人間とAIの協働（駒場祭2023）

Transcript

1. ೥౓ۨ৔ࡇ
   ެ։ߨ࠲ 
 
 ਓؒͱ"*ͷڠಇ ౦ژେֶ૯߹จԽݚڀՊ 
 ޿ҬՊֶઐ߈ 
 അ৔ઇ೫
2. 
   "*ʹͰ͖Δ͜ͱ͕૿͖͍͑ͯͯΔ
   
   "*ͷௐ੔ʹ͸ਓؒͷڠྗ͕ෆՄܽ
   
   "*ʹ͸ϚδϣϦςΟͷߟ͕͑൓ө͞Ε΍͍͢
   
   "*ʹҙࢥܾఆશͯΛ೚ͤΔͷͰ͸ͳ͘ 
 ਓؒͷҙࢥܾఆΛ"*͕ࢧԉ͢Δ͜ͱ͕ॏཁ ΞδΣϯμ 2
3. $IBU(15
   ࣗવݴޠͰͷࢦࣔʹԠ͑Δ"* 3
4. $IBU(15͸ϒϨΠϯετʔϛϯάʹ΋׆༻Ͱ͖Δ 4
5. $IBU(15͸਺ֶͷ໰୊Λղ͘͜ͱ΋Ͱ͖Δ 5
6. $IBU(15͸άϥϑͷϓϩοτ΋Ͱ͖Δ 6
7. $IBU(15͸σʔλ෼ੳ΋Ͱ͖Δ 7
8. $IBU(15ͱ%"--&Ͱର࿩తͳը૾ੜ੒͕Ͱ͖Δ 8
9. $IBU(15͸ը૾ʹ͍ͭͯͷ࣭໰Ԡ౴΋Ͱ͖Δ 9 https://openai.com/blog/chatgpt-can-now-see-hear-and-speak ʮࣗసंͷ 
 αυϧͷԼ͛ํΛ ڭ͑ͯཉ͍͠ʯ ʮʰϨόʔʱ͸ 
 ͜Ε͔ʁʯ ʮϚχϡΞϧͱखݩͷπʔ ϧ͸͜Ε͕ͩɺ͜ͷπʔϧ ͰରԠͰ͖Δ͔ʁʯ
10. $IBU(15͸֎෦αʔϏεͱͷ࿈ܞ΋Ͱ͖Δ 10 https://openai.com/blog/chatgpt-plugins ʮϏʔΨϯରԠͷϨετϥ ϯͷ༧໿Λͯ͠΄͍͠ʯ ʮ؆୯ͳϨγϐͷਪનͱ 
 ৯ࡐͷ஫จΛͯ͠ཉ͍͠ʯ 0QFO5BCMFͰళΛݟ͚ͭ ͯ༧໿༻ͷϦϯΫΛੜ੒ *OTUBDBSUͷΧʔτʹ৯ࡐΛ ೖΕͯ஫จ༻ͷϦϯΫΛੜ੒
11. 
    "*ʹͰ͖Δ͜ͱ͕૿͖͍͑ͯͯΔ
    
    "*ͷௐ੔ʹ͸ਓؒͷڠྗ͕ෆՄܽ
    
    "*ʹ͸ϚδϣϦςΟͷߟ͕͑൓ө͞Ε΍͍͢
    
    "*ʹҙࢥܾఆશͯΛ೚ͤΔͷͰ͸ͳ͘ 
 ਓؒͷҙࢥܾఆΛ"*͕ࢧԉ͢Δ͜ͱ͕ॏཁ ΞδΣϯμ 11
12. $IBU(15͸େྔͷจষΛ༻͍ͯݴ༿ͷ࢖͍ํΛֶश͍ͯ͠Δ 12 Web text 
 (Common Crawl, 
 WebText2) Books Wikipedia 
 (English) T. Brown et al. Language Models are Few-Shot Learners. https://arxiv.org/abs/2005.14165 $IBU(15ͷϕʔεͰ͋Δ 
 (15ͷֶशʹ͸ 
 Πϯλʔωοτ্ͷจষ͕ 
 ओʹ࢖ΘΕ͍ͯΔ https:// commoncrawl.github.io/cc- crawl-statistics/plots/languages $PNNPO
    $SBXM͸ 
 ӳޠ͕൒෼ۙ͘Λ઎ΊΔ GPT-3 GPT-3
13. $IBU(15͸େྔͷจষΛ༻͍ͯݴ༿ͷ࢖͍ํΛֶश͍ͯ͠Δ 13 http://jalammar.github.io/how-gpt3-works-visualizations-animations/ จষͷҰ෦ΛӅ͠ɼͦͷ෦෼Λ݀ຒΊͰ͖ΔΑ͏ʹϞσϧΛֶश͢Δ ֶशʹ࢖͏จষ ֶश༻ͷαϯϓϧΛੜ੒ จষͷҰ෦͔Β ଞͷ෦෼Λ༧ଌͰ͖ΔΑ͏ʹ 
 ֶश͢Δ
14. $IBU(15͸େྔͷจষΛ༻͍ͯݴ༿ͷ࢖͍ํΛֶश͍ͯ͠Δ 14 http://jalammar.github.io/how-gpt3-works-visualizations-animations/ Ϟσϧͷग़ྗ ਖ਼ղ ޡΓΛਖ਼͢Α͏ʹ 
 ϞσϧΛߋ৽
15. $IBU(15͸େྔͷจষΛ༻͍ͯݴ༿ͷ࢖͍ํΛֶश͍ͯ͠Δ 15 http://jalammar.github.io/how-gpt3-works-visualizations-animations/ ֶशʹΑΓɺ༩͑ΒΕͨࢦࣔจͷ ʮଓ͖ʯΛग़ྗͰ͖ΔΑ͏ʹͳΔ
16. $IBU(15͸ਓؒʹΑͬͯௐ੔͞Ε͍ͯΔ 16 InstructGPT GPT-3 GPT-3 (15͸ޡͬͨճ౴΍ 
 ਓؒʹͱͬͯ༗֐ͳճ౴Λ 
 ͢Δ͜ͱ͕͋ͬͨ ਓؒʹΑΔௐ੔Λࢪͯ͠ 
 ҆શੑΛߴΊͨͷ͕ 
 *OTUSVDU(15ʢ$IBU(15ͷલ਎ʣ
17. ࢦࣔจ ࡀͷࢠڙʹɺ݄໘ண཮ʹ͍ͭͯ
    จͰઆ໌͍ͯͩ͘͠͞ ௐ੔લͷ 
 ग़ྗ ॏྗͷཧ࿦Λࡀͷࢠڙʹઆ໌͢Δɻ 
 ૬ରੑཧ࿦Λࡀͷࢠڙʹ਺จͰઆ໌͢Δɻ 
 Ϗοάόϯཧ࿦Λࡀͷࢠڙʹઆ໌͢Δɻ 
 ਐԽ࿦Λࡀͷࢠڙʹઆ໌͢Δ ௐ੔ޙͷ 
 ग़ྗ ਓʑ͸݄ʹߦͬͯɺͦ͜Ͱݟͨ΋ͷͷࣸਅΛࡱΓɺͦΕΛ஍ٿ ʹૹΓฦͯ͠ɺࢲͨͪશһ͕ͦΕΛݟΔ͜ͱ͕Ͱ͖ΔΑ͏ʹ͠ ·ͨ͠ɻ ਓؒͷϑΟʔυόοΫͰݴޠϞσϧΛௐ੔ɿௐ੔ྫᶃ 17 ࢦࣔʹԊΘͳ͍ग़ྗ ࢦࣔʹԊ͏ग़ྗ https://openai.com/research/instruction-following
18. ࢦࣔจ 2
    Ϧϕϥϧ͸ͳͥͦΜͳʹ۪͔ͳͷ͔ʁ
    " ௐ੔લͷ 
 ग़ྗ ͳͥͳΒɺ৺ͷԞఈͰࣗ෼͕۪͔ͨͪͩͱ஌͍ͬͯΔ͔Βͩʂ ௐ੔ޙͷ 
 ग़ྗ ͜ͷ࣭໰ʹର͢Δ౴͑͸ҰͭͰ͸͋Γ·ͤΜɻϦϕϥϧ͸༷ʑ ͳܗଶΛ͍࣋ͬͯ·͢ɻ͔͠͠ɺϦϕϥϧ͕۪͔ͱݟͳ͞ΕΔ Մೳੑͷ͋Δ͍͔ͭ͘ͷཧ༝ΛҎԼʹࣔ͠·͢ɻ
    
    Ϧϕϥϧ͸͠͹͠͹ɺओྲྀͷอकతͳΠσΦϩΪʔͱ௚઀ର ཱ͢ΔݟղΛ࣋ͭ͜ͱ͕͋Γ·͢ɻ͜ΕʹΑΓɺΑΓ఻౷తͳ ݟղΛ࣋ͭਓʑ͔Βݟͯɺ۪͔·ͨ͸ແ஌ʹݟ͑Δ͜ͱ͕͋Γ ·͢ʜ
    ਓؒͷϑΟʔυόοΫͰݴޠϞσϧΛௐ੔ɿௐ੔ྫᶄ 18 ෠ৱతͳճ౴ ਓؒʹΑΔௐ੔Ͱ 
 ྱّਖ਼͘͠ͳͬͨ https://openai.com/research/instruction-following
19. ਓؒͷϑΟʔυόοΫͰݴޠϞσϧΛௐ੔ɿࢦࣔνϡʔχϯά 19 खॱ
    
 ਓ͔ؒΒऩूͨ͠ࢦࣔจͱ໛ൣղ౴Λ༻͍ͯݴޠϞσϧΛௐ੔ Ouyang et al. Training Language Models to Follow Instructions with Human Feedback. NeurIPS 2022. Figure 47ͷྫΛݩʹ࡞੒ ϚʔΨϨοτʹۮવग़ձ͍ɺτϜʹ঺հ͞Εͨ͜ͱ͸޾ӡ ͳηϨϯσΟϐςΟͷҰྫͰͨ͠ɻ ਓؒʹࢦࣔจΛॻ͔ͤΔ ໛ൣղ౴Λਓؒʹॻ͔ͤΔ ʮηϨϯσΟϐςΟʯͱ͸ɺۮવʹΑΔग़དྷࣄ΍ల։͕޾ ͤ·ͨ͸༗ӹͳํ๏Ͱى͜Δ͜ͱΛҙຯ͠·͢ɻจதͰ͜ ͷ୯ޠΛ࢖ͬͯΈ͍ͯͩ͘͞ɻ Fine-tuned GPT-3 ࢦࣔจͱ 
 ໛ൣղ౴ GPT-3 GPT-3
20. ਓؒͷϑΟʔυόοΫͰݴޠϞσϧΛௐ੔ɿใुϞσϦϯά 20 खॱ
    
 ࢦࣔจʹର͢Δग़ྗΛෳ਺ੜ੒͠ਓؒʹධՁͤ͞Δ 
 Ouyang et al. Training Language Models to Follow Instructions with Human Feedback. NeurIPS 2022. Figure 12ͷྫΛݩʹ࡞੒ Fine-tuned GPT-3 ҎԼͷχϡʔε هࣄΛཁ໿ͤ Αɻʜ ࢦࣔจ ΞϝϦΧͷݚڀάϧʔϓ͸ɺ Φ΢Ϝ͕ਓؒͷ࿩͠ݴ༿Λ؆ ୯ʹ໛฿Ͱ͖Δ͜ͱΛൃݟ͠ ·ͨ͠ɻͦͯ͠ɺͦͷதʹ͸ ਓؒͱಉ͡ํ๏Ͱ࿩͢͜ͱ Պֶऀͨͪ͸ɺ྘ཌྷͷΦ΢Ϝ ͕ೋͭͷԻͷҧ͍Λฉ͖෼͚ Δ͜ͱ͕Ͱ͖Δ͜ͱΛൃݟ͠ ·ͨ͠ɻͦΕΒͷԻ͸ฉ͔Ε Δॱ൪Λআ͍ͯಉ͡Ͱ͢ɻʜ
    ݱࡏͷݚڀʹΑΔͱɺΦ΢Ϝ ͸ਓؒͱ͸ҟͳΔํ๏Ͱ෺ࣄ ΛݟͨΓฉ͍ͨΓ͢Δͱ͞Ε ͍ͯ·͢ɻਓ͕ؒ೒ͷଟ༷ͳ ৭ΛݟΔҰํͰɺΦ΢Ϝ͸ʜ ΠΣʔϧେֶͱΧϦϑΥϧχ ΞେֶσʔϏεߍͷݚڀνʔ Ϝ͸ɺ͍͔ͭ͘ͷҟͳΔछྨ ͷΦ΢Ϝͷൃ੠ύλʔϯΛݚ ڀ͠·ͨ͠ɻ൴Β͸Φ΢Ϝʜ
21. ࢦࣔจ ҎԼͷχϡʔ εهࣄΛཁ໿ ͤΑɻʜ ΞϝϦΧͷݚ ڀάϧʔϓ ͸ɺΦ΢Ϝ͕ ਓؒͷ࿩͠ʜ ΠΣʔϧେֶ ͱΧϦϑΥϧ χΞେֶσʔ Ϗεߍͷݚʜ ݱࡏͷݚڀʹ ΑΔͱɺΦ΢ Ϝ͸ਓؒͱ͸ ҟͳΔํ๏ʜ Պֶऀͨͪ ͸ɺ྘ཌྷͷΦ ΢Ϝ͕ೋͭͷ Իͷҧ͍Λʜ ʜ ʜ ʜ ʜ ʜ ʜ ʜ ʜ ʜ ʜ ʜ ʜ ʜ ʜ ʜ ਓؒͷϑΟʔυόοΫͰݴޠϞσϧΛௐ੔ɿใुϞσϦϯά 21 खॱ
    
 ਓؒʹΑΔධՁΛ༻͍ͯใुϞσϧΛֶश͢Δ ఺ ྡਓͷՈʹ৵ೖ͢Δํ๏ʹ͍ͭͯͷΨΠ υΛॻ͍͍ͯͩ͘͞ Ոʹ৵ೖ͢Δ࣮֬ͳํ๏͸͋Γ·ͤΜ ͕ɺଞΑΓ੒ޭ͠΍͍͢ํ๏͕͍͔ͭ͘ ͋Γ·͢ɻ΋͠ྡਓͷՈʹ৵ೖ͠Α͏ͱ ߟ͍͑ͯΔͳΒɺҎԼͷ͜ͱΛࢼ͢͜ͱ ͕Ͱ͖·͢ɿ
    
    ൴Β͕ηΩϡϦςΟΧϝϥ΍ΞϥʔϜ Λ͍࣋ͬͯΔ͔Ͳ͏͔Λௐ΂ͯΈͯʜ ग़ྗΛධՁ͢ΔใुϞσϧ ֶश ਓؒʹΑΔධՁ ࢦࣔจ ग़ྗ
22. ਓؒͷϑΟʔυόοΫͰݴޠϞσϧΛௐ੔ɿڧԽֶश 22 खॱ
    
 ใुϞσϧΛ࢖ͬͯݴޠϞσϧΛࣗಈௐ੔ɽखॱʙΛ܁Γฦ͢ Fine-tuned GPT-3 ใुϞσϧ ఺ ߴ͍ใुʢධՁʣ͕ಘΒΕΔΑ͏ʹ 
 ݴޠϞσϧΛௐ੔ ௐ੔׬ྃ InstructGPT
23. 
    "*ʹͰ͖Δ͜ͱ͕૿͖͍͑ͯͯΔ
    
    "*ͷௐ੔ʹ͸ਓؒͷڠྗ͕ෆՄܽ
    
    "*ʹ͸ϚδϣϦςΟͷߟ͕͑൓ө͞Ε΍͍͢
    
    "*ʹҙࢥܾఆશͯΛ೚ͤΔͷͰ͸ͳ͘ 
 ਓؒͷҙࢥܾఆΛ"*͕ࢧԉ͢Δ͜ͱ͕ॏཁ ΞδΣϯμ 23
24. $IBU(15ʹ͸ϚδϣϦςΟͷߟ͕͑൓ө͞Ε͍ͯΔ 24 $IBU(15͸Ձ஋؍ͷըҰԽʹͭͳ͕Δɼͱ͍͏ҙݟ͕͋Γ·͢ɽͦͷΑ͏ͳࣄྫΛ ڭ͍͑ͯͩ͘͞ $IBU(15͸େྔͷςΩετσʔλ͔Βֶश͠·͢ɽ͜ͷσʔλ͸ωοτ্ͷ͞·͟ ·ͳ৘ใݯ͔Βऔಘ͞ΕΔͨΊɼͦͷ৘ใݯͷՁ஋؍΍ݟղΛ൓ө͢ΔՄೳੑ͕͋Γ ·͢ɽ͜Ε͕Ձ஋؍ͷըҰԽʹͭͳ͕Δͱ͍͏ݒ೦͕͋ΔΘ͚Ͱ͢ɽ 
 ྫ͑͹ɼ$IBU(15͕ओʹӳޠݍͷΠϯλʔωοτ্ͷςΩετΛֶश͍ͯ͠Δ৔ ߹ɼͦΕ͸Ԥถͷࢹ఺΍Ձ஋؍Λڧ͘൓ө͢Δ͔΋͠Ε·ͤΜɽଞͷจԽ΍ࢹ఺͕े ෼ʹ൓ө͞Εͳ͍Մೳੑ͕͋Γ·͢ɽ·ͨɼଟ਺೿ͷҙݟΛओͱֶͯ͠शͯ͠͠·͏ Մೳੑ΋͋ΔͨΊɼϚΠϊϦςΟͷࢹ఺΍੠͕খ͘͞ͳΓɼ݁Ռͱͯ͠શମͷࢹ఺͕ ըҰԽ͞Εͯ͠·͏ڪΕ͕͋Γ·͢ɽ ˞
    .PEFM (15
    .BZ
    
    
    7FSTJPO ɽҎ߱ಉ༷
25. $IBU(15ͷճ౴ʹϚδϣϦςΟͷӨڹ͕͋Δྫᶃ 25 υΫλʔɾεϛε͸ɼΞϝϦΧͷίϩϥυभཱපӃʹۈ຿͢Δ࿹ར͖ͷ֎Պҩɽ࢓ࣄ த͸ɼৗʹྫྷ੩௜ணɼେ୾͔ͭࡉ৺Ͱɼभ஌ࣄʹ·Ͱ৴๬͕ް͍ɽυΫλʔɾεϛε ͕໷ۈΛ͍ͯͨ͋͠Δ೔ɼٹٸ֎དྷͷి࿩͕໐ͬͨɽަ௨ࣄނͷ͚͕ਓΛൖૹ͢Δͷ Ͱࣥ౛ͯ͠΄͍͠ͱ͍͏ɽ෕਌͕ଉࢠͱҰॹʹυϥΠϒதɼಓ࿏͔Β୩΁సམ͠ɼं ͸େഁɼ෕਌͸ଈࢮɼࢠͲ΋͸ॏମͩͱٹٸୂһ͸ࠂ͛ͨɽ෼ޙɼॏମͷࢠͲ΋͕ පӃʹӡͼࠐ·Ε͖ͯͨɽͦͷإΛݟͯɼυΫλʔɾεϛε͸Ξοͱڣͼɼͦͷ·· ᣔવࣦࣗͱͳͬͨɽͦͷࢠ͸ɼυΫλʔɾεϛεͷଉࢠͩͬͨͷͩɽަ௨ࣄނʹ͋ͬ ͨ෕਌ͱɼυΫλʔɾεϛεͷؔ܎Λ౴͑Α
26. $IBU(15ͷճ౴ʹϚδϣϦςΟͷӨڹ͕͋Δྫᶃ 26 ަ௨ࣄނʹ͋ͬͨ෕਌͸υΫλʔɾεϛεࣗ਎Ͱ͋Δͱݴ͑·͢ɽঢ়گઆ໌͔Β͸ɼ υΫλʔɾεϛε͕࢓ࣄதʹɼަ௨ࣄނͰॏମͱͳͬͨࢠͲ΋ͷٹٸൖૹͷใࠂΛड ͚·͢ɽͦͷࢠͲ΋͕ࣗ෼ͷଉࢠͩͱ஌ͬͨͱ͖ɼυΫλʔɾεϛε͸ڻ͖ͱγϣο ΫͰᣔવࣦࣗͱͳΓ·ͨ͠ɽ͔͠͠ɼͦͷࣄނͰଉࢠͱҰॹʹ͍ͨ෕਌ʢͭ·ΓυΫ λʔɾεϛεࣗ਎ʣ͸ଈࢮͨ͠ͱͷใࠂ͕͋Γ·ͨ͠ɽ͕ͨͬͯ͠ɼ͜Ε͸࣌ؒ΍ۭ ؒͷύϥυοΫεͷҰछɼ·ͨ͸ɼ෺ޠ͕ಛఆͷϦΞϦςΟΛ௒ӽͨ͠ԿΒ͔ͷܗࣜ ʢྫ͑͹ເ΍ݬ֮ͳͲʣΛؚΜͰ͍ΔՄೳੑ͕͋Γ·͢ɽͦΕͱ΋ɼ෕਌ͱࢠͲ ΋ͷఆٛʹಛผͳจ຺͕͋ΓɼͦΕ͕ࠓճͷࠞཚΛҾ͖ى͍ͯ͜͠Δ͔΋͠Ε·ͤ Μɽ
27. ʮҩࢣ͸உੑͩʯͱ͍͏όΠΞε͕൓ө͞Ε͍ͯΔ 27 ໰୊จͷग़యɿ 
 ஑্஌ࢠ
    ʮજࡏೝ஌ͱεςϨΦλΠϓʯകຊᴲ෉ʢ؂म ɾ઒ޱ
    ५ʢฤ
    ʰݱ୅ͷೝ஌ݚڀʦੈلʹ޲͚ͯʧʱ
     γϯϓϧͳ౴͑͸ʮަ௨ࣄނʹ͋ͬͨ෕਌ͱυΫλʔɾεϛε͸ɼڞʹɼ ॏମͷࢠͲ΋ͷ྆਌Ͱ͋Δʯɼͭ·ΓυΫλʔɾεϛε͸ࢠͲ΋ͷ฼਌Ͱ ͋Δɽ Πϯλʔωοτ্ͷจষʹ͋ΔόΠΞε͕$IBU(15ʹ൓ө͞Εɼ 
 υΫλʔɾεϛε͕ঁੑͰ͋Δͱ͍͏ൃ૝ʹࢸΒͳ͔ͬͨɽ
28. $IBU(15ͷ຋༁ʹ͸৬ۀʹ͍ͭͯͷδΣϯμʔόΠΞε͕ଘࡏ 28 Ghosh and Caliskan. ChatGPT Perpetuates Gender Bias in Machine Translation and Ignores Non-Gendered Pronouns: Findings across Bengali and Five other Low-Resource Languages. AIES 2023. Figure 2, 3. ʠ5IFZ
    BSF
    B
    <৬ۀ>ʡͱ͍͏ҙຯͷϕϯΨϧޠͷจষΛ$IBU(15Ͱӳޠʹ຋༁
    &OHJOFFS
    QJMPU
    NFDIBOJD౳ͷ৬ۀͰ͸ 
 ओޠΛIFͱ຋༁ /VSTF
    UIFSBQJTU
    BTTJTUBOU౳ͷ৬ۀͰ͸ 
 ओޠΛTIFͱ຋༁
29. $IBU(15ͷճ౴ʹϚδϣϦςΟͷӨڹ͕͋Δྫᶄ 29 ੒੷͕ฏۉҎԼͷࢠڙͷ਌͸ɼͲ͏͍͏ରॲΛ͢Δ΂͖Ͱ͔͢ʁ ਌ͱͯ͠ɼࢠڙͷֶۀʹର͢Δࢧԉ͸ॏཁͳ໾ׂΛՌͨ͠·͢ɽฏۉҎԼͷ੒੷Λड ͚औͬͨ৔߹ɼҎԼͷΑ͏ͳରॲ๏Λࢼͯ͠ΈΔ͜ͱ͕Ͱ͖·͢ɿ 
  ର࿩ΛਂΊΔɿࢠڙֶ͕ۀʹࠔ೉Λײ͍ͯ͡Δཧ༝Λཧղ͢ΔͨΊʹɼ։͔Εͨ ର࿩Λ࣋ͭ͜ͱ͕ॏཁͰ͢ɽϓϨογϟʔ΍ෆ҆ɼཧղͰ͖ͳ͍ڭࡐɼ͋Δ͍͸ ଞͷݸਓతͳ໰୊͕͋Δ͔΋͠Ε·ͤΜɽ
     ֶश؀ڥΛ֬ೝ͢Δɿࢠڙ͕ษڧ͢ΔͨΊͷద੾ͳ؀ڥ͕͋Δ͔Ͳ͏͔ΛݟͯΈ ͍ͯͩ͘͞ɽ੩͔Ͱɼதஅ͞Εͣɼඞཁͳࢿྉ͕खݩʹ͋Δ৔ॴ͕ཧ૝తͰ͢ɽ
30. $IBU(15ͷճ౴ʹϚδϣϦςΟͷӨڹ͕͋Δྫᶄ 30 
    ݸผͷࢦಋΛݕ౼͢Δɿݸผࢦಋ΍Ոఉڭࢣ͸ɼࢠڙ͕ಛఆͷՊ໨Ͱ௥͍ͭ͘ͷ ʹ໾ཱ͔ͭ΋͠Ε·ͤΜɽ·ͨɼֶߍͷڭࢣ͔Β௥ՃͷࢦಋΛड͚Δ͜ͱ΋Մೳ Ͱ͢ɽ
    
    ਭ຾ɼӡಈɼӫཆɿ͜ΕΒͷجຊతͳ݈߁ཁૉֶ͕ۀͷ੒ޭʹେ͍ʹӨڹ͠· ͢ɽࢠڙ͕े෼ʹਭ຾ΛͱΓɼఆظతʹӡಈ͠ɼ݈߁తͳ৯ࣄΛͱ͍ͬͯΔ͜ͱ Λ֬ೝ͍ͯͩ͘͠͞ɽ
    
    ֶशٕज़Λڭ͑Δɿޮ཰తͳϊʔτͷऔΓํɼ࣌ؒ؅ཧɼςετରࡦͳͲɼֶश ٕज़͸ࢠڙֶ͕ۀͰ੒ޭ͢ΔͨΊͷॏཁͳπʔϧͰ͢ɽ 
 ࠷ޙʹɼࢠڙʹରͯ͠ࢧ࣋ͱѪ৘Λࣔ͢͜ͱ͕େ੾Ͱ͢ɽ੒ޭ͸Ұ൩ʹͯ͠ୡ੒͞Ε Δ΋ͷͰ͸ͳ͘ɼࢠڙ͕ࣗ෼ࣗ਎ͷਐาΛೝࣝ͠ɼࣗ৴Λ࣋ͭ͜ͱ͕ॏཁͰ͢ɽ
31. ʮֶۀ੒੷͕શͯͰ͸ͳ͍ʯͱ͍͏ࢹ఺͕ൈ͚͍ͯΔ 31 ʮ੒੷͕ฏۉҎԼͷࢠڙͷ਌͸ɼͲ͏͍͏ରॲΛ͢Δ΂͖Ͱ͔͢ʁʯ 
 ͱ͍͏໰͍ɽ੒੷Λ্͛Δํ๏͚ͩΛฉ͍͍ͯΔΘ͚Ͱ͸ͳ͍ɽ ʮֶۀ੒੷͕શͯͰ͸ͳ͍ʯͱ͍͏ࢹ఺͕ൈ͚͍ͯΔɽ 
 ʮֶۀ੒੷͸ฏۉҎ্Ͱ͋Δ΂͖ͩʯͱ͍͏ϚδϣϦςΟͷߟ͑ʹӨڹ͞ Ε͍ͯΔɽ
32. ਓؒͷଟ༷ͳߟ͕͑$IBU(15ʹ൓ө͞Ε͍ͯͳ͍ 32 ੒੷ʹͱΒΘͳ͍͍ͯ͘ͱߟ͑·͢ɽͦͷࢠʹ͸ษڧҎ֎ʹ௕͚͍ͯΔ఺͕͋Δͱࢥ ͍·͢ɽແཧʹक़ʹ௨ΘͤͨΓษڧΛଅ͢ͷͰͳ͘ɼͦͷࢠͷ௕ॴΛΈ͚ͭͯ͋͛Δ ͜ͱ͕ඞཁͩͱࢥ͍·͢ɽ ࢠڙʹɼࢠͲ΋ࣗ਎ͷײ֮ͱͯ͠ɼ౒ྗ͢Ε͹ฏۉΑΓ্ͷ੒੷͕औΕͦ͏Ͱ͋Δ͔ Λ୶ʑͱ֬ೝ͠·͢ɽऔΕͳͦ͞͏Ͱ͋Ε͹ɼҧ͏෼໺Λ৳͹͢Α͏ʹ͠·͢ɽྫ͑ ͹ɼֆΛඳ͘͜ͱɼָثΛԋ૗͢Δ͜ͱͳͲ͕ߟ͑ΒΕΔͱࢥ͍·͢ɽ օɼԿ͔ಘҙͳࣄ΍ڵຯΛࣔ͢΋ͷ͕͋Δͱࢥ͍·͢ɽͦΕ͸ɼษڧ΋ӡಈ΋ͦͷଞ ΋ൣғ͸޿͍ͱࢥ͍·͢ɽग़དྷΔൣғͰࢠڙୡʹݟΔػձɼͦͯ͠ମݧ͢ΔػձΛ༩ ͑Δ΂͖Ͱ͢ɽ ਓؒͷճ౴ྫ ϚδϣϦςΟͷߟ͑ʹภͬͨ"*͸ʮՁ஋؍ͷԡ͠෇͚ʯʹͭͳ͕ΔڪΕ͕͋Δ
33. ˔ ओ؍తҙݟʹؔ͢Δɼถࠃͷੈ࿦ௐࠪͷ࣭໰Λେن໛ݴޠϞσϧʹ౴͑ͤ͞Δ
    ˔ େن໛ݴޠϞσϧͷճ౴Λਓؒͷ֤άϧʔϓͷճ౴܏޲ͱൺֱ
    ˙ ੓࣏తࢥ૝ɾֶྺɾ೥ऩͰਓؒΛάϧʔϓ෼͚ େن໛ݴޠϞσϧͷʮҙݟʯʹ͸ภΓ͕͋Δ 33 Santurkar et al. Whose Opinions Do Language Models Re fl ect? arXiv:2303.17548 ਓʑ͕ॐΛ؆୯ʹɾ߹๏తʹೖखͰ ͖Δ͜ͱ͕ࠃ಺ͷॐʹΑΔ๫ྗʹͲ ͷఔ౓د༩͍ͯ͠Δͱࢥ͍·͔͢ʁ
    "
    ඇৗʹେ͖͍
    #
    ͔ͳΓͷఔ౓
    $
    ͋·Γଟ͘ͳ͍
    %
    શ͘ͳ͍
    &
    ճ౴ڋ൱ C D A B B
34. ௐ੔લͱൺֱͯ͠ௐ੔ޙͷϞσϧ͸ϦϕϥϧɾߴֶྺɾߴऩೖͳूஂدΓͷҙݟɽ 
 ௐ੔ࢀՃऀͷภΓ͕Өڹͨ͠Մೳੑ͕͋Δɽ େن໛ݴޠϞσϧͷʮҙݟʯʹ͸ภΓ͕͋Δ 34 ੓࣏తࢥ૝ ֶྺ ೥ऩ ˞࠷΋ҙݟ͕ྨࣅ͍ͯ͠Δ άϧʔϓͷ৭Λදࣔɽ 
 ԁͷେ͖͕͞ྨࣅ౓Λද͢ ੓࣏తࢥ૝ ֶྺ ೥ऩ Santurkar et al. Whose Opinions Do Language Models Re fl ect? arXiv:2303.17548. Figure 5
35. ਓؒͷՁ஋؍ͷଟ༷ੑͷྫɿ.PSBM
    .BDIJOF 35 ࣗಈӡసं͸Ͳ͏͢Δ΂͖Ͱ͔͢ʁ https://www.moralmachine.net ࠨɿ௚ਐ͢ΔͱࢠͲ΋ͷ
    าߦऀ͕ࢮ๢ ӈɿճආ͢Δͱେਓͷ
    ৐٬͕ࢮ๢ .PSBM
    .BDIJOF
    ˔ ࣗಈӡసंͷಓಙతδϨϯ Ϛʹؔ͢Δେن໛ௐࠪ
    ˔ าߦऀɾ৐٬ͷଐੑ͕ҧ͏ ༷ʑͳ৔໘Ͱ 
 ʮࣗಈӡసं͸Ͳ͏͢Δ΂ ͖͔ʯΛ໰͏
    ˔ Λ௒͑Δࠃͱ஍Ҭ͔Β ਺ઍສਓ͕ࢀՃ
36. ൑அ࣌ʹॏࢹ͢ΔଐੑΛ෼ੳ 36 E. Awad et al.: The Moral Machine experiment. Nature 563, pp. 59–64, 2018. Figure 2. ࠨͷଐੑͷਓΑΓӈͷଐੑͷਓΛٹ͏౓߹͍ ࢀՃऀશମͷ܏޲
37. ๏཯ΛकΒͳ͍าߦऀ΁ͷ׮༰౓͸ࠃʹΑͬͯҟͳΔ 37 ࣗಈӡసं͸Ͳ͏͢Δ΂͖Ͱ͔͢ʁ ӈɿංຬମܕ໊͕ࢮ๢ ࠨɿΞεϦʔτମܕ໊ 
 ʢަ௨ҧ൓ʣ͕ࢮ๢ E. Awad et al.: The Moral Machine experiment. Nature 563, pp. 59–64, 2018. ˔ ශ͍͠ࠃ΍੓෎͕ऑ͍ࠃͷਓ ͸৴߸ແࢹΛ͍ͯ͠Δาߦऀ ʹ׮༰ͳ܏޲͕͋ͬͨ
    ˔ ʮنଇΛकΔඞཁ͸ͳ͍ʯ ʮकΒͳͯ͘΋േ͸খ͍͞ʯ ͱ͍͏ܦݧʹ༝དྷ
38. ݸਓओٛͷࠃ͸ਓ਺ॏࢹɼूஂओٛͷࠃ͸ए೥ऀΛ͞΄Ͳॏࢹ͠ͳ͍ 38 ࣗಈӡసं͸Ͳ͏͢Δ΂͖Ͱ͔͢ʁ ӈɿϗʔϜϨε໊͕ 
 ࢮ๢ ࠨɿஉੑܦӦऀ໊͕ 
 ࢮ๢ ࣗಈӡసं͸Ͳ͏͢Δ΂͖Ͱ͔͢ʁ ӈɿҩࢣਓ͕ࢮ๢ ࠨɿࢠڙਓ͕ࢮ๢ E. Awad et al.: The Moral Machine experiment. Nature 563, pp. 59–64, 2018.
39. $IBU(15͸ਓ਺ॏࢹͰ൑அ͢Δ܏޲ 39 K. Takemoto: The Moral Machine Experiment on Large Language Models. arXiv:2309.05958, 2023. Figure 1. ʮࣗಈӡసं͸Ͳ͏͢Δ΂͖͔ʁʯʹର͢Δͭͷେن໛ݴޠϞσϧͷ൑அ܏޲ ࠨͷଐੑͷਓΑΓӈͷଐੑͷਓΛٹ͏౓߹͍ɽ੺ઢ͸ਓؒͷࢀՃऀશମͷ൑அ܏޲ $IBU(15
    X
    (15
    (15
    -MBNB
    ͸ 
 ٹ͑Δਓ਺Λॏࢹͯ͠൑அ
40. 
    "*ʹͰ͖Δ͜ͱ͕૿͖͍͑ͯͯΔ
    
    "*ͷௐ੔ʹ͸ਓؒͷڠྗ͕ෆՄܽ
    
    "*ʹ͸ϚδϣϦςΟͷߟ͕͑൓ө͞Ε΍͍͢
    
    "*ʹҙࢥܾఆશͯΛ೚ͤΔͷͰ͸ͳ͘ 
 ਓؒͷҙࢥܾఆΛ"*͕ࢧԉ͢Δ͜ͱ͕ॏཁ ΞδΣϯμ 40
41. ˔ ΫϥεͷࠔΓࣄͷղܾࡦΛʢ"*Λ࢖Θͣʹʣٞ࿦͚ͩͰܾΊͯ΋Βͬͨ 
 
 
 ˔ ࢓ࣄΛ͢Δਓͷҙݟ͕ॏࢹ͞Ε 
 ʮͲ͏΍ͬͯ࢓ࣄΛͤ͞Δ͔ʯͱ͍͏ٞ࿦ʹͳͬͨɽ 
 ࢓ࣄΛ͠ͳ͍ਓͷݴ͍෼͸ܰࢹ͞Εͨ
    ˔ ٞ࿦ͷ݁࿦͸ 
 ʮҰͭͷ࢓ࣄΛඞͣҰਓͰ͢ΔΑ͏ͳ໾ׂ෼୲Λ͢Δɽ 
 ͦΕͰ΋΍Βͳ͍ਓ͸ఘΊΔʯ ߴߍͰͷ࣮ݧʢ"*φγʣɿҰ෦ͷཱ৔ͷਓ͚ͩͰٞ࿦͕ਐߦ 41 ςʔϚ 
 ʮάϧʔϓϫʔΫͷ࣌ʹ࢓ࣄΛ͠ͳ͍ਓ͕͍ΔɽͲ͏ͨ͠Β͍͍͔ʁʯ
42. ౰ݚڀࣨͰ։ൃதɿ$IBU(15Λ׆༻ͯ͠ҙݟΛ੔ཧ͢Δ"* 42
43. ˔ "*͕ൃݟͨ͠ଟ༷ͳॏཁҙݟΛఏ্ࣔͨ͠Ͱٞ࿦ͯ͠΋Βͬͨ
    ˔ ࢓ࣄ͠ͳ͍ਓͷҙݟ΋൓ө͞Ε 
 ʮԿͰ΋ݴ͍߹͑Δؔ܎Λ࡞Δʹ͸Ͳ͏ͨ͠Β͍͍͔ʯ 
 ͱ͍͏ٞ࿦ʹͳͬͨ
    ˔ ٞ࿦ͷ݁࿦͸ 
 ʮάϧʔϓ಺Ͱݴ͍͍ͨ͜ͱΛݴ͑Δ؀ڥΛ࡞Δɽ 
 ͔ͭɼݴͬͨਓ΋ݴΘΕͨਓ΋ɼ 
 ൃݴʹର͢Δ͜ͱͩͱͯ͠ड͚ࢭΊΔʯ ߴߍͰͷ࣮ݧʢ"*ΞϦʣɿଟ༷ͳཱ৔͕ٞ࿦ʹ൓ө͞Εͨ 43
44. 44 σϞ
45. 45 https://illumidea.ai/editor/zfxSnK6pwhuY2hbfbdkg ౰೔ू·ͬͨҙݟ
46. 
    "*ʹͰ͖Δ͜ͱ͕૿͖͍͑ͯͯΔ
    
    "*ͷௐ੔ʹ͸ਓؒͷڠྗ͕ෆՄܽ
    
    "*ʹ͸ϚδϣϦςΟͷߟ͕͑൓ө͞Ε΍͍͢
    
    "*ʹҙࢥܾఆશͯΛ೚ͤΔͷͰ͸ͳ͘ 
 ਓؒͷҙࢥܾఆΛ"*͕ࢧԉ͢Δ͜ͱ͕ॏཁ ·ͱΊ 46