UDD-1 / AL_CYCLE1_REPORT.md
rain1024's picture
Restructure technical report to ACL format, add Phase 0 gold eval methodology
bbc68b5

AL Cycle 1 Report — Word Segmentation

Overview

Item Value
Cycle 1 (first gold annotation round)
Source Top 100 highest-uncertainty sentences from AL Cycle 0
Annotation tool Label Studio (project 4, labels.ankiren.com)
Annotator 1 person
Tasks imported 100
Tasks annotated 95 (5 not annotated)
Tasks exported 92 (3 skipped due to annotation gaps)
Output gold_ws_cycle1.txt (BIO format), ls_export_cycle1.json

Sentence Selection

Sentences were selected by CRF token marginal uncertainty scoring from AL Cycle 0:

  • CRF model: tree-1/models/word_segmentation/udd_ws_v1_1-20260212_065135/
  • Scored 20,000 sentences (dev + test splits)
  • Top 100 by composite score: mean uncertainty × (1 + boundary_weight) + 0.01 × n_long_tokens
  • These represent the hardest sentences for the CRF, not a random sample

Domain Distribution

Domain Count %
Fiction 28 30.4%
Wikipedia 28 30.4%
Non-fiction 24 26.1%
News 8 8.7%
Legal 4 4.3%

Fiction and Wikipedia are overrepresented because they contain more difficult segmentation cases (literary names, Hán-Việt compounds, foreign names).

Skipped Sentences

3 sentences skipped due to annotation gaps (syllables left untagged in Label Studio):

sent_id Issue
uvn-3488 Foreign product names (Derma peel One) left untagged
uvb-n-10657 Single syllable "là" left untagged
uvb-f-10312 Punctuation "»" left untagged

CRF Silver vs Gold Evaluation

Overall Metrics (92 sentences, 1,506 syllables)

Metric Score
Syllable Accuracy 0.8665 (1,305/1,506)
Word Precision 0.7934
Word Recall 0.7685
Word F1 0.7808

Context: The CRF baseline on silver data achieves Word F1 = 0.9834. The much lower F1 = 0.7808 on these 92 gold sentences confirms that AL successfully selected the most error-prone sentences.

Boundary Change Statistics

Change Count Meaning
B→I 83 Silver has B (word start), gold has I (continuation) → silver over-splits
I→B 118 Silver has I (continuation), gold has B (word start) → silver over-merges
Total changes 201 13.3% of all syllables have wrong boundary tags

Per-Domain Breakdown

Domain N Syllables Syl Acc Precision Recall F1
News 8 213 0.9061 0.8485 0.8296 0.8390
Non-fiction 24 414 0.8889 0.8235 0.8235 0.8235
Fiction 28 412 0.8786 0.8264 0.8041 0.8151
Legal 4 81 0.8395 0.7347 0.6923 0.7129
Wikipedia 28 386 0.8317 0.7289 0.6868 0.7072

Wikipedia is the hardest domain (F1 = 0.707) due to Hán-Việt classical text, foreign names, and specialized terminology. Legal is also low (F1 = 0.713) but with only 4 sentences.

Error Analysis

75 out of 92 sentences (81.5%) have at least one segmentation difference between silver and gold.

Error Type Summary

Error Type Count Description
Over-merge 134 CRF merged syllables across word boundaries
Over-split 119 CRF split syllables that belong to one word
Total 253

Over-merge is slightly more common (53%) than over-split (47%).

Error Category Taxonomy

1. Cross-boundary proper name merge (most frequent)

The CRF frequently merges the last syllable of a proper name with the next word:

Silver (wrong) Gold (correct) Domain
Nguyễn_Bính_Mưa_dầm Nguyễn_Bính | Mưa | dầm fiction
Chu_Du_tính_tình Chu_Du | tính_tình non-fiction
Tăng_Hoàng_Vinh_giữ Tăng_Hoàng_Vinh | giữ news
Khổng_Dĩnh_Đạt_thời Khổng_Dĩnh_Đạt | thời wikipedia
Hà_Tiên_trấn Hà_Tiên | trấn wikipedia
Quỳnh_nghiến Quỳnh | nghiến_răng fiction
Khải_nheo Khải | nheo_nheo fiction

Root cause: CRF sees proper name continuation (I-W) and extends it to the next syllable, especially when the next word starts with a common syllable.

2. Hán-Việt classical text boundary errors (Wikipedia-specific)

Many Wikipedia sentences contain classical Chinese quotations where the CRF fails completely:

Silver (wrong) Gold (correct) sent_id
thiên_địa_lưu thiên_địa | lưu uvw-11634
tất_hữu_tụng tất | hữu | tụng uvw-379
chi_dĩ_tụng chi | | tụng uvw-379
ẩm_thực_chi ẩm_thực | chi uvw-379
Đinh_Lê_nhị_gia Đinh_Lê | nhị_gia uvw-4040

Root cause: Classical Hán-Việt text uses mostly single-syllable words, but the CRF trained on modern Vietnamese aggressively merges adjacent syllables.

3. Foreign word/abbreviation merges

Silver (wrong) Gold (correct) sent_id
I_LOVE I | LOVE uvb-f-16922
BST_Love BST | Love uvn-16886
DNA_replication DNA | replication uvw-16915
Frankael-Zermelo_Tiên Frankael-Zermelo | Tiên_đề uvw-14316

Root cause: CRF treats consecutive Latin-script tokens as one word.

4. Compound word boundary shifts

CRF merges across true word boundaries, creating invalid compounds:

Silver (wrong) Gold (correct) Type
lợi_nhuận_tích + lũy lợi_nhuận + tích_lũy boundary shift
niên_đại_mẫu + vật niên_đại + mẫu_vật boundary shift
phóng_vệ_tinh phóng + vệ_tinh over-merge
quy_phạm_pháp_luật quy_phạm + pháp_luật over-merge (legal)
trình_bày_văn_bản trình_bày + văn_bản over-merge (legal)

5. Common compound under-merges (over-splits)

CRF fails to merge well-known compound words:

Silver (wrong) Gold (correct)
Ủy | ban Ủy_ban (committee)
lính | thú lính_thú (soldier)
mu | rùa mu_rùa (turtle shell)
trêu | ghẹo trêu_ghẹo (tease)
sương | mai sương_mai (morning dew)
mái | nhà mái_nhà (roof)
đầm | đuôi | đầm_đuôi_cá (fishtail dress)
cha | truyền_con | nối cha_truyền_con_nối (hereditary)
siêu | máy_tính siêu_máy_tính (supercomputer)
lượng | tử_lai lượng_tử (quantum) + lai

6. Proper name under-merges

Silver (wrong) Gold (correct)
Trương_Hán | Siêu Trương_Hán_Siêu
Nữ_minh | tinh Nữ_minh_tinh
Nicolas | Fatio_de | Duillier Nicolas_Fatio_de_Duillier
Nigeria_Ibrahim | Baré_Maïnassara Nigeria_Ibrahim_Baré_Maïnassara
World_Music | Awards World_Music_Awards

All Sentence-Level Differences

Fiction (20/28 sentences with differences)

uvb-f-10494: dạo_khúcdạo | khúc

uvb-f-1078: Mặt_Kỳ + Nam_xụMặt | Kỳ_Nam | xụ

uvb-f-11528: thần_linh_giữthần_linh | giữ

uvb-f-116: lính | thúlính_thú

uvb-f-12571: nỡ_tẩy + chaynỡ | tẩy_chay

uvb-f-14791: như + điên_sợnhư_điên | sợ

uvb-f-16030: Quỳnh_nghiến + răngQuỳnh | nghiến_răng

uvb-f-16202: Nữ_minh + tinh_khịt + mũiNữ_minh_tinh | khịt_mũi

uvb-f-16922: Ahern_PS + I_LOVE + YOU_ChươngAhern | PS | I | LOVE | YOU | Chương

uvb-f-17901: gia_nhân_túc_trực + chăn | lợngia_nhân | túc_trực | chăn_lợn

uvb-f-18098: Khải_nheo + nheoKhải | nheo_nheo

uvb-f-2195: người | thương + Bờ | xe | nước_tạc + chi_baongười_thương | Bờ_xe_nước | tạc | chi | bao

uvb-f-3070: Xin | chào_sĩ_quanXin_chào | sĩ_quan

uvb-f-6015: Liên_xua | tay + nói | gởLiên | xua_tay | nói_gở

uvb-f-6292: Trương_Hán | Siêu_bưng + khay | trà + lên_ánTrương_Hán_Siêu | bưng | khay_trà | lên | án

uvb-f-6893: Nguyễn_Bính_Mưa_dầmNguyễn_Bính | Mưa | dầm

uvb-f-7016: Nguyễn_Bính_Công + diều_mướnNguyễn_Bính | Công | diều | mướn

uvb-f-7435: Mặt_San + y_dơn | dớtMặt | San | y | dơn_dớt

uvb-f-8734: có_học + bơi | chó | học | bơi_chó

uvb-f-9174: tạo_nghiệp_lànhtạo | nghiệp_lành

Non-fiction (18/24 sentences with differences)

uvb-n-10024: Sống_biến + Lạnh_biếnSống | biến | Lạnh | biến

uvb-n-11587: phải_biếtphải | biết

uvb-n-11976: Lệ | thủy_trìnhLệ_thủy | trình

uvb-n-12321: sương | mai_phớt + mái | nhàsương_mai | phớt | mái_nhà

uvb-n-13233: lạnh_đặclạnh | đặc

uvb-n-13288: I._Lương + Huệ_vươngI. | Lương_Huệ_vương

uvb-n-16553: đương_ngủ | ngàyđương | ngủ_ngày

uvb-n-18226: Ngữ | đoạn + Khái_niệm_ngữ | đoạnNgữ_đoạn | Khái_niệm | ngữ_đoạn

uvb-n-18736: Lục | Giả_đápLục_Giả | đáp

uvb-n-19309: Lạc | Hồng_định + trêu | ghẹoLạc_Hồng | định | trêu_ghẹo

uvb-n-2201: quyết_phục_thùquyết | phục_thù

uvb-n-3769: Chu_Du_tính_tìnhChu_Du | tính_tình

uvb-n-3847: răng | nanh_sắc | nhọn + triết_gia_lùirăng_nanh | sắc_nhọn | triết_gia | lùi

uvb-n-3874: thành | thực_cầu | mong + cao_minh_phủ | chínhthành_thực | cầu_mong | cao_minh | phủ_chính

uvb-n-3968: đồng | tự + làm_âmđồng_tự | làm | âm

uvb-n-511: Ủy | ban + Hành | chánhỦy_ban | Hành_chánh

uvb-n-6010: chơi_tròchơi | trò

uvb-n-7402: mu | rùamu_rùa

uvb-n-7669: Cung | Khôn_Thái + Cần_Thánh_chỗCung_Khôn_Thái | Cần_Thánh | chỗ

Wikipedia (22/28 sentences with differences)

uvw-10198: Hữu_thưởng + vu_đại_quốcHữu | thưởng | vu | đại_quốc

uvw-11634: Thệ_tâm + thiên_địa_lưu + xỉ_giang + sơn_thổ + thiệt_hồngThệ | tâm | thiên_địa | lưu | xỉ | giang_sơn | thổ | thiệt | hồng

uvw-13194: Quân_xướng + thần_họaQuân | xướng | thần | họa

uvw-13430: Đường | thẳngĐường_thẳng

uvw-14316: Hệ_tiên | đề (×2) + đề_số | học + Frankael-Zermelo_Tiên | đề_chọn + Tiên | đề + tiên | đềHệ_tiên_đề (×2) | số_học | Frankael-Zermelo | Tiên_đề (×2) | chọn | tiên_đề

uvw-1451: Quan_Âm_Diệu_ThiệnQuan_Âm | Diệu_Thiện

uvw-15121: Nhà_Habsburg + bắt_đầu_tích | lũy + cha | truyền_con | nốiNhà | Habsburg | bắt_đầu | tích_lũy | cha_truyền_con_nối

uvw-15306: Tống_tiêu_diệtTống | tiêu_diệt

uvw-15392: Hải_quốc_văn_kiến | lục_》Hải_quốc_văn_kiến_lục |

uvw-15619: Long | Hồ_dinh_đổi + Hà_Tiên_trấnLong_Hồ | dinh | đổi | Hà_Tiên | trấn

uvw-16915: DNA_replicationDNA | replication

uvw-17218: Điểm_nóng | chảyĐiểm | nóng_chảy

uvw-17235: Khổng_Dĩnh | Đạt_thời + Hoa_Hạ_viKhổng_Dĩnh_Đạt | thời | Hoa_Hạ | vi

uvw-18245: ngăn_ắc + quy_ổn_địnhngăn | ắc_quy | ổn_định

uvw-19010: nhà_toán_học + Nicolas | Fatio_de | Duilliernhà | toán_học | Nicolas_Fatio_de_Duillier

uvw-2178: Đồng_nguyên + chất_mềm + đồng_tươiĐồng | nguyên_chất | mềm | đồng | tươi

uvw-3214: Thiên_địa_hòa | xướng + chi_tượng + giao | hòaThiên_địa | hòa_xướng | chi | tượng | giao_hòa

uvw-379: Phục | Hy_ghi + ẩm_thực_chi + tất_hữu_tụng + cố_thụ + chi_dĩ_tụngPhục_Hy | ghi | ẩm_thực | chi (×2) | tất | hữu | tụng (×2) | cố | thụ |

uvw-4040: Đinh_Lê_nhị_gia + võng_đạo + an_quyết + đại_phất | thế + toán | sốĐinh_Lê | nhị_gia | võng | đạo | an | quyết | thế_đại | phất | toán_số

uvw-415: Chỉ | Chỉ_dã

uvw-4428: Nhu_vu + trinh_cátNhu | vu | trinh | cát

uvw-5791: Kỳ_địa + quảng_nhi + quyết_thổ + cao_nhi + sảng | khảiKỳ | địa | quảng | nhi (×2) | quyết | thổ | cao | sảng_khải

uvw-6347: miệt_hôn_điếm + chi_khốn + chi_phongmiệt | hôn_điếm | chi (×2) | khốn | phong

uvw-7555: Quân_MỹQuân | Mỹ

uvw-8056: Nigeria_Ibrahim | Baré_MaïnassaraNigeria_Ibrahim_Baré_Maïnassara

uvw-8082: Grand | Prix_Award + MTV_Video | Music | Awards_Japan + World_Music | Awards + Japan_Gold | Disc_AwardGrand_Prix_Award | MTV_Video_Music_Awards_Japan | World_Music_Awards | Japan_Gold_Disc_Award

News (7/8 sentences with differences)

uvn-16886: đầm | đuôi | + thân | áo + BST_Loveđầm_đuôi_cá | thân_áo | BST | Love

uvn-18693: niên_đại_mẫu | vậtniên_đại | mẫu_vật

uvn-18725: phóng_vệ_tinhphóng | vệ_tinh

uvn-19673: Tăng_Hoàng_Vinh_giữTăng_Hoàng_Vinh | giữ

uvn-4740: lợi_nhuận_tích | lũylợi_nhuận | tích_lũy

uvn-6121: of_refusing + Matt_really + new_leaf + and_hasof | refusing | Matt | really | new | leaf | and | has

uvn-964: siêu | máy_tính + lượng | tử_laisiêu_máy_tính | lượng_tử | lai

Legal (3/4 sentences with differences)

vlc-10445: miễn_thị_thựcmiễn | thị_thực

vlc-6220: Ủy | ban_thường_vụ + trình_bày_văn_bản + quy_phạm_pháp_luật + Chủ_tịch | nướcỦy_ban | thường_vụ | trình_bày | văn_bản | quy_phạm | pháp_luật | Chủ_tịch_nước

vlc-7243: phong_quân (×2) + hàm_sĩ_quan (×2) + tại | ngũphong (×2) | quân_hàm (×2) | sĩ_quan (×2) | tại_ngũ

Key Takeaways

  1. AL selection works: The 92 gold sentences have Word F1 = 0.78, much lower than the 0.98 silver baseline, confirming that uncertainty scoring correctly identifies hard cases.

  2. Cross-boundary merge is the #1 error: CRF extends proper name spans into the next word, creating invalid compounds like Nguyễn_Bính_Mưa_dầm.

  3. Wikipedia is hardest: F1 = 0.707, driven by Hán-Việt classical text (single-syllable words mismerged) and foreign names.

  4. Over-merge slightly > over-split: 134 vs 119 errors. The CRF has a slight bias toward merging.

  5. Legal domain has systematic issues: 4-syllable compound over-merges (quy_phạm_pháp_luật, trình_bày_văn_bản) and boundary shifts on quân_hàm/sĩ_quan.

  6. Foreign text always over-merged: English words (I_LOVE, DNA_replication, new_leaf) are treated as Vietnamese compounds.

Recommendations for Cycle 2

  1. Add proper name boundary features to CRF: capitalization patterns, known name lists
  2. Add foreign word detection: Latin-script tokens should default to single-word segmentation
  3. Increase legal domain sampling: Only 4 sentences in Cycle 1, need more legal-specific gold data
  4. Classical Hán-Việt: Consider special handling or separate model for classical text passages
  5. Retrain CRF with the 92 gold corrections mixed into training data, then re-score remaining 19,900 sentences for Cycle 2 selection