{ "table_of_contents": [ { "title": "A Survey on Generative Diffusion Models", "heading_level": null, "page_id": 0, "polygon": [ [ 86.54435670375824, 55.77545166015625 ], [ 525.268346786499, 55.77545166015625 ], [ 525.268346786499, 79.77545166015625 ], [ 86.54435670375824, 79.77545166015625 ] ] }, { "title": "1 INTRODUCTION", "heading_level": null, "page_id": 0, "polygon": [ [ 48.275001525878906, 268.34808349609375 ], [ 139.82582092285156, 268.34808349609375 ], [ 139.82582092285156, 279.57178688049316 ], [ 48.275001525878906, 279.57178688049316 ] ] }, { "title": "2 PRELIMINARIES", "heading_level": null, "page_id": 1, "polygon": [ [ 47.00848290324211, 246.11063075065613 ], [ 144.1628658771515, 246.11063075065613 ], [ 144.1628658771515, 257.7747802734375 ], [ 47.00848290324211, 257.7747802734375 ] ] }, { "title": "2.1 Notions and Definitions", "heading_level": null, "page_id": 1, "polygon": [ [ 47.878739565610886, 268.6039295196533 ], [ 177.51866161823273, 267.0570545196533 ], [ 177.51866161823273, 278.4881591796875 ], [ 47.878739565610886, 278.4881591796875 ] ] }, { "title": "2.1.1 Time and States", "heading_level": null, "page_id": 1, "polygon": [ [ 48.00010299682617, 287.373291015625 ], [ 148.70734602212906, 287.18165373802185 ], [ 147.79759216308594, 296.873291015625 ], [ 47.076856434345245, 298.28013253211975 ] ] }, { "title": "2.1.2 Forward / Reverse Process, and Transition Kernel", "heading_level": null, "page_id": 1, "polygon": [ [ 47.06732654571533, 494.55403661727905 ], [ 287.9966998100281, 494.55403661727905 ], [ 287.9966998100281, 505.0154113769531 ], [ 47.06732654571533, 505.0154113769531 ] ] }, { "title": "2.2 Background", "heading_level": null, "page_id": 1, "polygon": [ [ 311.0387849807739, 197.5656967163086 ], [ 390.6561985015869, 197.5656967163086 ], [ 390.6561985015869, 207.40899658203125 ], [ 311.0387849807739, 207.40899658203125 ] ] }, { "title": "2.2.1 Denoised Diffusion Probabilistic Models (DDPM)", "heading_level": null, "page_id": 1, "polygon": [ [ 311.804203748703, 290.21337389945984 ], [ 546.3936157226562, 290.21337389945984 ], [ 546.3936157226562, 300.41705322265625 ], [ 311.804203748703, 300.41705322265625 ] ] }, { "title": "2.2.2 Score SDE Formulation", "heading_level": null, "page_id": 2, "polygon": [ [ 310.63380575180054, 462.4605407714844 ], [ 442.3590817451477, 462.4605407714844 ], [ 442.3590817451477, 473.0755205154419 ], [ 310.63380575180054, 473.0755205154419 ] ] }, { "title": "2.3 Conditional Diffusion Probabilistic Models", "heading_level": null, "page_id": 3, "polygon": [ [ 47.66822934150696, 604.9064025878906 ], [ 261.5980529785156, 603.7531299591064 ], [ 261.5980529785156, 614.4064025878906 ], [ 47.66822934150696, 615.0247163772583 ] ] }, { "title": "3 ALGORITHM IMPROVEMENT", "heading_level": null, "page_id": 3, "polygon": [ [ 310.60228872299194, 135.48687744140625 ], [ 465.111750125885, 135.48687744140625 ], [ 465.111750125885, 146.48687744140625 ], [ 310.60228872299194, 146.99901008605957 ] ] }, { "title": "3.1 Sampling Acceleration", "heading_level": null, "page_id": 3, "polygon": [ [ 310.78796195983887, 322.2880859375 ], [ 437.1243591308594, 322.2880859375 ], [ 437.1243591308594, 331.7880859375 ], [ 310.78796195983887, 331.7880859375 ] ] }, { "title": "3.1.1 Knowledge Distillation", "heading_level": null, "page_id": 3, "polygon": [ [ 311.21745443344116, 412.96624755859375 ], [ 435.4048767089844, 412.96624755859375 ], [ 435.4048767089844, 422.46624755859375 ], [ 311.21745443344116, 422.46624755859375 ] ] }, { "title": "3.1.2 Training Schedule", "heading_level": null, "page_id": 4, "polygon": [ [ 47.796280950307846, 533.3675537109375 ], [ 155.94562017917633, 532.3196339607239 ], [ 155.94562017917633, 542.8675537109375 ], [ 47.796280950307846, 543.7435097694397 ] ] }, { "title": "3.1.3 Training-Free Sampling", "heading_level": null, "page_id": 5, "polygon": [ [ 47.80248221755028, 111.69244384765625 ], [ 177.79452681541443, 111.69244384765625 ], [ 177.79452681541443, 121.3444082736969 ], [ 47.80248221755028, 121.3444082736969 ] ] }, { "title": "3.1.4 Merging Diffusion and Other Generative Models", "heading_level": null, "page_id": 5, "polygon": [ [ 311.33779549598694, 216.30099749565125 ], [ 543.7611458301544, 216.30099749565125 ], [ 543.7611458301544, 226.040283203125 ], [ 311.33779549598694, 226.040283203125 ] ] }, { "title": "3.2 Diffusion Process Design", "heading_level": null, "page_id": 5, "polygon": [ [ 310.873575925827, 360.3977994918823 ], [ 449.7694866657257, 360.3977994918823 ], [ 449.7694866657257, 369.9301452636719 ], [ 310.873575925827, 369.9301452636719 ] ] }, { "title": "3.2.1 Latent Space", "heading_level": null, "page_id": 5, "polygon": [ [ 311.1778392791748, 535.6183776855469 ], [ 400.0317850112915, 535.6183776855469 ], [ 400.0317850112915, 545.1183624267578 ], [ 311.1778392791748, 545.1183624267578 ] ] }, { "title": "3.2.2 Emerging Forward Processes", "heading_level": null, "page_id": 6, "polygon": [ [ 48.00006103515625, 123.16748046875 ], [ 203.9444317817688, 122.21408879756927 ], [ 203.9444317817688, 132.66748046875 ], [ 48.00006103515625, 132.98185980319977 ] ] }, { "title": "3.2.3 Diffusion Models on non-Euclidean space", "heading_level": null, "page_id": 6, "polygon": [ [ 48.00006103515625, 492.141357421875 ], [ 255.05750155448914, 492.141357421875 ], [ 255.05750155448914, 501.641357421875 ], [ 48.00006103515625, 501.641357421875 ] ] }, { "title": "3.3 Likelihood Optimization", "heading_level": null, "page_id": 6, "polygon": [ [ 311.5372574329376, 383.3085014820099 ], [ 442.38751220703125, 383.3085014820099 ], [ 442.38751220703125, 392.89599609375 ], [ 310.3419449329376, 392.89599609375 ] ] }, { "title": "3.3.1 MLE Training", "heading_level": null, "page_id": 6, "polygon": [ [ 311.6392865180969, 488.3831481933594 ], [ 398.0222096443176, 487.8095898628235 ], [ 398.0222096443176, 497.8831481933594 ], [ 311.6392865180969, 498.8696894645691 ] ] }, { "title": "3.3.2 Hybrid Loss", "heading_level": null, "page_id": 6, "polygon": [ [ 307.10386204719543, 627.1071166992188 ], [ 393.7680308818817, 627.1071166992188 ], [ 393.7680308818817, 636.8405728340149 ], [ 307.10386204719543, 636.8405728340149 ] ] }, { "title": "3.4 Bridging Distributions", "heading_level": null, "page_id": 7, "polygon": [ [ 47.783613950014114, 44.471435546875 ], [ 170.7970428466797, 44.2112894654274 ], [ 170.7970428466797, 53.971435546875 ], [ 47.783613950014114, 55.07215243577957 ] ] }, { "title": "4 APPLICATION", "heading_level": null, "page_id": 7, "polygon": [ [ 47.9302693605423, 241.103759765625 ], [ 132.4382940530777, 241.103759765625 ], [ 132.4382940530777, 252.37904334068298 ], [ 47.9302693605423, 252.37904334068298 ] ] }, { "title": "4.1 Image Generation", "heading_level": null, "page_id": 7, "polygon": [ [ 48.000091552734375, 379.72540497779846 ], [ 151.71952039003372, 379.72540497779846 ], [ 150.9515838623047, 389.8841857910156 ], [ 47.01022928953171, 389.8841857910156 ] ] }, { "title": "4.1.1 Text condition", "heading_level": null, "page_id": 7, "polygon": [ [ 48.000091552734375, 481.2515158653259 ], [ 136.65509194135666, 481.2515158653259 ], [ 136.65509194135666, 491.141357421875 ], [ 48.000091552734375, 491.141357421875 ] ] }, { "title": "4.1.2 Image condition", "heading_level": null, "page_id": 7, "polygon": [ [ 311.4767768383026, 44.46331787109375 ], [ 410.0967700481415, 43.67128112912178 ], [ 410.0967700481415, 53.96331787109375 ], [ 311.4767768383026, 54.296885669231415 ] ] }, { "title": "4.2 3D Generation", "heading_level": null, "page_id": 7, "polygon": [ [ 309.4453876018524, 194.53417718410492 ], [ 400.68747782707214, 194.53417718410492 ], [ 400.68747782707214, 204.68133115768433 ], [ 309.4453876018524, 204.68133115768433 ] ] }, { "title": "4.2.1 3D Data Condition", "heading_level": null, "page_id": 7, "polygon": [ [ 311.22172236442566, 273.9081416130066 ], [ 420.24310302734375, 273.9081416130066 ], [ 420.24310302734375, 283.9860534667969 ], [ 311.22172236442566, 283.9860534667969 ] ] }, { "title": "4.2.2 2D Diffusion prior", "heading_level": null, "page_id": 7, "polygon": [ [ 312.0000915527344, 502.0212206840515 ], [ 416.19046354293823, 502.0212206840515 ], [ 416.19046354293823, 512.3568835258484 ], [ 312.0000915527344, 512.3568835258484 ] ] }, { "title": "4.3 Video Generation", "heading_level": null, "page_id": 7, "polygon": [ [ 312.0000915527344, 617.8960747718811 ], [ 413.80873918533325, 617.8960747718811 ], [ 413.80873918533325, 628.1890258789062 ], [ 312.0000915527344, 628.1890258789062 ] ] }, { "title": "4.4 Medical Analysis", "heading_level": null, "page_id": 8, "polygon": [ [ 47.47505530714989, 497.1264343261719 ], [ 147.79560470581055, 496.91334199905396 ], [ 147.79560470581055, 506.6264343261719 ], [ 47.47505530714989, 508.1649127006531 ] ] }, { "title": "4.4.1 In-distribution Analysis", "heading_level": null, "page_id": 8, "polygon": [ [ 47.812933176755905, 591.9715576171875 ], [ 175.70028591156006, 590.9186182022095 ], [ 175.70028591156006, 601.4715576171875 ], [ 47.812933176755905, 601.6587495803833 ] ] }, { "title": "4.4.2 Cross-distribution Generation", "heading_level": null, "page_id": 8, "polygon": [ [ 311.52784609794617, 497.1185302734375 ], [ 467.01995301246643, 497.1185302734375 ], [ 467.01995301246643, 507.60716342926025 ], [ 311.52784609794617, 507.60716342926025 ] ] }, { "title": "4.5 Text Generation", "heading_level": null, "page_id": 9, "polygon": [ [ 47.89712905883789, 44.471435546875 ], [ 143.62078392505646, 44.23121961951256 ], [ 143.62078392505646, 53.971435546875 ], [ 47.89712905883789, 54.731070935726166 ] ] }, { "title": "4.5.1 Discrete Generation", "heading_level": null, "page_id": 9, "polygon": [ [ 47.602409273386, 188.1763916015625 ], [ 163.00697326660156, 188.1763916015625 ], [ 163.00697326660156, 197.97099137306213 ], [ 47.602409273386, 197.97099137306213 ] ] }, { "title": "4.5.2 Latent Generation", "heading_level": null, "page_id": 9, "polygon": [ [ 47.99998474121094, 365.6183166503906 ], [ 155.554265499115, 365.6183166503906 ], [ 154.58045959472656, 375.1183166503906 ], [ 47.36612731218338, 375.1183166503906 ] ] }, { "title": "4.6 Time Series Generation", "heading_level": null, "page_id": 9, "polygon": [ [ 47.81795346736908, 617.4262390136719 ], [ 176.8294677734375, 616.9878616333008 ], [ 176.8294677734375, 626.9262390136719 ], [ 47.81795346736908, 627.492241859436 ] ] }, { "title": "4.7 Audio Generation", "heading_level": null, "page_id": 9, "polygon": [ [ 311.98889207839966, 191.2834997177124 ], [ 414.125, 191.2834997177124 ], [ 414.125, 201.29400634765625 ], [ 311.98889207839966, 201.29400634765625 ] ] }, { "title": "4.8 Molecule Design", "heading_level": null, "page_id": 9, "polygon": [ [ 310.7162826061249, 526.4540710449219 ], [ 409.66949462890625, 526.4540710449219 ], [ 409.66949462890625, 535.9540710449219 ], [ 310.7162826061249, 535.9540710449219 ] ] }, { "title": "4.8.1 Unconditional Generation", "heading_level": null, "page_id": 10, "polygon": [ [ 48.0, 44.46356201171875 ], [ 185.9832090139389, 44.00969249010086 ], [ 185.9832090139389, 53.96356201171875 ], [ 48.0, 55.040925085544586 ] ] }, { "title": "4.8.2 Cross-modal Generation", "heading_level": null, "page_id": 10, "polygon": [ [ 47.958161383867264, 219.14922451972961 ], [ 182.00698852539062, 219.14922451972961 ], [ 182.00698852539062, 229.00830078125 ], [ 47.958161383867264, 229.00830078125 ] ] }, { "title": "4.9 Graph Generation", "heading_level": null, "page_id": 10, "polygon": [ [ 47.57976096868515, 456.74822998046875 ], [ 151.4739990234375, 455.5087080001831 ], [ 151.4739990234375, 466.24822998046875 ], [ 47.57976096868515, 466.77670669555664 ] ] }, { "title": "5 CONCLUSIONS & DISCUSSIONS", "heading_level": null, "page_id": 10, "polygon": [ [ 310.30382537841797, 162.86865234375 ], [ 484.8111877441406, 161.60406303405762 ], [ 484.8111877441406, 173.86865234375 ], [ 310.30382537841797, 173.86865234375 ] ] }, { "title": "5.1 Conclusions", "heading_level": null, "page_id": 10, "polygon": [ [ 311.6214487552643, 178.50366532802582 ], [ 391.5245056152344, 176.95679032802582 ], [ 391.5245056152344, 188.06402587890625 ], [ 311.6214487552643, 188.51985561847687 ] ] }, { "title": "5.2 Comparison to Existing Surveys", "heading_level": null, "page_id": 10, "polygon": [ [ 310.53794145584106, 310.6576108932495 ], [ 481.4483685493469, 310.6576108932495 ], [ 481.4483685493469, 320.56458806991577 ], [ 310.53794145584106, 320.56458806991577 ] ] }, { "title": "5.3 Limitations and Future Directions", "heading_level": null, "page_id": 10, "polygon": [ [ 310.0482966899872, 465.2762360572815 ], [ 487.4131486415863, 465.2762360572815 ], [ 487.4131486415863, 475.4089660644531 ], [ 310.0482966899872, 475.4089660644531 ] ] }, { "title": "5.3.1 Challenges Under Data Limitations", "heading_level": null, "page_id": 10, "polygon": [ [ 310.05522751808167, 479.74424743652344 ], [ 489.4029846191406, 479.74424743652344 ], [ 489.4029846191406, 489.9430847167969 ], [ 310.05522751808167, 489.9430847167969 ] ] }, { "title": "5.3.2 Controllable Distribution-based Generation", "heading_level": null, "page_id": 10, "polygon": [ [ 311.49600076675415, 619.3520660400391 ], [ 522.6752686500549, 619.3520660400391 ], [ 522.6752686500549, 629.0228776931763 ], [ 311.49600076675415, 629.0228776931763 ] ] }, { "title": "5.3.4 Integration with Machine Learning Fields", "heading_level": null, "page_id": 11, "polygon": [ [ 47.552402436733246, 221.78400564193726 ], [ 250.39852595329285, 221.78400564193726 ], [ 250.39852595329285, 231.99801635742188 ], [ 47.552402436733246, 231.99801635742188 ] ] }, { "title": "ACKNOWLEDGMENTS", "heading_level": null, "page_id": 11, "polygon": [ [ 47.72047045826912, 440.74884033203125 ], [ 153.17500340938568, 439.70752716064453 ], [ 153.17500340938568, 451.74884033203125 ], [ 47.72047045826912, 452.0945177078247 ] ] }, { "title": "REFERENCES", "heading_level": null, "page_id": 11, "polygon": [ [ 47.70960000157356, 638.9708709716797 ], [ 115.5488691329956, 637.578085899353 ], [ 115.5488691329956, 649.9708709716797 ], [ 47.70960000157356, 650.6412677764893 ] ] }, { "title": "APPENDIX A\nSAMPLING ALGORITHMS", "heading_level": null, "page_id": 19, "polygon": [ [ 48.274993896484375, 42.301812529563904 ], [ 169.24319458007812, 40.754937529563904 ], [ 169.24319458007812, 67.81805419921875 ], [ 48.274993896484375, 68.85641348361969 ] ] }, { "title": "A.1 Unconditional Sampling", "heading_level": null, "page_id": 19, "polygon": [ [ 47.13164645433426, 166.92729949951172 ], [ 181.01898193359375, 166.92729949951172 ], [ 181.01898193359375, 179.08849954605103 ], [ 47.13164645433426, 179.08849954605103 ] ] }, { "title": "Algorithm 1 Ancestral Sampling [10]\n \u223c N (0, )", "heading_level": null, "page_id": 19, "polygon": [ [ 47.345088601112366, 204.97373056411743 ], [ 208.030264377594, 204.97373056411743 ], [ 208.030264377594, 238.04889965057373 ], [ 47.345088601112366, 238.04889965057373 ] ] }, { "title": "Algorithm 2 Annealed Langevin Dynamics Sampling [227]", "heading_level": null, "page_id": 19, "polygon": [ [ 47.80808615684509, 338.0934462547302 ], [ 298.825644493103, 338.0934462547302 ], [ 298.825644493103, 351.02008056640625 ], [ 47.80808615684509, 351.02008056640625 ] ] }, { "title": "A.1.3 Predictor-Corrector Sampling", "heading_level": null, "page_id": 19, "polygon": [ [ 46.752010345458984, 494.0372157096863 ], [ 202.67897033691406, 494.0372157096863 ], [ 202.67897033691406, 506.00312089920044 ], [ 46.752010345458984, 506.00312089920044 ] ] }, { "title": "A.2.1 Labeled Condition", "heading_level": null, "page_id": 19, "polygon": [ [ 312.0, 53.8285049200058 ], [ 425.85349702835083, 53.8285049200058 ], [ 425.85349702835083, 68.62255859375 ], [ 312.0, 68.62255859375 ] ] }, { "title": "Algorithm 5 Classifier-free Guidance Sampling [48]", "heading_level": null, "page_id": 19, "polygon": [ [ 312.0, 271.3610472679138 ], [ 533.6283845901489, 271.3610472679138 ], [ 533.6283845901489, 283.12429332733154 ], [ 312.0, 283.12429332733154 ] ] }, { "title": "A.3 Unlabeled Condition", "heading_level": null, "page_id": 19, "polygon": [ [ 310.8816375732422, 412.28538179397583 ], [ 429.86498737335205, 412.28538179397583 ], [ 429.86498737335205, 424.3767628669739 ], [ 310.8816375732422, 424.3767628669739 ] ] }, { "title": "APPENDIX B\nEVALUATION METRIC", "heading_level": null, "page_id": 20, "polygon": [ [ 48.274993896484375, 43.06063359975815 ], [ 151.7906036376953, 41.51375859975815 ], [ 151.7906036376953, 67.81805419921875 ], [ 48.274993896484375, 68.23908323049545 ] ] }, { "title": "B.1 Inception Score (IS)", "heading_level": null, "page_id": 20, "polygon": [ [ 47.450108885765076, 72.63043212890625 ], [ 163.38544464111328, 71.45107465982437 ], [ 163.38544464111328, 82.13043212890625 ], [ 47.450108885765076, 83.08496206998825 ] ] }, { "title": "B.2 Frechet Inception Distance (FID)", "heading_level": null, "page_id": 20, "polygon": [ [ 47.892040371894836, 313.3484191894531 ], [ 218.1142727136612, 312.2117793560028 ], [ 218.1142727136612, 322.8484191894531 ], [ 47.892040371894836, 323.8404858112335 ] ] }, { "title": "B.3 Negative Log Likelihood (NLL)", "heading_level": null, "page_id": 20, "polygon": [ [ 47.83203399181366, 516.1134181022644 ], [ 209.5189666748047, 516.1134181022644 ], [ 209.5189666748047, 526.7204895019531 ], [ 47.83203399181366, 526.7204895019531 ] ] }, { "title": "APPENDIX C\nBENCHMARKS", "heading_level": null, "page_id": 20, "polygon": [ [ 312.27496337890625, 41.99120011925697 ], [ 382.30535888671875, 41.99120011925697 ], [ 382.30535888671875, 67.818115234375 ], [ 312.27496337890625, 67.818115234375 ] ] }, { "title": "C.1 Benchmarks on CelebA-64", "heading_level": null, "page_id": 20, "polygon": [ [ 311.50610518455505, 156.2813812494278 ], [ 458.49649357795715, 156.2813812494278 ], [ 458.49649357795715, 166.96435546875 ], [ 311.50610518455505, 166.96435546875 ] ] }, { "title": "TABLE 6\nBenchmarks on CIFAR-10 (NFE < 1000)", "heading_level": null, "page_id": 21, "polygon": [ [ 365.79296875, 85.2161865234375 ], [ 510.2563920021057, 85.2161865234375 ], [ 510.2069396972656, 102.2161865234375 ], [ 365.79296875, 103.08226203918457 ] ] }, { "title": "TABLE 4\nBenchmarks on ImageNet-64", "heading_level": null, "page_id": 21, "polygon": [ [ 120.751021027565, 98.59209716320038 ], [ 229.4511011838913, 98.59209716320038 ], [ 229.4511011838913, 116.3841552734375 ], [ 120.751021027565, 116.3841552734375 ] ] }, { "title": "TABLE 7\nDetails for Improved Diffusion Methods", "heading_level": null, "page_id": 22, "polygon": [ [ 237.3690185546875, 123.71935093402863 ], [ 375.2003788948059, 122.17247593402863 ], [ 374.63299560546875, 141.61712646484375 ], [ 236.19656610488892, 142.13862526416779 ] ] }, { "title": "TABLE 9\nNotions in Diffusion Systems", "heading_level": null, "page_id": 24, "polygon": [ [ 123.09701538085938, 84.692138671875 ], [ 224.90499877929688, 83.67956811189651 ], [ 224.90499877929688, 101.692138671875 ], [ 122.57156717777252, 102.74860382080078 ] ] } ], "page_stats": [ { "page_id": 0, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 241 ], [ "Line", 97 ], [ "Footnote", 7 ], [ "Text", 6 ], [ "PageHeader", 3 ], [ "SectionHeader", 2 ], [ "TextInlineMath", 1 ] ] }, { "page_id": 1, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 665 ], [ "Line", 143 ], [ "Text", 12 ], [ "Equation", 7 ], [ "SectionHeader", 6 ], [ "PageHeader", 2 ], [ "TextInlineMath", 2 ] ] }, { "page_id": 2, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 626 ], [ "Line", 147 ], [ "Equation", 4 ], [ "Text", 4 ], [ "TextInlineMath", 3 ], [ "PageHeader", 2 ], [ "Figure", 1 ], [ "Caption", 1 ], [ "SectionHeader", 1 ], [ "FigureGroup", 1 ] ] }, { "page_id": 3, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 512 ], [ "Line", 148 ], [ "Text", 13 ], [ "Equation", 4 ], [ "SectionHeader", 4 ], [ "PageHeader", 2 ], [ "TextInlineMath", 2 ] ] }, { "page_id": 4, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 234 ], [ "Line", 98 ], [ "Text", 6 ], [ "PageHeader", 3 ], [ "Figure", 1 ], [ "SectionHeader", 1 ] ] }, { "page_id": 5, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 366 ], [ "Line", 141 ], [ "Text", 10 ], [ "SectionHeader", 4 ], [ "PageHeader", 2 ], [ "Equation", 2 ], [ "TextInlineMath", 2 ] ] }, { "page_id": 6, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 295 ], [ "Line", 117 ], [ "Text", 11 ], [ "SectionHeader", 5 ], [ "PageHeader", 2 ], [ "Equation", 2 ], [ "TextInlineMath", 1 ] ] }, { "page_id": 7, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 254 ], [ "Line", 116 ], [ "Text", 11 ], [ "SectionHeader", 9 ], [ "PageHeader", 2 ] ] }, { "page_id": 8, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 238 ], [ "Line", 97 ], [ "SectionHeader", 3 ], [ "Text", 3 ], [ "PageHeader", 2 ], [ "Figure", 1 ] ] }, { "page_id": 9, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 251 ], [ "Line", 116 ], [ "Text", 7 ], [ "SectionHeader", 6 ], [ "PageHeader", 2 ] ] }, { "page_id": 10, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 232 ], [ "Line", 114 ], [ "SectionHeader", 9 ], [ "Text", 8 ], [ "PageHeader", 2 ] ] }, { "page_id": 11, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 433 ], [ "Line", 117 ], [ "ListItem", 20 ], [ "Text", 5 ], [ "SectionHeader", 3 ], [ "PageHeader", 2 ], [ "ListGroup", 2 ] ] }, { "page_id": 12, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 615 ], [ "Line", 123 ], [ "ListItem", 36 ], [ "PageHeader", 2 ], [ "ListGroup", 2 ], [ "Text", 1 ] ] }, { "page_id": 13, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 628 ], [ "Line", 123 ], [ "ListItem", 37 ], [ "PageHeader", 2 ], [ "Text", 2 ], [ "ListGroup", 2 ] ] }, { "page_id": 14, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 516 ], [ "Line", 123 ], [ "ListItem", 36 ], [ "PageHeader", 2 ], [ "ListGroup", 2 ], [ "Text", 1 ] ] }, { "page_id": 15, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 539 ], [ "Line", 123 ], [ "ListItem", 34 ], [ "PageHeader", 2 ], [ "ListGroup", 2 ], [ "Text", 1 ] ] }, { "page_id": 16, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 517 ], [ "Line", 123 ], [ "ListItem", 30 ], [ "PageHeader", 2 ], [ "ListGroup", 2 ], [ "Text", 1 ] ] }, { "page_id": 17, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 496 ], [ "Line", 123 ], [ "ListItem", 32 ], [ "PageHeader", 2 ], [ "ListGroup", 2 ] ] }, { "page_id": 18, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 414 ], [ "Line", 102 ], [ "ListItem", 33 ], [ "PageHeader", 2 ], [ "Text", 2 ], [ "ListGroup", 2 ] ] }, { "page_id": 19, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 919 ], [ "Line", 215 ], [ "SectionHeader", 8 ], [ "TextInlineMath", 5 ], [ "PageHeader", 3 ], [ "Text", 2 ], [ "Equation", 2 ], [ "ListItem", 1 ], [ "Table", 1 ] ] }, { "page_id": 20, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 408 ], [ "Line", 112 ], [ "Text", 8 ], [ "SectionHeader", 6 ], [ "Equation", 3 ], [ "PageHeader", 2 ], [ "Table", 2 ] ] }, { "page_id": 21, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 465 ], [ "Line", 117 ], [ "Table", 3 ], [ "PageHeader", 2 ], [ "SectionHeader", 2 ], [ "Text", 1 ] ] }, { "page_id": 22, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 872 ], [ "Line", 67 ], [ "PageHeader", 2 ], [ "SectionHeader", 1 ], [ "Table", 1 ] ] }, { "page_id": 23, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 771 ], [ "Line", 70 ], [ "PageHeader", 2 ], [ "Text", 2 ], [ "Table", 1 ], [ "Caption", 1 ], [ "TableGroup", 1 ] ] }, { "page_id": 24, "text_extraction_method": "pdftext", "block_counts": [ [ "Span", 357 ], [ "Line", 96 ], [ "PageHeader", 2 ], [ "SectionHeader", 1 ], [ "Table", 1 ] ] } ], "debug_data_path": "debug_data\\Cao \u7b49 - 2023 - A survey on generative diffusion model" }