Structure of NLP Papers

Slide 1

Slide 1 text

4USVDUVSFPG/-11BQFST 6LZP)POEB /"*45/-1-BC .BZ

Slide 2

Slide 2 text

1 0WFSWJFX • 5JQTPOIPXUPTUSVDUVSFZPVSQBQFS • 4USVDUVSFPGTFDUJPOT • 4USVDUVSFJOTJEFPGFBDITFDUJPO • .BOZUIBOLTUPNZDPBVUIPST NFOUPST BOEUIJTBXFTPNFUVUPSJBM • IUUQXXXQIPOUSPODPNTMJEFTOFVCJHOMQUVUPSJBMQEG

Slide 3

Slide 3 text

2 $BWFBUT • 5IJTUVUPSJBMBTTVNFTUIBUZPVSQBQFSTBSF • *O/-1PSPUIFSDMPTFEPNBJOTTVDIBT$7BOE.- • /PUTVSWFZUIFNFEFNPQBQFST • 5IJTUVUPSJBMEPFTOPUDPWFSBMMTFDUJPOT • 4FFUIFMJOLCFMPXGPSUIFOFXNBOEBUPSZTFDUJPOT • -JNJUBUJPOT &UIJDBM 3JTLT FUD • IUUQTBDMSPMMJOHSFWJFXPSHSFTQPOTJCMF/-1SFTFBSDI • 5IFUJQTBSFCBTFEPONZQFSTPOBMQSBDUJDFBOEQSFGFSFODF • $PNNFOUTBOETVHHFTUJPOTBSFXFMDPNF

Slide 4

Slide 4 text

8IBUEPQBQFSTMPPLMJLF

Slide 5

Slide 5 text

4 3PVHITLFUDI 8IBUEPQBQFSTMPPLMJLF Abstract Introduction Methods Related Work Experiments Conclusion References !Consists of many different sections !Each section seems to have different roles and contents "‍‍What should we write in each section? "‍‍How to make sure the consistency?

Slide 6

Slide 6 text

5 'JOFSHSBJOFETLFUDI 8IBUEPQBQFSTMPPLMJLF Methods Related Work Experiments References !Papers are intentionally redundant for better understanding !Repeat of the basic arguments in different resolution "Just focus on structuring the basic arguments! 1. Background 2. Methods 3. Experiments Methods Experiments Background Methods Experiments Background Background Methods Experiments

Slide 7

Slide 7 text

6 #BTJDBSHVNFOUT #BDLHSPVOE Related Work Experiments References Methods Experiments Background Methods Experiments Background Background Methods Experiments Methods

Slide 8

Slide 8 text

7 #BTJDBSHVNFOUT #BDLHSPVOE • *OUSPEVDFTUIFQSPCMFNTZPVBEESFTTJOUIJTQBQFS (PBMTBOETFUUJOHTPGUIFUBTL #SJFGTLFUDIPGQSFWJPVTBQQSPBDIFT • %FUBJMFEEFTDSJQUJPOHPFTUP3FMBUFE8PSL 3FNBJOJOHQSPCMFNT UPBEESFTTJOUIJTQBQFS v Describe in this order

Slide 9

Slide 9 text

8 #BTJDBSHVNFOUT .FUIPET Related Work Experiments References Methods Experiments Background Methods Experiments Background Methods Background Methods Experiments

Slide 10

Slide 10 text

9 #BTJDBSHVNFOUT .FUIPET • %FTDSJCFTIPXUPTPMWFUIFQSPCMFNT *OUVJUJWFFYQMBOBUJPO • )PXZPVSNFUIPETTPMWFUIFQSPCMFNT %FUBJMTPGZPVSNFUIPET v Describe in this order

Slide 11

Slide 11 text

10 #BTJDBSHVNFOUT &YQFSJNFOUT Related Work Experiments References Methods Experiments Background Methods Experiments Background Methods Background Methods Experiments

Slide 12

Slide 12 text

11 #BTJDBSHVNFOUT &YQFSJNFOUT • %FNPOTUSBUFT UIBUZPVSNFUIPETTPMWFUIFQSPCMFNT PSOPU q &YQFSJNFOUBMTFUUJOHT q 0WFSBMMDPNQBSJTPOXJUICBTFMJOFTBOEQSFWJPVTNPEFMT • 5PWFSJGZUIBUZPVSNFUIPETBDIJFWFUIFHPBM CFUUFS PSOPU q "CMBUJPOTUVEJFT • 5PTFFXIJDIQBSUPGZPVSNFUIPETDPOUSJCVUFT PSOPU q 'PDVTFEBOBMZTJT • 5PGVSUIFSJOWFTUJHBUFIPX ZPVSNFUIPETBDIJFWFUIFHPBMCFUUFS PSOPU JF XIFUIFSZPVSNFUIPETTPMWFUIFQSPCMFN ZPVQPJOUFEPVU q FUD

Slide 13

Slide 13 text

12 4VNNBSZTPGBS • 1BQFSTSFQFBUFEMZFYQMBJO (PBMTBOEQSPCMFNT #BDLHSPVOE 4PMVUJPOT .FUIPET %FNPOTUSBUJPO &YQFSJNFOUT • 5IJTSFQFBUFOTVSFT • $MFBSFYQMBOBUJPOPGZPVSBSHVNFOUT • $POTJTUFODZUISPVHIUIFFOUJSFQBQFS

Slide 14

Slide 14 text

-FUTTFFUIFTUSVDUVSFPGBOBDUVBMQBQFS

Slide 15

Slide 15 text

14 &YBNQMF • Hereafter we use this paper as the example • https://arxiv.org/abs/2202.06417

Slide 16

Slide 16 text

15 &YBNQMF 0VUMJOF Abstract Introduction Methods Related Work Experiments Conclusion References Background • Slightly different in the colored sections • But basically the same as the structure we have seen above

Slide 17

Slide 17 text

16 3PVHITLFUDI "CTUSBDU • Background • Methods • Experiments !Explanation in order

Slide 18

Slide 18 text

17 'JOFSHSBJOFETLFUDI "CTUSBDU ! Explanation in order 1. Task and goal • Text generation for app. 2. Prev. approaches • Stochastic sampling • Modified objective function 3. Remaining problems • Lack of coherence • Anisotropic dist. of rep. • Solutions • Calibrate the rep. space • Diversity-aware decoding • Demonstration • Outperform SOTA 1. 2. 3.

Slide 19

Slide 19 text

18 &YBNQMF *OUSPEVDUJPO !5IFTBNFTUSVDUVSF BTUIF"CTUSBDU • #VUXJUINPSFEFUBJMT • *OIJHIFSSFTPMVUJPO • Background • Methods • Experiments

Slide 20

Slide 20 text

19 &YBNQMF #BDLHSPVOE • &YUFOEFE #BDLHSPVOE • &MBCPSBUJPOPOUIFUBTLTFUUJOHT • 4PNFUJNFT#BDLHSPVOEJTFYUFOEFEBOEDVUPVUGPSFMBCPSBUJPO • 5PJOUSPEVDFGPSNBMUBTLTFUUJOHT • 5PFYQMBJOQSFWJPVTMZVOLOPXOQSPCMFNT • FUD Methods Related Work Experiments References Methods Experiments Background Methods Experiments Background Background Methods Experiments Background

Slide 21

Slide 21 text

20 &YBNQMF .FUIPEPMPHZ ! Explanation in order 1. Intuitive explanation on how your methods solve the problem 2. Then describe the details 1. 2. 1. 2.

Slide 22

Slide 22 text

21 &YBNQMF &YQFSJNFOUT • %PDVNFOU(FOFSBUJPO • FYQFSJNFOUBMTFUUJOHT • 3FTVMUT • )VNBO&WBMVBUJPO • 0QFO%PNBJO%JBMPHVF(FOFSBUJPO • FYQFSJNFOUBMTFUUJOHT • SFTVMUT • 'VSUIFS"OBMZTJT q Experimental settings q Overall comparison q Ablation study q Experimental settings q Overall comparison q Ablation study q Focused analysis

Slide 23

Slide 23 text

22 0WFSBMMDPNQBSJTPO &YQFSJNFOUT q Overall comparison • Demonstrates that the proposed method achieves the goal better than the SOTA methods

Slide 24

Slide 24 text

23 q Ablation study • Two proposed methods • Each of them works well and works the best in combination "CMBUJPOTUVEZ &YQFSJNFOUT

Slide 25

Slide 25 text

24 'PDVTFEBOBMZTJT &YQFSJNFOUT q Focused analysis • Proposed methods improve the overall performance, but how? ü Do they work as you hypothesized? ü Do they solve the problem you pointed out? Ø Confirming here that the methods mitigate the anisotropic distribution of token representations

Slide 26

Slide 26 text

25 'PDVTFEBOBMZTJT &YQFSJNFOUT q Focused analysis • Qualitative analysis

Slide 27

Slide 27 text

26 &YBNQMF $PODMVTJPO !4BNFBT"CTU*OUSP CVU • -FTTGPDVTPO#BDLHSPVOE • 1SPCMFNPOMZ • .PSFGPDVTPO.FUIPETBOE&YQ • .PSFDPODJTFUIBO"CTU • +VTUUPTVNNBSJ[FSFFNQIBTJ[F ZPVSDPOUSJCVUJPOT • 'VUVSF8PSL

Slide 28

Slide 28 text

27 &YBNQMF 3FGFSFODFT • 6TFBVOJGJFEGPSNBU ! *O"$- *O/""$- " *O"$- *O1SPDFFEJOHTPGUIFUI"OOVBM.FFUJOHPGUIF"TTPDJBUJPOGPS $PNQVUBUJPOBM-JOHVJTUJDT *O1SPDFFEJOHTPGUIF$POGFSFODFPGUIF/PSUI"NFSJDBO$IBQUFS PGUIF"TTPDJBUJPOGPS$PNQVUBUJPOBM-JOHVJTUJDT • IUUQTZPTIJUBLBVTIJLVOFUIPX@UP@XSJUF@SFGFSFODFTIUNM

Slide 29

Slide 29 text

28 &YBNQMF "QQFOEJY3FMBUFE8PSL • #SJFGEFTDSJQUJPOPGQSFWJPVTBQQSPBDIFTBOEEJGGFSFODF GSPNZPVST • 3FMBUFE8PSLPGUFODPNFTSJHIUBGUFS*OUSPEVDUJPO • 4PVOETGBJSBTJUJTBQBSUPG#BDLHSPVOE • #VUQFSTPOBMMZ*EPOPUFNQMPZUIBUTUSVDUVSF • 1SFWJPVTBQQSPBDIFTBSFOPUZPVSNBJOQPJOUTDPOUSJCVUJPOT • 3FMBUFE8PSLUFOETUPCFUPPMPOH Ø+VTUHPPOUP.FUIPETGJSTUBOEHFUCBDLCFGPSF$PODMVTJPO v3FMBUFE8PSLJOBQQFOEJYIFSFJTBOFYUSFNFDBTFBOEOPUSFDPNNFOEFE

Slide 30

Slide 30 text

"OZRVFTUJPOTTPGBS

Slide 31

Slide 31 text

30 5JQT 8SJUF"CTUSBDUGJSTU • "CTUSBDU≈ 0VUMJOF • %POUCFXPSSJFEJGUIJTTUFQUBLFTUPPMPOH • 3FNFNCFSUIBUQBQFSTDPOTJTUPG"CTUSBDUJOEJGGFSFOUSFTPMVUJPOT • *GZPVBSFEPOFXJUIUIJT ZPVBSFBMNPTUUIFSF • 4IBSFZPVSESBGUPG"CTUSBDUXJUIZPVSDPBVUIPSTBOESFWJTF • 3FWJTFJUCBTFEPOGFFECBDLCFGPSFHPJOHPOUPUIFPUIFSTFDUJPOT • %POUIBWFUPCFTFOTJUJWFBCPVUMFOHUIBUUIFESBGUTUBHF

Slide 32

Slide 32 text

31 5JQT 8SJUF"CTUSBDUCFGPSFTUBSUJOHSFTFBSDI • +VTUBTIPSUNFNPJTGJOF • *UIFMQTZPVUP 4NPPUIMZNPWFPOUPXSJUJOHQBQFST ,FFQJONJOEZPVSIZQPUIFTJTEVSJOHSFTFBSDI ≈ ,FFQJONJOEXIBUUPEPOFYU • 8IBUJTUIFHPBMBOEXIBUJTUIFSFNBJOJOHQSPCMFN #BDLHSPVOE • 8IBUJTUIFQPTTJCMFTPMVUJPO UPUIFQSPCMFN .FUIPET • )PXTIPVMEZPVEFNPOTUSBUF UIBUZPVSNFUIPETPMWFTUIFQSPCMFN &YQFSJNFOUT 3PVHIMZKVEHFXIFUIFSZPVSSFTFBSDIJTJOUFSFTUJOHPSOPUCFGPSFXBTUJOH UJNFPOJNQMFNFOUBUJPOFYQFSJNFOUT

Slide 33

Slide 33 text

32 5JQT %POPUDPOGVTFFOETXJUINFBOT ! "DPNNPONJTUBLFJOGJOEJOHUIFQSPCMFNTUPTPMWF ! :PVSNFUIPESFTPVSDFFUDJTUIFNFBOT PGTPMWJOHQSPCMFNT • 5IFNFUIPESFTPVSDFFUDJUTFMGJTOPUUIFFOE Ø/PUVTJOH9JTOPUBQSPCMFNUPTPMWF " 9IBTOPUCFFOVTFEJOUIFQSFWJPVTXPSL TPXFVTF9 FH $POUSBTUJWFMFBSOJOHIBTOPUCFFOVTFEJOUFYUHFOFSBUJPO TPXFVTFJU # 9DBOTPMWFUIJTQSPCMFNMFGUCZUIFQSFWJPVTXPSL TPXFVTF9 FH $POUSBTUJWFMFBSOJOHDBOTPMWFUIFBOJTPUSPQJDEJTUPGSFQ TPXFVTFJU

Slide 34

Slide 34 text

33 8IZEPFTJUNBUUFS %POPUDPOGVTFFOETXJUINFBOT X Current Performance ! 9IBTOPUCFFOVTFEJOUIFQSFWJPVTXPSL TPXFVTF9 X Better Performance Apply X ü Overall performance ü Ablation study ✘ Focused analysis • As you don't have any guess on how/why X works here • "X does something awesome!" • You don't have any hypotheses to test regarding how/why X works ! Shallow contribution

Slide 35

Slide 35 text

34 8IZEPFTJUNBUUFS %POPUDPOGVTFFOETXJUINFBOT Problem X Current Performance ! 9DBOTPMWFUIJTQSPCMFNMFGUCZUIFQSFWJPVTXPSL TPXFVTF9 Problem X Better Performance Apply X ü Overall performance ü Ablation study ü Focused analysis • You can check how X solves the problem based on hypotheses • That gives the explanation of why X works here ! Deeper insight into the limitation/generalization of X ! More critical contribution

Slide 36

Slide 36 text

35 5BLFIPNFNFTTBHFT! ü 1BQFSTSFQFBUFEMZFYQMBJO (PBMTBOEQSPCMFNT #BDLHSPVOE 4PMVUJPOT .FUIPET %FNPOTUSBUJPO &YQFSJNFOUT ü 8SJUF"CTUSBDU JTI CFGPSFBOZUIJOH ü %POPUDPOGVTFFOETXJUINFBOT