1. Alon Lavie, Jaime Carbonell, Lori Levin,
AVENUE/LETRAS: Learning-based MT Approaches for Languages with Limited Resources
Alon Lavie, Jaime Carbonell, Lori Levin,
Bob Frederking
Joint work with:
Erik Peterson, Christian Monson, Ariadna Font-Llitjos, Alison Alvarez, Roberto Aranovich

2. Why Machine Translation for Languages with Limited Resources?
We are in the age of information explosion
The internet+web+Google  anyone can get the information they want anytime…
But what about the text in all those other languages?
How do they read all this English stuff?
How do we read all the stuff that they put online?
MT for these languages would Enable:
Better government access to native indigenous and minority communities
Better minority and native community participation in information-rich activities (health care, education, government) without giving up their languages.
Civilian and military applications (disaster relief)
Language preservation
Sep 22, 2006
Learning-based MT with Limited Resources

3. AVENUE/LETRAS Funding
Started in 2000 with small amount of DARPA/TIDES funding (NICE)
AVENUE project funded by 5-year NSF ITR grant ( )
Follow-on LETRAS project funded by NSF HLC Program grant ( )
Collaboration funding sources:
Mapudungun (MINEDUC, Chile)
Hebrew (ISF, Israel)
Brazilian Portuguese & Native Langs. (Brazilian Gov.)
Inupiaq (NSF, Polar Programs)
Sep 22, 2006
Learning-based MT with Limited Resources

4. Learning-based MT with Limited Resources
CMU’s AVENUE Approach
Elicitation: use bilingual native informants to create a small high-quality word-aligned bilingual corpus of translated phrases and sentences
Building Elicitation corpora from feature structures
Feature Detection and Navigation
Transfer-rule Learning: apply ML-based methods to automatically acquire syntactic transfer rules for translation between the two languages
Learn from major language to minor language
Translate from minor language to major language
XFER + Decoder:
XFER engine produces a lattice of possible transferred structures at all levels
Decoder searches and selects the best scoring combination
Rule Refinement: refine the acquired rules via a process of interaction with bilingual informants
Morphology Learning
Word and Phrase bilingual lexicon acquisition
Sep 22, 2006
Learning-based MT with Limited Resources

5. Learning-based MT with Limited Resources
AVENUE MT Approach
Interlingua
Semantic Analysis
Sentence Planning
Syntactic Parsing
Transfer Rules
Text Generation
AVENUE: Automate Rule Learning
Source
(e.g. Quechua)
Target
(e.g. English)
Direct: SMT, EBMT
Sep 22, 2006
Learning-based MT with Limited Resources

6. AVENUE Architecture Sep 22, 2006
Elicitation
Morphology
Rule Learning
Run-Time System
Rule Refinement
Learning
Module
Learned Transfer Rules
Word-Aligned Parallel Corpus
INPUT TEXT
Translation
Correction
Tool
Run Time Transfer System
Rule
Refinement
Module
Elicitation Corpus
Morphology
Analyzer
Learning Module
Handcrafted
rules
Decoder
Elicitation Tool
Lexical Resources
OUTPUT TEXT
Sep 22, 2006
Learning-based MT with Limited Resources

7. Transfer Rule Formalism
;SL: the old man, TL: ha-ish ha-zaqen
NP::NP [DET ADJ N] -> [DET N DET ADJ] (
(X1::Y1)
(X1::Y3)
(X2::Y4)
(X3::Y2)
((X1 AGR) = *3-SING)
((X1 DEF = *DEF)
((X3 AGR) = *3-SING)
((X3 COUNT) = +)
((Y1 DEF) = *DEF)
((Y3 DEF) = *DEF)
((Y2 AGR) = *3-SING)
((Y2 GENDER) = (Y4 GENDER)) )
Type information
Part-of-speech/constituent information
Alignments
x-side constraints
y-side constraints
xy-constraints,
e.g. ((Y1 AGR) = (X1 AGR))
Sep 22, 2006
Learning-based MT with Limited Resources

8. Transfer Rule Formalism (II)
;SL: the old man, TL: ha-ish ha-zaqen
NP::NP [DET ADJ N] -> [DET N DET ADJ] (
(X1::Y1)
(X1::Y3)
(X2::Y4)
(X3::Y2)
((X1 AGR) = *3-SING)
((X1 DEF = *DEF)
((X3 AGR) = *3-SING)
((X3 COUNT) = +)
((Y1 DEF) = *DEF)
((Y3 DEF) = *DEF)
((Y2 AGR) = *3-SING)
((Y2 GENDER) = (Y4 GENDER)) )
Value constraints
Agreement constraints
Sep 22, 2006
Learning-based MT with Limited Resources

9. Transfer Rules  Transfer TreesNP
PP NP1
NP P Adj N
N1 ke eka aXyAya N
jIvana NP
NP PP
Adj N P NP
one chapter of N1 N
life
; NP1 ke NP2 -> NP2 of NP1
; Ex: jIvana ke eka aXyAya
; life of (one) chapter
; ==> a chapter of life ;
{NP,12}
NP::NP : [PP NP1] -> [NP1 PP] (
(X1::Y2)
(X2::Y1)
; ((x2 lexwx) = 'kA') )
{NP,13}
NP::NP : [NP1] -> [NP1]
(X1::Y1)
{PP,12}
PP::PP : [NP Postp] -> [Prep NP]
Sep 22, 2006
Learning-based MT with Limited Resources

10. Rule Learning - Overview
Goal: Acquire Syntactic Transfer Rules
Use available knowledge from the source side (grammatical structure)
Three steps:
Flat Seed Generation: first guesses at transfer rules; flat syntactic structure
Compositionality Learning: use previously learned rules to learn hierarchical structure
Constraint Learning: refine rules by learning appropriate feature constraints
Sep 22, 2006
Learning-based MT with Limited Resources

11. Flat Seed Rule Generation
Learning Example: NP
Eng: the big apple
Heb: ha-tapuax ha-gadol
Generated Seed Rule:
NP::NP [ART ADJ N]  [ART N ART ADJ]
((X1::Y1)
(X1::Y3)
(X2::Y4)
(X3::Y2))
Sep 22, 2006
Learning-based MT with Limited Resources

12. Compositionality Learning
Initial Flat Rules:
S::S [ART ADJ N V ART N]  [ART N ART ADJ V P ART N]
((X1::Y1) (X1::Y3) (X2::Y4) (X3::Y2) (X4::Y5) (X5::Y7) (X6::Y8))
NP::NP [ART ADJ N]  [ART N ART ADJ]
((X1::Y1) (X1::Y3) (X2::Y4) (X3::Y2))
NP::NP [ART N]  [ART N]
((X1::Y1) (X2::Y2))
Generated Compositional Rule:
S::S [NP V NP]  [NP V P NP]
((X1::Y1) (X2::Y2) (X3::Y4))
Sep 22, 2006
Learning-based MT with Limited Resources

13. Learning-based MT with Limited Resources
Constraint Learning
Input: Rules and their Example Sets
S::S [NP V NP]  [NP V P NP] {ex1,ex12,ex17,ex26}
((X1::Y1) (X2::Y2) (X3::Y4))
NP::NP [ART ADJ N]  [ART N ART ADJ] {ex2,ex3,ex13}
((X1::Y1) (X1::Y3) (X2::Y4) (X3::Y2))
NP::NP [ART N]  [ART N] {ex4,ex5,ex6,ex8,ex10,ex11}
((X1::Y1) (X2::Y2))
Output: Rules with Feature Constraints:
S::S [NP V NP]  [NP V P NP]
((X1::Y1) (X2::Y2) (X3::Y4)
(X1 NUM = X2 NUM)
(Y1 NUM = Y2 NUM)
(X1 NUM = Y1 NUM))
Sep 22, 2006
Learning-based MT with Limited Resources

14. Learning-based MT with Limited Resources
AVENUE Prototypes
General XFER framework under development for past three years
Prototype systems so far:
German-to-English, Dutch-to-English
Chinese-to-English
Hindi-to-English
Hebrew-to-English
Portuguese-to-English
In progress or planned:
Mapudungun-to-Spanish
Quechua-to-Spanish
Inupiaq-to-English
Native-Brazilian languages to Brazilian Portuguese
Sep 22, 2006
Learning-based MT with Limited Resources

15. Learning-based MT with Limited Resources
Mapudungun
Indigenous Language of Chile and Argentina
~ 1 Million Mapuche Speakers
Sep 22, 2006
Learning-based MT with Limited Resources

16. Learning-based MT with Limited Resources
Collaboration
Eliseo Cañulef
Rosendo Huisca
Hugo Carrasco
Hector Painequeo
Flor Caniupil
Luis Caniupil Huaiquiñir
Marcela Collio Calfunao
Cristian Carrillan Anton
Salvador Cañulef
Mapuche Language Experts
Universidad de la Frontera (UFRO)
Instituto de Estudios Indígenas (IEI)
Institute for Indigenous Studies
Chilean Funding
Chilean Ministry of Education (Mineduc)
Bilingual and Multicultural Education Program
Carolina Huenchullan Arrúe
Claudio Millacura Salas
Sep 22, 2006
Learning-based MT with Limited Resources

17. Learning-based MT with Limited Resources
Accomplishments
Corpora Collection
Spoken Corpus
Collected: Luis Caniupil Huaiquiñir
Medical Domain
3 of 4 Mapudungun Dialects
120 hours of Nguluche
30 hours of Lafkenche
20 hours of Pwenche
Transcribed in Mapudungun
Translated into Spanish
Written Corpus
~ 200,000 words
Bilingual Mapudungun – Spanish
Historical and newspaper text
nmlch-nmjm1_x_0405_nmjm_00:
M: <SPA>no pütokovilu kay ko
C: no, si me lo tomaba con agua
M: chumgechi pütokoki femuechi pütokon pu <Noise>
C: como se debe tomar, me lo tomé pués
nmlch-nmjm1_x_0406_nmlch_00:
M: Chengewerkelafuymiürke
C: Ya no estabas como gente entonces!
Sep 22, 2006
Learning-based MT with Limited Resources

18. Learning-based MT with Limited Resources
Accomplishments
Developed At UFRO
Bilingual Dictionary with Examples
1,926 entries
Spelling Corrected Mapudungun Word List
117,003 fully-inflected word forms
Segmented Word List
15,120 forms
Stems translated into Spanish
Sep 22, 2006
Learning-based MT with Limited Resources

19. Learning-based MT with Limited Resources
Accomplishments
Developed at LTI using Mapudungun language resources from UFRO
Spelling Checker
Integrated into OpenOffice
Hand-built Morphological Analyzer
Prototype Machine Translation Systems
Rule-Based
Example-Based
Website: LenguasAmerindias.org
Sep 22, 2006
Learning-based MT with Limited Resources

20. Challenges for Hebrew MT
Paucity in existing language resources for Hebrew
No publicly available broad coverage morphological analyzer
No publicly available bilingual lexicons or dictionaries
No POS-tagged corpus or parse tree-bank corpus for Hebrew
No large Hebrew/English parallel corpus
Scenario well suited for CMU transfer-based MT framework for languages with limited resources
Sep 22, 2006
Learning-based MT with Limited Resources

21. Hebrew-to-English MT Prototype
Initial prototype developed within a two month intensive effort
Accomplished:
Adapted available morphological analyzer
Constructed a preliminary translation lexicon
Translated and aligned Elicitation Corpus
Learned XFER rules
Developed (small) manual XFER grammar as a point of comparison
System debugging and development
Evaluated performance on unseen test data using automatic evaluation metrics
Sep 22, 2006
Learning-based MT with Limited Resources

22. Transfer Rules Transfer Engine Decoder Source Input
בשורה הבאה
Transfer Rules
{NP1,3}
NP1::NP1 [NP1 "H" ADJ] -> [ADJ NP1]
((X3::Y1)
(X1::Y2)
((X1 def) = +)
((X1 status) =c absolute)
((X1 num) = (X3 num))
((X1 gen) = (X3 gen))
(X0 = X1))
Preprocessing
Morphology
English Language Model
Transfer Engine
Translation Lexicon
N::N |: ["$WR"] -> ["BULL"]
((X1::Y1)
((X0 NUM) = s)
((Y0 lex) = "BULL"))
N::N |: ["$WRH"] -> ["LINE"]
((Y0 lex) = "LINE"))
Decoder
Translation Output Lattice
(0 1
(1 1
(2 2
(1 2 "THE
(0 2 "IN
(0 4 "IN THE NEXT
English Output
in the next line

23. Learning-based MT with Limited Resources
Morphology Example
Input word: B$WRH
| B$WRH |
|-----B-----|$WR|--H--|
|--B--|-H--|--$WRH---|
Sep 22, 2006
Learning-based MT with Limited Resources

24. Learning-based MT with Limited Resources
Morphology Example
Y0: ((SPANSTART 0) Y1: ((SPANSTART 0) Y2: ((SPANSTART 1)
(SPANEND 4) (SPANEND 2) (SPANEND 3)
(LEX B$WRH) (LEX B) (LEX $WR)
(POS N) (POS PREP)) (POS N)
(GEN F) (GEN M)
(NUM S) (NUM S)
(STATUS ABSOLUTE)) (STATUS ABSOLUTE))
Y3: ((SPANSTART 3) Y4: ((SPANSTART 0) Y5: ((SPANSTART 1)
(SPANEND 4) (SPANEND 1) (SPANEND 2)
(LEX $LH) (LEX B) (LEX H)
(POS POSS)) (POS PREP)) (POS DET))
Y6: ((SPANSTART 2) Y7: ((SPANSTART 0)
(SPANEND 4) (SPANEND 4)
(LEX $WRH) (LEX B$WRH)
(POS N) (POS LEX))
(GEN F)
(NUM S)
(STATUS ABSOLUTE))
Sep 22, 2006
Learning-based MT with Limited Resources

25. Learning-based MT with Limited Resources
Example Translation
Input:
לאחר דיונים רבים החליטה הממשלה לערוך משאל עם בנושא הנסיגה
After debates many decided the government to hold referendum in issue the withdrawal
Output:
AFTER MANY DEBATES THE GOVERNMENT DECIDED TO HOLD A REFERENDUM ON THE ISSUE OF THE WITHDRAWAL
Sep 22, 2006
Learning-based MT with Limited Resources

26. Sample Output (dev-data)
maxwell anurpung comes from ghana for israel four years ago and since worked in cleaning in hotels in eilat
a few weeks ago announced if management club hotel that for him to leave israel according to the government instructions and immigration police
in a letter in broken english which spread among the foreign workers thanks to them hotel for their hard work and announced that will purchase for hm flight tickets for their countries from their money
Sep 22, 2006
Learning-based MT with Limited Resources

27. Challenges and Future Directions
Automatic Transfer Rule Learning:
Learning mappings for non-compositional structures
Effective models for rule scoring for
Decoding: using scores at runtime
Pruning the large collections of learned rules
Learning Unification Constraints
In the absence of morphology or POS annotated lexica
Integrated Xfer Engine and Decoder
Improved models for scoring tree-to-tree mappings, integration with LM and other knowledge sources in the course of the search
Sep 22, 2006
Learning-based MT with Limited Resources

28. Challenges and Future Directions
Our approach for learning transfer rules is applicable to the large parallel data scenario, subject to solutions for several big challenges:
No elicitation corpus  break-down parallel sentences into reasonable learning examples
Working with less reliable automatic word alignments rather than manual alignments
Effective use of reliable parse structures for ONE language (i.e. English) and automatic word alignments in order to decompose the translation of a sentence into several compositional rules.
Effective scoring of resulting very large transfer grammars, and scaled up transfer + decoding
Sep 22, 2006
Learning-based MT with Limited Resources

29. Future Research Directions
Automatic Rule Refinement
Morphology Learning
Feature Detection and Corpus Navigation …
Sep 22, 2006
Learning-based MT with Limited Resources

30. Implications for MT with Vast Amounts of Parallel Data
Phrase-to-phrase MT ill suited for long-range reorderings  ungrammatical output
Recent work on hierarchical Stat-MT [Chiang, 2005] and parsing-based MT [Melamed et al, 2005] [Knight et al]
Learning general tree-to-tree syntactic mappings is equally problematic:
Meaning is a hybrid of complex, non-compositional phrases embedded within a syntactic structure
Some constituents can be translated in isolation, others require contextual mappings
Sep 22, 2006
Learning-based MT with Limited Resources

31. Learning-based MT with Limited Resources
Evaluation Results
Test set of 62 sentences from Haaretz newspaper, 2 reference translations
System
BLEU
NIST P R
METEOR
No Gram
0.0616
3.4109
0.4090
0.4427
0.3298
Learned
0.0774
3.5451
0.4189
0.4488
0.3478
Manual
0.1026
3.7789
0.4334
0.4474
0.3617
Sep 22, 2006
Learning-based MT with Limited Resources

32. Hebrew-English: Test Suite Evaluation
Grammar
BLEU
METEOR
Baseline (NoGram)
0.0996
0.4916
Learned Grammar
0.1608
0.5525
Manual Grammar
0.1642
0.5320
Sep 22, 2006
Learning-based MT with Limited Resources

33. Learning-based MT with Limited Resources
QuechuaSpanish MT
V-Unit: funded Summer project in Cusco (Peru) June-August 2005 [preparations and data collection started earlier]
Intensive Quechua course in Centro Bartolome de las Casas (CBC)
Worked together with two Quechua native and one non-native speakers on developing infrastructure (correcting elicited translations, segmenting and translating list of most frequent words)
Sep 22, 2006
Learning-based MT with Limited Resources

34. Quechua  Spanish Prototype MT System
Stem Lexicon (semi-automatically generated): 753 lexical entries
Suffix lexicon: 21 suffixes
(150 Cusihuaman)
Quechua morphology analyzer
25 translation rules
Spanish morphology generation module
User-Studies: 10 sentences, 3 users (2 native, 1 non-native)
Sep 22, 2006
Learning-based MT with Limited Resources

35. Learning-based MT with Limited Resources
The Transfer Engine
Analysis
Source text is parsed into its grammatical structure. Determines transfer application ordering.
Example:
他 看 书。(he read book) S
NP VP
N V NP
他 看 书
Transfer
A target language tree is created by reordering, insertion, and deletion.
he read DET N
a book
Article “a” is inserted into object NP. Source words translated with transfer lexicon.
Generation
Target language constraints are checked and final translation produced.
E.g. “reads” is chosen over “read” to agree with “he”.
Final translation:
“He reads a book”
Sep 22, 2006
Learning-based MT with Limited Resources

36. Learning-based MT with Limited Resources
The Transfer Engine
Some Unique Features:
Works with either learned or manually-developed transfer grammars
Handles rules with or without unification constraints
Supports interfacing with servers for Morphological analysis and generation
Can handle ambiguous source-word analyses and/or SL segmentations represented in the form of lattice structures
Sep 22, 2006
Learning-based MT with Limited Resources

37. Learning-based MT with Limited Resources
The Lattice Decoder
Simple Stack Decoder, similar in principle to SMT/EBMT decoders
Searches for best-scoring path of non-overlapping lattice arcs
Scoring based on log-linear combination of scoring components (no MER training yet)
Scoring components:
Standard trigram LM
Fragmentation: how many arcs to cover the entire translation?
Length Penalty
Rule Scores (not fully integrated yet)
Sep 22, 2006
Learning-based MT with Limited Resources

38. Learning-based MT with Limited Resources
Outline
Rationale for learning-based MT
Roadmap for learning-based MT
Framework overview
Elicitation
Learning transfer Rules
Automatic rule refinement
Example prototypes
Implications for MT with vast parallel data
Conclusions and future directions
Sep 22, 2006
Learning-based MT with Limited Resources

39. Data Elicitation for Languages with Limited Resources
Rationale:
Large volumes of parallel text not available  create a small maximally-diverse parallel corpus that directly supports the learning task
Bilingual native informant(s) can translate and align a small pre-designed elicitation corpus, using elicitation tool
Elicitation corpus designed to be typologically and structurally comprehensive and compositional
Transfer-rule engine and new learning approach support acquisition of generalized transfer-rules from the data
Sep 22, 2006
Learning-based MT with Limited Resources

40. Elicitation Tool: English-Chinese Example
Sep 22, 2006
Learning-based MT with Limited Resources

41. Elicitation Tool: English-Chinese Example
Sep 22, 2006
Learning-based MT with Limited Resources

42. Elicitation Tool: English-Hindi Example
Sep 22, 2006
Learning-based MT with Limited Resources

43. Elicitation Tool: English-Arabic Example
Sep 22, 2006
Learning-based MT with Limited Resources

44. Elicitation Tool: Spanish-Mapudungun Example
Sep 22, 2006
Learning-based MT with Limited Resources

45. Designing Elicitation Corpora
What do we want to elicit?
Diversity of linguistic phenomena and constructions
Syntactic structural diversity
How do we construct an elicitation corpus?
Typological Elicitation Corpus based on elicitation and documentation work of field linguists (e.g. Comrie 1977, Bouquiaux 1992): initial corpus size ~1000 examples
Structural Elicitation Corpus based on representative sample of English phrase structures: ~120 examples
Organized compositionally: elicit simple structures first, then use them as building blocks
Goal: minimize size, maximize linguistic coverage
Sep 22, 2006
Learning-based MT with Limited Resources

46. Typological Elicitation Corpus
Feature Detection
Discover what features exist in the language and where/how they are marked
Example: does the language mark gender of nouns? How and where are these marked?
Method: compare translations of minimal pairs – sentences that differ in only ONE feature
Elicit translations/alignments for detected features and their combinations
Dynamic corpus navigation based on feature detection: no need to elicit for combinations involving non-existent features
Sep 22, 2006
Learning-based MT with Limited Resources

47. Typological Elicitation Corpus
Initial typological corpus of about 1000 sentences was manually constructed
New construction methodology for building an elicitation corpus using:
A feature specification: lists inventory of available features and their values
A definition of the set of desired feature structures
Schemas define sets of desired combinations of features and values
Multiplier algorithm generates the comprehensive set of feature structures
A generation grammar and lexicon: NLG generator generates NL sentences from the feature structures
Sep 22, 2006
Learning-based MT with Limited Resources

48. Structural Elicitation Corpus
Goal: create a compact diverse sample corpus of syntactic phrase structures in English in order to elicit how these map into the elicited language
Methodology:
Extracted all CFG “rules” from Brown section of Penn TreeBank (122K sentences)
Simplified POS tag set
Constructed frequency histogram of extracted rules
Pulled out simplest phrases for most frequent rules for NPs, PPs, ADJPs, ADVPs, SBARs and Sentences
Some manual inspection and refinement
Resulting corpus of about 120 phrases/sentences representing common structures
See [Probst and Lavie, 2004]
Sep 22, 2006
Learning-based MT with Limited Resources

49. Learning-based MT with Limited Resources
Outline
Rationale for learning-based MT
Roadmap for learning-based MT
Framework overview
Elicitation
Learning transfer Rules
Automatic rule refinement
Example prototypes
Implications for MT with vast parallel data
Conclusions and future directions
Sep 22, 2006
Learning-based MT with Limited Resources

50. Flat Seed Rule Generation
Create a “flat” transfer rule specific to the sentence pair, partially abstracted to POS
Words that are aligned word-to-word and have the same POS in both languages are generalized to their POS
Words that have complex alignments (or not the same POS) remain lexicalized
One seed rule for each translation example
No feature constraints associated with seed rules (but mark the example(s) from which it was learned)
Sep 22, 2006
Learning-based MT with Limited Resources

51. Compositionality Learning
Detection: traverse the c-structure of the English sentence, add compositional structure for translatable chunks
Generalization: adjust constituent sequences and alignments
Two implemented variants:
Safe Compositionality: there exists a transfer rule that correctly translates the sub-constituent
Maximal Compositionality: Generalize the rule if supported by the alignments, even in the absence of an existing transfer rule for the sub-constituent
Sep 22, 2006
Learning-based MT with Limited Resources

52. Learning-based MT with Limited Resources
Constraint Learning
Goal: add appropriate feature constraints to the acquired rules
Methodology:
Preserve general structural transfer
Learn specific feature constraints from example set
Seed rules are grouped into clusters of similar transfer structure (type, constituent sequences, alignments)
Each cluster forms a version space: a partially ordered hypothesis space with a specific and a general boundary
The seed rules in a group form the specific boundary of a version space
The general boundary is the (implicit) transfer rule with the same type, constituent sequences, and alignments, but no feature constraints
Sep 22, 2006
Learning-based MT with Limited Resources

53. Constraint Learning: Generalization
The partial order of the version space:
Definition: A transfer rule tr1 is strictly more general than another transfer rule tr2 if all f-structures that are satisfied by tr2 are also satisfied by tr1.
Generalize rules by merging them:
Deletion of constraint
Raising two value constraints to an agreement constraint, e.g.
((x1 num) = *pl), ((x3 num) = *pl) 
((x1 num) = (x3 num))
Sep 22, 2006
Learning-based MT with Limited Resources

54. Automated Rule Refinement
Bilingual informants can identify translation errors and pinpoint the errors
A sophisticated trace of the translation path can identify likely sources for the error and do “Blame Assignment”
Rule Refinement operators can be developed to modify the underlying translation grammar (and lexicon) based on characteristics of the error source:
Add or delete feature constraints from a rule
Bifurcate a rule into two rules (general and specific)
Add or correct lexical entries
See [Font-Llitjos, Carbonell & Lavie, 2005]
Sep 22, 2006
Learning-based MT with Limited Resources

55. Learning-based MT with Limited Resources
Outline
Rationale for learning-based MT
Roadmap for learning-based MT
Framework overview
Elicitation
Learning transfer Rules
Automatic rule refinement
Example prototypes
Implications for MT with vast parallel data
Conclusions and future directions
Sep 22, 2006
Learning-based MT with Limited Resources

56. Learning-based MT with Limited Resources
Outline
Rationale for learning-based MT
Roadmap for learning-based MT
Framework overview
Elicitation
Learning transfer Rules
Automatic rule refinement
Learning Morphology
Example prototypes
Implications for MT with vast parallel data
Conclusions and future directions
Sep 22, 2006
Learning-based MT with Limited Resources

57. Implications for MT with Vast Amounts of Parallel Data
Example:
他 经常 与 江泽民 总统 通 电话
He freq with J Zemin Pres via phone
He freq talked with President J Zemin over the phone
Sep 22, 2006
Learning-based MT with Limited Resources

58. Implications for MT with Vast Amounts of Parallel Data
Example:
他 经常 与 江泽民 总统 通 电话
He freq with J Zemin Pres via phone
He freq talked with President J Zemin over the phone
NP1
NP2
NP3
NP1
NP2
NP3
Sep 22, 2006
Learning-based MT with Limited Resources

59. Learning-based MT with Limited Resources
Conclusions
There is hope yet for wide-spread MT between many of the worlds language pairs
MT offers a fertile yet extremely challenging ground for learning-based approaches that leverage from diverse sources of information:
Syntactic structure of one or both languages
Word-to-word correspondences
Decomposable units of translation
Statistical Language Models
AVENUE’s XFER approach provides a feasible solution to MT for languages with limited resources
Promising approach for addressing the fundamental weaknesses in current corpus-based MT for languages with vast resources
Sep 22, 2006
Learning-based MT with Limited Resources

60. Learning-based MT with Limited Resources
Sep 22, 2006
Learning-based MT with Limited Resources

61. Mapudungun-to-Spanish Example
English
I didn’t see Maria
Mapudungun
pelafiñ Maria
Spanish
No vi a María
Sep 22, 2006
Learning-based MT with Limited Resources

62. Mapudungun-to-Spanish Example
English
I didn’t see Maria
Mapudungun
pelafiñ Maria
pe -la -fi -ñ Maria
see -neg -3.obj -1.subj.indicative Maria
Spanish
No vi a María
No vi a María
neg see.1.subj.past.indicative acc Maria
Sep 22, 2006
Learning-based MT with Limited Resources

63. Learning-based MT with Limited Resources
pe-la-fi-ñ Maria V pe
Sep 22, 2006
Learning-based MT with Limited Resources

64. Learning-based MT with Limited Resources
pe-la-fi-ñ Maria V pe
VSuff
Negation = + la
Sep 22, 2006
Learning-based MT with Limited Resources

65. Learning-based MT with Limited Resources
pe-la-fi-ñ Maria V pe
VSuffG
Pass all features up
VSuff la
Sep 22, 2006
Learning-based MT with Limited Resources

66. Learning-based MT with Limited Resources
pe-la-fi-ñ Maria V pe
VSuffG
VSuff
object person = 3
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

67. Learning-based MT with Limited Resources
pe-la-fi-ñ Maria V pe
VSuffG
Pass all features up from both children
VSuffG
VSuff
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

68. Learning-based MT with Limited Resources
pe-la-fi-ñ Maria V pe
VSuffG
VSuff
person = 1
number = sg
mood = ind
VSuffG
VSuff ñ
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

69. Learning-based MT with Limited Resources
pe-la-fi-ñ Maria V
VSuffG pe
VSuffG
VSuff
Pass all features up from both children
VSuffG
VSuff ñ
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

70. Learning-based MT with Limited Resources
pe-la-fi-ñ Maria
Pass all features up from both children V
Check that:
1) negation = +
2) tense is undefined V
VSuffG pe
VSuffG
VSuff
VSuffG
VSuff ñ
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

71. Learning-based MT with Limited Resources
pe-la-fi-ñ Maria V NP V
VSuffG N
person = 3
number = sg
human = + pe
VSuffG
VSuff N
VSuffG
VSuff ñ
Maria
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

72. Learning-based MT with Limited Resources
pe-la-fi-ñ Maria S
Check that NP is human = +
Pass features up from V VP V NP V
VSuffG N pe
VSuffG
VSuff N
VSuffG
VSuff ñ
Maria
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

73. Learning-based MT with Limited Resources
Transfer to Spanish: Top-Down S S VP VP V NP V
VSuffG N pe
VSuffG
VSuff N
VSuffG
VSuff ñ
Maria
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

74. Learning-based MT with Limited Resources
Transfer to Spanish: Top-Down
Pass all features
to Spanish side S S VP VP V NP V
“a” NP V
VSuffG N pe
VSuffG
VSuff N
VSuffG
VSuff ñ
Maria
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

75. Learning-based MT with Limited Resources
Transfer to Spanish: Top-Down S S
Pass all features
down VP VP V NP V
“a” NP V
VSuffG N pe
VSuffG
VSuff N
VSuffG
VSuff ñ
Maria
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

76. Learning-based MT with Limited Resources
Transfer to Spanish: Top-Down S S
Pass object features
down VP VP V NP V
“a” NP V
VSuffG N pe
VSuffG
VSuff N
VSuffG
VSuff ñ
Maria
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

77. Learning-based MT with Limited Resources
Transfer to Spanish: Top-Down S S VP VP V NP V
“a” NP V
VSuffG N
Accusative marker on objects is introduced because human = + pe
VSuffG
VSuff N
VSuffG
VSuff ñ
Maria
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

78. Learning-based MT with Limited Resources
Transfer to Spanish: Top-Down S S VP
VP::VP [VBar NP] -> [VBar "a" NP]
( (X1::Y1) (X2::Y3)
((X2 type) = (*NOT* personal))
((X2 human) =c +)
(X0 = X1)
((X0 object) = X2)
(Y0 = X0)
((Y0 object) = (X0 object))
(Y1 = Y0)
(Y3 = (Y0 object))
((Y1 objmarker person) = (Y3 person))
((Y1 objmarker number) = (Y3 number))
((Y1 objmarker gender) = (Y3 ender))) VP V NP V
“a” NP V
VSuffG N pe
VSuffG
VSuff N
VSuffG
VSuff ñ
Maria
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

79. Learning-based MT with Limited Resources
Transfer to Spanish: Top-Down S S
Pass person, number, and mood features to Spanish Verb VP VP V NP V
“a” NP
Assign tense = past V
VSuffG N
“no” V pe
VSuffG
VSuff N
VSuffG
VSuff ñ
Maria
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

80. Learning-based MT with Limited Resources
Transfer to Spanish: Top-Down S S VP VP V NP V
“a” NP V
VSuffG N
“no” V pe
VSuffG
VSuff N
VSuffG
VSuff ñ
Maria
Introduced because negation = +
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

81. Learning-based MT with Limited Resources
Transfer to Spanish: Top-Down S S VP VP V NP V
“a” NP V
VSuffG N
“no” V pe
VSuffG
VSuff N
ver
VSuffG
VSuff ñ
Maria
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

82. Learning-based MT with Limited Resources
Transfer to Spanish: Top-Down S S VP VP V NP V
“a” NP V
VSuffG N
“no” V pe
VSuffG
VSuff N
ver vi
VSuffG
VSuff ñ
Maria
person = 1
number = sg
mood = indicative
tense = past
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

83. Learning-based MT with Limited Resources
Transfer to Spanish: Top-Down S S
Pass features over to Spanish side VP VP V NP V
“a” NP V
VSuffG N
“no” V N pe
VSuffG
VSuff N vi N
VSuffG
VSuff ñ
Maria
María
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

84. Learning-based MT with Limited Resources
I Didn’t see Maria S S VP VP V NP V
“a” NP V
VSuffG N
“no” V N pe
VSuffG
VSuff N vi N
VSuffG
VSuff ñ
Maria
María
VSuff fi la
Sep 22, 2006
Learning-based MT with Limited Resources

85. Learning-based MT with Limited Resources
Sep 22, 2006
Learning-based MT with Limited Resources