towards interactive and automatic refinement of translation rules
DESCRIPTION
Towards Interactive and Automatic Refinement of Translation Rules. Ariadna Font Llitjós PhD Thesis Proposal Jaime Carbonell (advisor) Alon Lavie (co-advisor) Lori Levin Bonnie Dorr (Univ. Maryland) 5 November 2004. Outline. Introduction Thesis statement and scope Preliminary Research - PowerPoint PPT PresentationTRANSCRIPT
Towards Interactive and Automatic Refinement of
Translation Rules
Ariadna Font Llitjós PhD Thesis Proposal
Jaime Carbonell (advisor)
Alon Lavie (co-advisor)
Lori Levin
Bonnie Dorr (Univ. Maryland)
5 November 2004
Interactive and Automatic Rule Refinement 2
Outline
• Introduction
• Thesis statement and scope
• Preliminary Research– Interactive elicitation of error information– A framework for automatic rule adaptation
• Proposed Research
• Contributions and Thesis Timeline
Interactive and Automatic Rule Refinement 3
Machine Translation (MT)
• Source Language (SL) sentence: Gaudi was a great artist
In Spanish, it translates as: Gaudi era un gran artista
• MT System outputs :*Gaudi estaba un artista grande
*Gaudi era un artista grande
Interactive and Automatic Rule Refinement 4
Spanish Adjectives Completed Work Automatic Rule
Adaptation
NP
DET N ADJ
NP
DET ADJ N
a big house una casa grande
NP
DET ADJ N
NP
DET ADJ N
a great artist un gran artista
General order: grande big in size
Exception: gran exceptional
Interactive and Automatic Rule Refinement 5
Commercial and Online Systems
Correct Translation: Gaudi era un gran artista
• Systran, Babelfish (Altavista), WorldLingo, Translated.net : *Gaudi era gran artista
• ImTranslation: *El Gaudi era un gran artista
• 1-800-Translate *Gaudi era un fenomenal artista
Interactive and Automatic Rule Refinement 6
• Current solutions:
Manual post-editing [Allen, 2003]
Automated post-edition module (APE) [Allen & Hogan, 2000]
Post-editing
Interactive and Automatic Rule Refinement 7
Drawbacks of Current Methods
• Manual post-editing Corrections do not generalize Gaudi era un artista grande
Juan es un amigo grande (Juan is a great friend)
Era una oportunidad grande (It is a great
opportunity)
• APE Humans need to predict all the errors ahead of time and code for the post-editing rules + new error
Interactive and Automatic Rule Refinement 8
My Solution
• Automate post-editing efforts by feeding them back into the MT system.
• Possible alternatives:Automatic learning of post-editing rules
+ system independent
- several thousands of sentences might need to be corrected for the same error
Automatic refinement of translation rules+ attacks the core of the problem- for transfer-based MT systems (need rules to fix!)
Interactive and Automatic Rule Refinement 9
[Corston-Oliver & Gammon, 2003]
[Imamura et al. 2003]
[Menezes & Richardson, 2001]
Related Work
[Callison-Burch, 2004]
[Su et al. 1995]
[Brill, 1993]
[Gavaldà, 2000]
Post-editing
Rule Adaptation
Machine TranslationMy Thesis
No pre-existing training data required
No human reference translations required
Use Non-expert user feedback
Interactive and Automatic Rule Refinement 10
Resource-poor Scenarios (AVENUE)
- Lack of electronic parallel data - Lack of manual grammar (or very small initial grammar)
Need to validate elicitation corpus and automatically learned translation rules
Why bother? - Indigenous communities have difficult access to
crucial information that directly affects their life (such as land laws, plagues, health warnings, etc.)
- Preservation of their language and culture
MapudungunQuechuaAymara
Interactive and Automatic Rule Refinement 11
How is MT possible for resource-poor languages?
Bilingual speakers
Interactive and Automatic Rule Refinement 12
AVENUE Project Overview
Learning
Module
Transfer Rules
Lexical Resources
Run Time Transfer System
Lattice
Word-Aligned Parallel Corpus
Elicitation Tool
Elicitation Corpus
Elicitation Rule Learning
Run-Time System
Handcrafted rules
Morphology
Morpho-logical analyzer
Interactive and Automatic Rule Refinement 13
My Thesis
Learning
Module
Transfer Rules
Lexical Resources
Run Time Transfer System
Lattice
Translation
Correction
Tool
Word-Aligned Parallel Corpus
Elicitation Tool
Elicitation Corpus
Elicitation Rule Learning
Run-Time System
Rule Refinement
Rule
Refinement
Module
Handcrafted rules
Morphology
Morpho-logical analyzer
Recycle corrections of Machine Translation output back into the system
by refining and expanding existing translation rules
Interactive and Automatic Rule Refinement 15
Thesis Statement
- Given a rule-based Transfer MT system, we can extract useful information from non-expert bilingual speakers about the corrections required to make MT output acceptable..
- We can automatically refine and expand translation rules, given corrected and aligned translation pairs and some error information, to improve coverage and overall MT quality.
Interactive and Automatic Rule Refinement 16
Assumptions
• No parallel training data available
• No human reference translations available
• The SL sentence needs to be fully parsed by the translation grammar.
• Bilingual speakers can give enough information about the MT errors.
Interactive and Automatic Rule Refinement 17
Scope
Types of errors that:
• Focus 1: can be refined fully automatically just by using correction information.
• Focus 2: can be refined fully automatically using correction and error information.
• Focus 3: require a reasonable amount of further user interaction and can be solved by available correction and error information.
Interactive and Automatic Rule Refinement 18
Technical Challenges
Elicit minimal MT information from non-expert users
Refine and expand a translation rules
minimally
Manually written Automatically Learned
Automatic Evaluation of Refinement process
Preliminary Work
• Interactive elicitation of error information
• A framework for automatic rule adaptation
Interactive and Automatic Rule Refinement 20
Interactive Elicitation of MT Errors
Goal: • Simplify MT correction task maximally
Challenges:• Find appropriate level of granularity for MT error
classification
• Design a user-friendly graphic user interface with: • SL sentence (e.g. I see them)• TL sentence (e.g. Yo veo los)• word-to-word alignments (I-yo, see-veo, them-los)• (context)
Interactive and Automatic Rule Refinement 21
MT Error Typology for RR (simplified)
Missing word
Extra word
Wrong word order
Incorrect word
Wrong agreement
Completed Work Interactive elicitation of error information
Local vs Long distance
Word vs. phrase
+ Word change
Sense
Form
Selectional restrictions
Idiom
Missing constraint
Extra constraint
Interactive and Automatic Rule Refinement 22
TCTool (Demo)• Add a word• Delete a word• Modify a word• Change word order
Actions:
Interactive elicitation of error information
Interactive and Automatic Rule Refinement 23
1st Eng2Spa User Study [LREC 2004]
• MT error classification 9 linguistically-motivated classes:
word order, sense, agreement error (number, person, gender, tense), form, incorrect word and no translation
Interactive elicitation of error information
Completed Work
precision recall F1
error detection 90% 89% 89%
error classification 72% 71% 72%
Interactive and Automatic Rule Refinement 24
Automatic Rule Refinement Framework
• Find best RR operations given a:• Grammar (G), • Lexicon (L), • (Set of) Source Language sentence(s) (SL), • (Set of) Target Language sentence(s) (TL), • Its Parse tree (P), and • Minimal correction of TL (TL’)
such that TQ2 > TQ1• Which can also be expressed as:
max TQ(TL|TL’,P,SL,RR(G,L))
Completed Work Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 25
1. Refine a translation rule:R0 R1 (R0 modified, either made more
specific or more general)
Types of Refinement OperationsCompleted Work Automatic Rule
Adaptation
R0: NP
DET N ADJ
NP
DET ADJ N
R1:
a nice house
una casa bonito
NP
DET N ADJ
NP
DET ADJ N
a nice house
una casa bonita
N gender = ADJ gender
Interactive and Automatic Rule Refinement 26
2. Bifurcate a translation rule:R0 R0 (same, general rule)
R1 (R0 modified, specific rule)
Types of Refinement Operations (2)Completed Work Automatic Rule
Adaptation
R0: NP
DET N ADJ
NP
DET ADJ N
NP
DET ADJ N
NP
DET ADJ N
R1:
a nice house una casa bonita
a great artist un gran artista
Interactive and Automatic Rule Refinement 27
Formalizing Error Information
Wi = error
Wi’ = correction
Wc = clue word
• Example:
SL: the red car - TL: *el auto roja TL’: el auto rojo
Wi = roja Wi’ = rojo Wc = auto
need to agree
Completed Work Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 28
Triggering Feature Detection
Comparison at the feature level to detect triggering feature(s)
Delta function: (Wi,Wi’)
Examples:(rojo,roja) = {gender}
(comiamos,comia) = {person,number}
(mujer,guitarra) = {}
If set is empty, need to
postulate a new binary feature
Completed Work Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 29
Deciding on the Refinement Op
Given:
- Action performed by the user (add, delete,
modify, change word order) , and
- Error information is available (clue word, word alignments, etc.)
Refinement Action
Completed Work Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 30
Rule Refinement Operations
Modify Add Delete Change W Order
+Wc –Wc +Wc –Wc +al –al Wi Wc Wi(…) Wc –Wc
= +rule –rule +al –al =Wi WiWi’ RuleLearner
POSi=POSi’ POSiPOSi’
Proposed Work
- Batch and Interactive mode
- User studies
- Evaluation
Interactive and Automatic Rule Refinement 32
Rule Refinement ExampleAutomatic Rule Adaptation
Change word order
SL: Gaudí was a great artist
TL: Gaudí era un artista grande
Corrected TL (TL’): Gaudí era un gran artista
Interactive and Automatic Rule Refinement 33Refinement Operation Typology
1. Error Information Elicitation
Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 34
2. Variable Instantiation from Log File
Correcting Actions:
1. Word order change (artista grande grande artista):
Wi = grande
2. Edited grande into gran:
Wi’ = gran
identified artist as clue word Wc = artist
In this case, even if user had not identified Wc,
refinement process would have been the same
Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 35
3. Retrieve Relevant Lexical Entries
• No lexical entry for great gran
• Duplicate lexical entry great-grande and change TL side:
ADJ::ADJ |: [great] -> [gran]
((X1::Y1)
((x0 form) = great)
((y0 agr num) = sg)
((y0 agr gen) = masc))
(Morphological analyzer: grande = gran)
Automatic Rule Adaptation
ADJ::ADJ |: [great] -> [grande]
((X1::Y1)
((x0 form) = great)
((y0 agr num) = sg)
((y0 agr gen) = masc))
Interactive and Automatic Rule Refinement 36
4. Finding Triggering Feature(s)
Feature function: (Wi, Wi’) = need to postulate a new binary feature: feat1
5. Blame assignment
tree: <((S,1 (NP,2 (N,5:1 "GAUDI") )
(VP,3 (VB,2 (AUX,17:2 "ERA") )
(NP,8 (DET,0:3 "UN")
(N,4:5 "ARTISTA")
(ADJ,5:4 "GRANDE") ) ) ) )>
Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 37
6. Variable Instantiation in the Rules
Wi = grande POSi = ADJ = Y3, y3
Wc = artista POSc = N = Y2, y2
{NP,8} NP::NP : [DET ADJ N] -> [DET N ADJ]( (X1::Y1) (X2::Y3) (X3::Y2) ((x0 def) = (x1 def)) (x0 = x3) ((y1 agr) = (y2 agr)) ; det-noun agreement ((y3 agr) = (y2 agr)) ; adj-noun agreement (y2 = x3) )
Automatic Rule Adaptation
(R0)
Interactive and Automatic Rule Refinement 38
7. Refining Rules
{NP,8’}
NP::NP : [DET ADJ N] -> [DET ADJ N]
( (X1::Y1) (X2::Y2) (X3::Y3)
((x0 def) = (x1 def))
(x0 = x3)
((y1 agr) = (y3 agr)) ; det-noun agreement
((y2 agr) = (y3 agr)) ; adj-noun agreement
(y2 = x3)
((y2 feat1) =c + ))
Automatic Rule Adaptation
(R1)
Interactive and Automatic Rule Refinement 39
8. Refining Lexical Entries
ADJ::ADJ |: [great] -> [grande]((X1::Y1)((x0 form) = great)((y0 agr num) = sg)((y0 agr gen) = masc)((y0 feat1) = -))
ADJ::ADJ |: [great] -> [gran]((X1::Y1)((x0 form) = great)((y0 agr num) = sg)((y0 agr gen) = masc)((y0 feat1) = +))
Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 40
Done? Not yet
NP,8 (R0) ADJ(grande) [feat1 = -]
NP,8’ (R1) ADJ(gran)
[feat1 =c +] [feat1 = +]
Need to restrict application of general rule (R0) to just post-nominal ADJ
Automatic Rule Adaptation
un artista grande
un artista gran
un gran artista
*un grande artista
Interactive and Automatic Rule Refinement 41
Add Blocking Constraint
NP,8 (R0) ADJ(grande) [feat1 = -] [feat1 = -]
NP,8’ (R1) ADJ(gran)
[feat1 =c +] [feat1 = +]
Can we also eliminate incorrect translations automatically?
Automatic Rule Adaptation
un artista grande
*un artista gran
un gran artista
*un grande artista
Interactive and Automatic Rule Refinement 42
Making the grammar tighter
• If Wc = artista Add [feat1= +] to N(artista) Add agreement constraint to NP,8 (R0)
between N and ADJ ((N feat1) = (ADJ feat1))
Automatic Rule Adaptation
*un artista grande
*un artista gran
un gran artista
*un grande artista
Interactive and Automatic Rule Refinement 43
Batch Mode Implementation
• For Refinement operations of errors that can be refined :
– Fully automatically just by using correction information (Focus 1)
– Fully automatically using correction and error information (Focus 2)
Proposed Work Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 44
Rule Refinement Operations
Modify Add Delete Change W Order
+Wc –Wc +Wc –Wc +al –al Wi Wc Wi(…) Wc –Wc
= +rule –rule +al –al =Wi WiWi’ RuleLearner
POSi=POSi’ POSiPOSi’
Interactive and Automatic Rule Refinement 45
Modify Add Delete Change W Order
+Wc –Wc +Wc –Wc +al –al Wi Wc Wi(…) Wc –Wc
= +rule –rule +al –al =Wi WiWi’ RuleLearner
POSi=POSi’ POSiPOSi’
Rule Refinement Operations
Focus 1
It is a nice house – Es una casa bonito Es una casa bonita
John and Mary fell – Juan y Maria cayeron Juan y Maria se cayeron
Gaudi was a great artist – Gaudi era un artista grande Gaudi era un gran artista
Interactive and Automatic Rule Refinement 46
Modify Add Delete Change W Order
+Wc –Wc +Wc –Wc +al –al Wi Wc Wi(…) Wc –Wc
= +rule –rule +al –al =Wi WiWi’ RuleLearner
POSi=POSi’ POSiPOSi’
Rule Refinement Operations
Focus 1
Gaudi was a great artist – Gaudi era un artista grande Gaudi era un gran artista
Es una casa bonito Es una casa bonita
J y M cayeron J y M se cayeron
I will help him fix the car – Ayudaré a él a arreglar el auto
Le ayudare a arreglar el auto
Interactive and Automatic Rule Refinement 47
Modify Add Delete Change W Order
+Wc –Wc +Wc –Wc +al –al Wi Wc Wi(…) Wc –Wc
= +rule –rule +al –al =Wi WiWi’ RuleLearner
POSi=POSi’ POSiPOSi’
Rule Refinement Operations
Focus 1
I will help him fix the car – Ayudaré a él a arreglar el auto
Le ayudare a arreglar el auto
I would like to go – Me gustaria que ir
Me gustaria ir
Interactive and Automatic Rule Refinement 48
Modify Add Delete Change W Order
+Wc –Wc +Wc –Wc +al –al Wi Wc Wi(…) Wc –Wc
= +rule –rule +al –al =Wi WiWi’ RuleLearner
POSi=POSi’ POSiPOSi’
Rule Refinement Operations
Focus 1 & 2
I am proud of you – Estoy orgullosa tu Estoy orgullosa de ti
PP PREP NP
Interactive and Automatic Rule Refinement 49
Interactive Mode Implementation
• Extra error information is required More sentences need to be evaluated
(and corrected) by users Relevant Minimal Pairs (MP)
– Focus 3: types of errors that require a reasonable amount of further user interaction and can be solved by available correction and error information.
Proposed Work Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 50
Modify Add Delete Change W Order
+Wc –Wc +Wc –Wc +al –al Wi Wc Wi(…) Wc –Wc
= +rule –rule +al –al =Wi WiWi’ RuleLearner
POSi=POSi’ POSiPOSi’
Rule Refinement Operations
Focus 3
Wally plays the guitar – Wally juega la guitarra Wally toca la guitarra
I saw the woman – Vi la mujer Vi a la mujer
I see them – Veo los Los veo
Interactive and Automatic Rule Refinement 51
Example Requiring Minimal PairAutomatic Rule Adaptation
1. Run SL sentence through the transfer engine
I see them *veo los los veo
2. Wi = los but no Wi’ nor Wc
Need a minimal pair to determine appropriate refinement:
I see cars veo autos
3. Triggering feature(s): (los,autos) = {pos}
PRON(los)[pos=pron] N(autos)[pos=n]
Proposed Work
Interactive and Automatic Rule Refinement 52
Refining and Adding Constraints
VP,3: VP NP VP NP
VP,3’: VP NP NP VP + [NP pos =c pron]
• Percolate triggering features up to the constituent level:
NP: PRON PRON + [NP pos = PRON pos]
• Block application of general rule (VP,3):
VP,3: VP NP VP NP + [NP pos = (*NOT* pron)]
Proposed Work
Interactive and Automatic Rule Refinement 53
Generalization Power
• Have other example sentences with same error that would be translated correctly after refinement!
Interactive and Automatic Rule Refinement 54
Rule Refinement Operations
Modify Add Delete Change W Order
+Wc –Wc +Wc –Wc +al –al Wi Wc Wi(…) Wc –Wc
= +rule –rule +al –al =Wi WiWi’ RuleLearner
POSi=POSi’ POSiPOSi’
Outside Scope of Thesis
John read the book – A Juan leyó el libro Juan leyó el libro
Where are you from? – Donde eres tu de? De donde eres tu?
Interactive and Automatic Rule Refinement 55
User Studies
• TCTool: new MT classification (Eng2Spa)
• Different language pair (Mapudungun or Quechua Spanish)
• Manual vs Learned grammars
• Batch vs Interactive mode (+Active Learning)
• Amount of error information elicited
Proposed Work
Interactive and Automatic Rule Refinement 58
Evaluation of Refined MT System
• Evaluate best translation Automatic evaluation metrics (BLEU, NIST, METEOR)
• Evaluate candidate list precision (+parsimony)
Interactive and Automatic Rule Refinement 59
Evaluate Best translation
Hypothesis file (translations to be evaluated automatically)
Raw MT output:• Best sentence (picked by user to be correct or
requiring the least amount of correction) Refined MT output:• Use METOR score at sentence level to pick best
candidate from the list
Run all automatic metrics on the new hypothesis file using user corrections as reference translations.
Interactive and Automatic Rule Refinement 60
Evaluate Candidate List
• Precision: “tp” binary {0,1}
tp + fp total number of TC
SLTLTLTL
SLTLTLTL
SLTLTLTLTLTL
SLTLTLTLTLTLTL
=
Interactive and Automatic Rule Refinement 61
Expected Contributions
• An efficient online GUI to display translations and alignments and solicit pinpoint fixes from non-expert bilingual users.
• An expandable set of rule refinement operators– triggered by user corrections,– to automatically refine and expand different types
of grammars.
• A mechanism to automatically evaluate rule refinements with user corrections as reference translations.
Interactive and Automatic Rule Refinement 62
Thesis Timeline
Research components Duration (months)
Back-end implementation 7
User Studies 3
Resource-poor language (data + manual grammar) 2
Adapt system to new language pair 1
Active Learning methods 1
Evaluation 1
Write and defend thesis 3
Total 18
Expected graduation date: May 2006
Interactive and Automatic Rule Refinement 63
References
Add references:
• Related work
• Probst et al. 2002
• AL
Interactive and Automatic Rule Refinement 64
Thanks!
Questions?
Interactive and Automatic Rule Refinement 65
Some Questions
• Is the refinement process deterministic?
Interactive and Automatic Rule Refinement 66
Others
• TCTool Demo Simulation
• RR operation patterns
• Automatic Evaluation feasibility study
• AMTA paper results
• BLEU, NIST and METEOR
• Precision, recall and F1
Interactive and Automatic Rule Refinement 67
Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 68
Automatic Rule AdaptationSL + best TL picked by user
Interactive and Automatic Rule Refinement 70
Automatic Rule AdaptationChanging “grande” into “gran”
Interactive and Automatic Rule Refinement 71
Automatic Rule Adaptation
Back to main
Interactive and Automatic Rule Refinement 72
1
2
3
Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 73
Input to RR module
sl: I see them
tl: VEO LOS
tree: <((S,0 (VP,3 (VP,1 (V,1:2 "VEO") )
(NP,0 (PRON,2:3 "LOS") ) ) ) )>
- User correction log file - Transfer engine output (+ parse tree):
Automatic Rule Adaptation
sl: I see cars
tl: VEO AUTOS
tree: <((S,0 (VP,3 (VP,1 (V,1:2 "VEO") )
(NP,2 (N,1:3 “AUTOS") ) ) ) )>
Interactive and Automatic Rule Refinement 74
Types of RR Operations
• Grammar:– R0 R0 + R1 [=R0’ + constr] Cov[R0] Cov[R0,R1]– R0 R1[=R0 + constr= -]
R2[=R0’ + constr=c +] Cov[R0] Cov[R1,R2]
– R0 R1 [=R0 + constr] Cov[R0] Cov[R1]
• Lexicon– Lex0 Lex0 + Lex1[=Lex0 + constr] – Lex0 Lex1[=Lex0 + constr]– Lex0 Lex0 + Lex1[Lex0 + TLword] Lex1 (adding lexical item)
bifurcate
refine
Completed Work Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 75
Manual vs Learned Grammars
[AMTA 2004]
Automatic Rule Adaptation
NIST BLEU METEOR
Manual grammar 4.3 0.16 0.6
Learned grammar 3.7 0.14 0.55
• Manual inspection:
• Automatic MT Evaluation:
Interactive and Automatic Rule Refinement 76
Human Oracle experiment
• As a feasibility experiment, compared raw output with manually corrected MT:
statistically significant (confidence interval test)
• These is an upper-bound on how much difference we should expect any refinement approach to make.
Automatic Rule Adaptation
Completed Work
Interactive and Automatic Rule Refinement 77
Active Learning
• Minimize the number of examples a human annotator must label [Cohn et al. 1994] usually by processing examples in order of usefulness.
.
Minimize the number of Minimal Pairs presented to users
Proposed Work Automatic Rule Adaptation
Interactive and Automatic Rule Refinement 78
Order deterministic?
• Application of Rule Refinement operations is not deterministic, it directly depends on:– The order in which it sees the corrected
sentences
• Example:• 1st agr constraint + bifurcate (WWO)
– C-set
• Reverse order– C-set (!=)