to VP[+eventive]

  • 1.103509

1.371476 admit NP was Ved to VP[-eventive] 0.318550 1.463886

Train model Train model Train model Train model Train model Test on held-out data

slide-113
SLIDE 113

32

Verb Complement Future Acc. Past Acc.

abhor NP Ved VPing

  • 0.503955

0.413169 abhor NP was Ved to VP[+eventive] 0.134924

  • 1.559801

absolve NP Ved to VP[+eventive] 0.948428

  • 2.079783

accept NP Ved VPing 4.774069 1.883071 accept NP Ved to VP[-eventive] 2.434219

  • 1.854628

accept NP was Ved to VP[+eventive] 2.946932

  • 2.002958

acclaim NP Ved VPing

  • 2.137957

0.221483 acclaim NP Ved to VP[+eventive]

  • 2.549958
  • 0.554269

acclaim NP was Ved to VP[-eventive] 1.382240

  • 0.742686

add NP Ved VPing 3.664288

  • 3.777042

add NP Ved to VP[+eventive] 0.503324

  • 0.172519

add NP was Ved to VP[+eventive] 1.878762

  • 2.685818

address NP Ved VPing 1.876711 3.596447 address NP was Ved to VP[+eventive] 0.928784

  • 1.928204

admire NP Ved VPing

  • 0.070897
  • 0.475992

admit NP Ved VPing

  • 0.690028

4.566390 admit NP Ved to VP[+eventive]

  • 3.257618

0.955866 admit NP Ved to VP[-eventive] 0.373650

  • 2.930481

admit NP was Ved to VP[+eventive]

  • 1.103509

1.371476 admit NP was Ved to VP[-eventive] 0.318550 1.463886

Train model Train model Train model Train model Train model Test on held-out data

slide-114
SLIDE 114

32

Verb Complement Future Acc. Past Acc.

abhor NP Ved VPing

  • 0.503955

0.413169 abhor NP was Ved to VP[+eventive] 0.134924

  • 1.559801

absolve NP Ved to VP[+eventive] 0.948428

  • 2.079783

accept NP Ved VPing 4.774069 1.883071 accept NP Ved to VP[-eventive] 2.434219

  • 1.854628

accept NP was Ved to VP[+eventive] 2.946932

  • 2.002958

acclaim NP Ved VPing

  • 2.137957

0.221483 acclaim NP Ved to VP[+eventive]

  • 2.549958
  • 0.554269

acclaim NP was Ved to VP[-eventive] 1.382240

  • 0.742686

add NP Ved VPing 3.664288

  • 3.777042

add NP Ved to VP[+eventive] 0.503324

  • 0.172519

add NP was Ved to VP[+eventive] 1.878762

  • 2.685818

address NP Ved VPing 1.876711 3.596447 address NP was Ved to VP[+eventive] 0.928784

  • 1.928204

admire NP Ved VPing

  • 0.070897
  • 0.475992

admit NP Ved VPing

  • 0.690028

4.566390 admit NP Ved to VP[+eventive]

  • 3.257618

0.955866 admit NP Ved to VP[-eventive] 0.373650

  • 2.930481

admit NP was Ved to VP[+eventive]

  • 1.103509

1.371476 admit NP was Ved to VP[-eventive] 0.318550 1.463886

Train model Train model Train model Train model Train model Test on held-out data

slide-115
SLIDE 115

32

Verb Complement Future Acc. Past Acc.

abhor NP Ved VPing

  • 0.503955

0.413169 abhor NP was Ved to VP[+eventive] 0.134924

  • 1.559801

absolve NP Ved to VP[+eventive] 0.948428

  • 2.079783

accept NP Ved VPing 4.774069 1.883071 accept NP Ved to VP[-eventive] 2.434219

  • 1.854628

accept NP was Ved to VP[+eventive] 2.946932

  • 2.002958

acclaim NP Ved VPing

  • 2.137957

0.221483 acclaim NP Ved to VP[+eventive]

  • 2.549958
  • 0.554269

acclaim NP was Ved to VP[-eventive] 1.382240

  • 0.742686

add NP Ved VPing 3.664288

  • 3.777042

add NP Ved to VP[+eventive] 0.503324

  • 0.172519

add NP was Ved to VP[+eventive] 1.878762

  • 2.685818

address NP Ved VPing 1.876711 3.596447 address NP was Ved to VP[+eventive] 0.928784

  • 1.928204

admire NP Ved VPing

  • 0.070897
  • 0.475992

admit NP Ved VPing

  • 0.690028

4.566390 admit NP Ved to VP[+eventive]

  • 3.257618

0.955866 admit NP Ved to VP[-eventive] 0.373650

  • 2.930481

admit NP was Ved to VP[+eventive]

  • 1.103509

1.371476 admit NP was Ved to VP[-eventive] 0.318550 1.463886

Train model Train model Train model Train model Train model Test on held-out data

slide-116
SLIDE 116

TALK OUTLINE ▪ Introduction ▪ Three Hypotheses ▪ Data Collection ▪ Model Design

33

slide-117
SLIDE 117

TALK OUTLINE ▪ Introduction ▪ Three Hypotheses ▪ Data Collection ▪ Model Design ▪ Analysis and Results

33

slide-118
SLIDE 118

34

slide-119
SLIDE 119

34

slide-120
SLIDE 120

34

slide-121
SLIDE 121

Lexical Only

34

slide-122
SLIDE 122

Lexical Only

34

Structural Only

slide-123
SLIDE 123

Lexical Only

34

Structural Only

slide-124
SLIDE 124

Lexical Only

34

Structural Only

slide-125
SLIDE 125

34

slide-126
SLIDE 126

34

slide-127
SLIDE 127

35

slide-128
SLIDE 128

35

slide-129
SLIDE 129

36

slide-130
SLIDE 130

36

slide-131
SLIDE 131

36

slide-132
SLIDE 132

36

slide-133
SLIDE 133

37

slide-134
SLIDE 134

37

slide-135
SLIDE 135

38

slide-136
SLIDE 136

38

slide-137
SLIDE 137

38

slide-138
SLIDE 138

38

slide-139
SLIDE 139

38

slide-140
SLIDE 140

39

slide-141
SLIDE 141

39

slide-142
SLIDE 142

39

slide-143
SLIDE 143

39

slide-144
SLIDE 144

39

slide-145
SLIDE 145

39

slide-146
SLIDE 146

CONCLUSION

Both constructional and lexical models do fit the data, but in different ways, mixed models less so.

40

slide-147
SLIDE 147

CONCLUSION

Both constructional and lexical models do fit the data, but in different ways, mixed models less so. These models capture fine-grained information about verbal semantics in areas related to temporality.

40

slide-148
SLIDE 148

CONCLUSION

Both constructional and lexical models do fit the data, but in different ways, mixed models less so. These models capture fine-grained information about verbal semantics in areas related to temporality. Lexicon-scale datasets of verb features like this can enable us to empirically test theoretical possibilities.

40

slide-149
SLIDE 149

Thank you! Data is available at megaattitude.io

41

slide-150
SLIDE 150

REFERENCES

Abusch, Dorit. 1997. Sequence of T ense and T emporal De Re. Linguistics and Philosophy 20:1-50. An, Hannah Youngeun, and Aaron Steven White. 2020. The lexical and grammatical sources of neg-raising inferences. In Proceedings of the Society for Computation in Linguistics, to appear. Grano, Thomas. 2012. Control and Restructuring at the Syntax-Semantics Interface. Doctoral Dissertation, University of Chicago. Grano, Thomas. 2017. The logic of intention reports. Journal of Semantics 34:587-632. Kratzer, Angelika. 1998. More structural analogies between pronouns and tenses. Proceedings from Semantics and Linguistic Theory, 8:92-110. Moulton, Kier. 2009. Natural Selection and the Syntax of Clausal Complementation. Doctoral Dissertation, University of Massachusetts Amherst. Landau, Idan. 2001. Elements of Control: Structure and meaning in infinitival constructions. Dordrecht: Springer Science & Business Media. Ogihara, T

  • shiyuki. 1995. The Semantics of T

ense in Embedded Clauses. Linguistic Inquiry 26:663-679. Partee, Barbara H. 1973. Some structural analogies between tenses and pronouns in English. Journal of Philosophy 70:601-609. Pearson, Hazel. 2016. The semantics of partial control. Natural Language & Linguistic Theory 34:691–738. Pesetsky, David. 1991. Zero syntax: vol. 2: Infinitives. Stowell, Tim. 1982. The tense of infinitives. Linguistic Inquiry 13:561–570. White, Aaron Steven, and Kyle Rawlins. 2016. A computational model of S-selection. Semantics and Linguistic Theory 26:641–663. White, Aaron Steven, and Kyle Rawlins. 2018. The role of veridicality and factivity in clause selection. In Proceedings of the 48th Annual Meeting of the North East Linguistic Society, to appear. Amherst, MA: GLSA Publications. Wurmbrand, Susi. 2001. Infinitives: Restructuring and clause structure. Berlin: Mouton de Gruyter. Wurmbrand, Susi. 2014. T ense and aspect in English infinitives. Linguistic Inquiry 45:403–447.

42

slide-151
SLIDE 151

APPENDICES

43

slide-152
SLIDE 152

44

ANNOTATOR INSTRUCTIONS

slide-153
SLIDE 153

45

slide-154
SLIDE 154

46

slide-155
SLIDE 155

47

slide-156
SLIDE 156

48