{"noiseShare":0.5219,"topics":[{"claims":[748,901,1137,1215,1289,1561,1644,1922,2669],"deltaPp":-4.1959,"id":0,"keywords":["convergence","gradient","stochastic","optimization","descent"],"label":"Gradient Descent Convergence","labelFull":"Theoretical Convergence Guarantees for Gradient Descent","papers":7503,"papersByYear":[403,490,621,705,737,898,935,1125,1589],"rel":-0.3875,"reviewed":true,"share":[10.8275,9.6609,9.5656,9.2193,8.4577,8.0697,6.6824,6.2779,6.6316],"size":13086,"slope":-0.005641},{"claims":[440,738,957,973,1140,1194,1291,1454,1699],"deltaPp":-2.1547,"id":1,"keywords":["adversarial","attacks","robustness","backdoor","examples"],"label":"Adversarial Robustness","labelFull":"Techniques for Adversarial Robustness and Defense","papers":4561,"papersByYear":[220,348,415,414,470,505,590,699,900],"rel":-0.3645,"reviewed":true,"share":[5.9108,6.8612,6.3925,5.4139,5.3936,4.5381,4.2167,3.9007,3.7561],"size":9886,"slope":-0.003788},{"claims":[250,534,813,828,1053,1375,1440,1666,1661],"deltaPp":0.0474,"id":2,"keywords":["graph","node","networks","neural","classification"],"label":"Graph Neural Networks","labelFull":"Graph Neural Networks for Node and Graph Classification","papers":4044,"papersByYear":[118,226,350,362,420,539,573,685,771],"rel":0.015,"reviewed":true,"share":[3.1703,4.4558,5.3913,4.7339,4.8198,4.8436,4.0952,3.8225,3.2177],"size":9620,"slope":-0.000699},{"claims":[310,567,636,808,948,1148,1337,1329,1180],"deltaPp":-1.7717,"id":3,"keywords":["segmentation","semantic","instance","medical","image"],"label":"Universal Image Segmentation","labelFull":"Universal Image Segmentation: Semantic, Instance, Panoptic","papers":4288,"papersByYear":[168,287,340,419,481,603,666,667,657],"rel":-0.3925,"reviewed":true,"share":[4.5137,5.6585,5.2372,5.4793,5.5199,5.4188,4.7599,3.7221,2.742],"size":8263,"slope":-0.002319},{"claims":[589,741,854,797,840,915,846,836,992],"deltaPp":-6.3055,"id":4,"keywords":["networks","neural","relu","deep","activation"],"label":"Layerwise Training","labelFull":"Layerwise Training for Efficient Deep Neural Networks","papers":4750,"papersByYear":[347,424,490,510,525,589,571,571,723],"rel":-0.6763,"reviewed":true,"share":[9.3229,8.3596,7.5478,6.6693,6.0248,5.293,4.0809,3.1864,3.0174],"size":7410,"slope":-0.008175},{"claims":[67,109,190,261,392,710,1067,1791,2505],"deltaPp":3.0868,"id":5,"keywords":["protein","molecular","quantum","prediction","design"],"label":"Molecular Machine Learning","labelFull":"Machine Learning for Molecular Prediction and Design","papers":2833,"papersByYear":[42,55,88,115,166,286,403,668,1010],"rel":2.7356,"reviewed":true,"share":[1.1284,1.0844,1.3555,1.5039,1.905,2.5701,2.8802,3.7277,4.2152],"size":7092,"slope":0.004065},{"claims":[10,19,15,52,129,567,1129,1602,2064],"deltaPp":4.879,"id":6,"keywords":["diffusion","denoising","sampling","process","latent"],"label":"Efficient Diffusion Models","labelFull":"Efficient Training and Sampling for Diffusion Models","papers":3232,"papersByYear":[9,14,13,33,66,316,643,911,1227],"rel":20.1778,"reviewed":true,"share":[0.2418,0.276,0.2002,0.4315,0.7574,2.8397,4.5955,5.0837,5.1208],"size":5587,"slope":0.007523},{"claims":[241,269,363,536,573,675,789,757,1268],"deltaPp":-0.1183,"id":7,"keywords":["reinforcement","learning","offline","policy","actor-critic"],"label":"Offline Reinforcement Learning","labelFull":"Offline Reinforcement Learning: Algorithms and Benchmarks","papers":3519,"papersByYear":[147,180,234,313,354,403,478,492,918],"rel":-0.03,"reviewed":true,"share":[3.9495,3.5489,3.6044,4.0931,4.0624,3.6215,3.4162,2.7455,3.8312],"size":5471,"slope":-0.000622},{"claims":[285,370,427,531,613,605,693,718,593],"deltaPp":-2.1264,"id":8,"keywords":["pose","motion","human","estimation","camera"],"label":"3D Pose Estimation","labelFull":"3D Pose Estimation for Humans, Objects, and Cameras","papers":2559,"papersByYear":[136,176,205,247,311,332,386,400,366],"rel":-0.582,"reviewed":true,"share":[3.6539,3.47,3.1577,3.23,3.569,2.9835,2.7587,2.2321,1.5275],"size":4835,"slope":-0.002211},{"claims":[193,242,357,422,422,570,694,701,788],"deltaPp":-0.4785,"id":9,"keywords":["super-resolution","restoration","denoising","image","reconstruction"],"label":"Image Restoration","labelFull":"Deep Learning for Image Restoration and Enhancement","papers":2541,"papersByYear":[102,126,172,215,225,311,393,455,542],"rel":-0.1746,"reviewed":true,"share":[2.7405,2.4842,2.6494,2.8116,2.5821,2.7948,2.8087,2.5391,2.262],"size":4389,"slope":-0.000241},{"claims":[110,140,255,320,336,546,668,768,1023],"deltaPp":0.742,"id":10,"keywords":["causal","counterfactual","treatment","effect","variables"],"label":"Causal Discovery Benchmarking","labelFull":"Benchmarking and Evaluation of Causal Discovery Algorithms","papers":1955,"papersByYear":[53,67,112,148,166,257,290,343,519],"rel":0.5211,"reviewed":true,"share":[1.424,1.321,1.7252,1.9354,1.905,2.3095,2.0726,1.9141,2.166],"size":4166,"slope":0.000969},{"claims":[128,208,267,452,589,660,594,588,535],"deltaPp":-1.208,"id":11,"keywords":["vision","transformers","visual","image","classification"],"label":"Self-Supervised Vision Transformers","labelFull":"Self-Supervised Vision Transformers: Efficient Training","papers":3116,"papersByYear":[113,177,224,339,426,487,459,453,438],"rel":-0.3979,"reviewed":true,"share":[3.036,3.4897,3.4504,4.4331,4.8887,4.3763,3.2804,2.5279,1.828],"size":4021,"slope":-0.001352},{"claims":[376,355,445,420,487,575,514,532,298],"deltaPp":-3.1211,"id":12,"keywords":["face","facial","recognition","identity","expression"],"label":"Face Representation Learning","labelFull":"Identity-Preserving Face Representation Learning","papers":1572,"papersByYear":[139,138,167,155,185,201,211,229,147],"rel":-0.8357,"reviewed":true,"share":[3.7346,2.7208,2.5724,2.0269,2.123,1.8063,1.508,1.2779,0.6135],"size":4002,"slope":-0.003194},{"claims":[116,151,163,284,333,446,583,771,981],"deltaPp":0.5214,"id":13,"keywords":["audio","speech","music","sound","acoustic"],"label":"Audio Foundation Models","labelFull":"Large-Scale Self-Supervised Audio Foundation Models","papers":1350,"papersByYear":[39,62,52,99,127,153,183,259,376],"rel":0.4976,"reviewed":true,"share":[1.0478,1.2224,0.801,1.2946,1.4574,1.3749,1.3079,1.4453,1.5692],"size":3828,"slope":0.000641},{"claims":[60,114,210,310,389,506,592,686,959],"deltaPp":1.6623,"id":14,"keywords":["distillation","teacher","student","knowledge","feature"],"label":"Knowledge Distillation","labelFull":"Knowledge Distillation for Model Compression and Transfer","papers":2147,"papersByYear":[29,54,101,158,209,279,335,397,585],"rel":2.1333,"reviewed":true,"share":[0.7792,1.0647,1.5558,2.0662,2.3984,2.5072,2.3942,2.2154,2.4415],"size":3826,"slope":0.002037},{"claims":[18,39,66,83,156,374,605,1009,1369],"deltaPp":3.5743,"id":15,"keywords":["vision-language","visual","reasoning","text","alignment"],"label":"Vision-Language Model Evaluation","labelFull":"Evaluation, Robustness, and Optimization of VLMs","papers":2559,"papersByYear":[15,29,50,67,104,256,407,678,953],"rel":8.8692,"reviewed":true,"share":[0.403,0.5718,0.7702,0.8762,1.1935,2.3005,2.9088,3.7835,3.9773],"size":3719,"slope":0.004939},{"claims":[19,25,44,45,99,200,729,1122,1385],"deltaPp":3.8706,"id":16,"keywords":["fine-tuning","enables","framework"],"label":"LLM Fine-Tuning Frameworks","labelFull":"Frameworks for Fine-Tuning Large Language Models","papers":2637,"papersByYear":[15,23,33,35,75,138,509,785,1024],"rel":9.6045,"reviewed":true,"share":[0.403,0.4535,0.5083,0.4577,0.8607,1.2401,3.6378,4.3806,4.2736],"size":3668,"slope":0.005718},{"claims":[153,174,237,289,370,451,529,642,756],"deltaPp":-0.9835,"id":17,"keywords":["state-of-the-art","benchmark","outperforms","real-world","synthetic"],"label":"Benchmark Evaluation","labelFull":"State-of-the-Art Performance Evaluation on Standard Benchmarks","papers":3487,"papersByYear":[150,172,225,276,360,440,511,623,730],"rel":-0.244,"reviewed":true,"share":[4.0301,3.3912,3.4658,3.6093,4.1313,3.954,3.6521,3.4766,3.0466],"size":3601,"slope":-0.000493},{"claims":[30,55,80,166,288,554,768,916,662],"deltaPp":1.0903,"id":18,"keywords":["rendering","radiance","gaussian","splatting","view synthesis"],"label":"3D Gaussian Splatting","labelFull":"3D Gaussian Splatting for Real-Time Neural Rendering","papers":1994,"papersByYear":[20,37,55,92,164,314,429,493,390],"rel":2.0292,"reviewed":true,"share":[0.5373,0.7295,0.8472,1.2031,1.882,2.8217,3.066,2.7511,1.6276],"size":3519,"slope":0.002747},{"claims":[56,79,97,138,121,141,307,685,1892],"deltaPp":3.3453,"id":19,"keywords":["agents","ai","llm","systems","games"],"label":"LLM Agents","labelFull":"Large Language Model Agents: Methods and Benchmarks","papers":2018,"papersByYear":[33,55,72,98,103,95,186,362,1014],"rel":3.7732,"reviewed":true,"share":[0.8866,1.0844,1.1091,1.2815,1.182,0.8537,1.3293,2.0201,4.2319],"size":3516,"slope":0.0027},{"claims":[66,120,184,237,280,359,444,729,1090],"deltaPp":1.6419,"id":20,"keywords":["navigation","manipulation","robotic","locomotion","environments"],"label":"Multi-Task Robot Learning","labelFull":"Robot Learning for Navigation, Manipulation, and Locomotion","papers":2137,"papersByYear":[47,82,105,153,166,208,259,421,696],"rel":1.3002,"reviewed":true,"share":[1.2628,1.6167,1.6174,2.0008,1.905,1.8692,1.8511,2.3493,2.9047],"size":3509,"slope":0.001517},{"claims":[1,5,1,9,44,384,987,1082,817],"deltaPp":2.6358,"id":21,"keywords":["diffusion","text-to-image","video","generation","latent"],"label":"Diffusion Model Adaptation","labelFull":"Adapting Diffusion Models for Image, Video, and Control","papers":2424,"papersByYear":[1,4,1,7,26,261,689,797,638],"rel":97.9851,"reviewed":true,"share":[0.0269,0.0789,0.0154,0.0915,0.2984,2.3454,4.9242,4.4475,2.6627],"size":3330,"slope":0.005953},{"claims":[44,71,94,130,161,259,505,683,1136],"deltaPp":1.4828,"id":22,"keywords":["time series","forecasting","multivariate","weather","autoregressive"],"label":"Neural Time Series Forecasting","labelFull":"Deep Neural Models for Multivariate Time Series Forecasting","papers":1487,"papersByYear":[29,41,55,77,84,132,217,310,542],"rel":1.903,"reviewed":true,"share":[0.7792,0.8084,0.8472,1.0069,0.964,1.1862,1.5509,1.7299,2.262],"size":3083,"slope":0.001714},{"claims":[276,313,360,330,310,349,323,367,435],"deltaPp":-2.7507,"id":23,"keywords":["variational","posterior","gaussian","inference","bayesian"],"label":"Variational Bayesian Neural Networks","labelFull":"Scalable Variational Inference for Bayesian Neural Networks","papers":1929,"papersByYear":[151,191,213,209,187,218,207,240,313],"rel":-0.678,"reviewed":true,"share":[4.057,3.7658,3.281,2.7331,2.146,1.959,1.4794,1.3393,1.3063],"size":3063,"slope":-0.003777},{"claims":[119,164,249,299,307,388,403,371,697],"deltaPp":-0.1164,"id":24,"keywords":["policy","markov decision","decision processes","mdps","gradient"],"label":"Off-Policy Reinforcement Learning","labelFull":"Off-Policy RL: Optimization and Evaluation","papers":1976,"papersByYear":[82,108,159,201,200,242,240,244,500],"rel":-0.0528,"reviewed":true,"share":[2.2031,2.1293,2.4492,2.6285,2.2952,2.1747,1.7153,1.3616,2.0867],"size":2997,"slope":-0.000782},{"claims":[112,160,275,354,443,502,431,381,293],"deltaPp":-1.0862,"id":25,"keywords":["label","semi-supervised","unlabeled","noise","learning"],"label":"Learning With Noisy Labels","labelFull":"Learning With Noisy Labels: Semi-supervised Pseudo-Labeling","papers":1924,"papersByYear":[76,113,169,233,275,317,272,240,229],"rel":-0.532,"reviewed":true,"share":[2.0419,2.2279,2.6032,3.0469,3.1558,2.8487,1.944,1.3393,0.9557],"size":2951,"slope":-0.001421},{"claims":[245,272,309,298,383,353,322,453,311],"deltaPp":-2.0675,"id":26,"keywords":["depth","stereo","estimation","monocular","disparity"],"label":"Monocular Depth Estimation","labelFull":"Monocular Depth Estimation from Single Images","papers":1415,"papersByYear":[106,115,147,122,171,179,165,223,187],"rel":-0.726,"reviewed":true,"share":[2.8479,2.2674,2.2643,1.5954,1.9624,1.6086,1.1792,1.2444,0.7804],"size":2946,"slope":-0.002249},{"claims":[153,246,318,381,449,420,368,297,298],"deltaPp":-1.4574,"id":27,"keywords":["detection","object","detectors","coco","localization"],"label":"Object Detection","labelFull":"State-of-the-Art Object Detection Methods and Benchmarks","papers":1696,"papersByYear":[85,132,163,198,246,249,217,208,198],"rel":-0.6382,"reviewed":true,"share":[2.2837,2.6025,2.5108,2.5893,2.823,2.2376,1.5509,1.1607,0.8263],"size":2930,"slope":-0.002071},{"claims":[123,165,257,275,329,407,451,450,386],"deltaPp":-0.7825,"id":28,"keywords":["shape","surface","reconstruction","3d","mesh"],"label":"Neural Implicit Surface Reconstruction","labelFull":"Neural Implicit Surface Reconstruction for 3D Shapes","papers":1756,"papersByYear":[72,98,154,154,187,256,267,292,276],"rel":-0.4045,"reviewed":true,"share":[1.9344,1.9322,2.3722,2.0139,2.146,2.3005,1.9082,1.6295,1.1519],"size":2843,"slope":-0.00078},{"claims":[19,64,76,115,171,272,428,749,932],"deltaPp":2.0575,"id":29,"keywords":["multimodal","modality","fusion","alignment","information"],"label":"Modality-Aware Multimodal Fusion","labelFull":"Modality-Aware Multimodal Fusion and Alignment","papers":1768,"papersByYear":[16,43,53,79,109,158,263,451,596],"rel":4.786,"reviewed":true,"share":[0.4299,0.8478,0.8164,1.0331,1.2509,1.4198,1.8796,2.5167,2.4874],"size":2826,"slope":0.002625},{"claims":[62,86,112,194,200,233,368,574,957],"deltaPp":1.301,"id":30,"keywords":["attention","heads","softmax","linear","sparse"],"label":"Efficient Self-Attention","labelFull":"Efficient and Sparse Self-Attention Mechanisms for Transformers","papers":1792,"papersByYear":[44,61,68,122,142,157,249,354,595],"rel":1.1005,"reviewed":true,"share":[1.1822,1.2027,1.0474,1.5954,1.6296,1.4109,1.7796,1.9754,2.4832],"size":2786,"slope":0.001467},{"claims":[117,160,214,191,275,247,326,466,707],"deltaPp":0.2496,"id":31,"keywords":["quantization","compression","quantized","post-training","error"],"label":"Neural Network Quantization","labelFull":"Neural Network Quantization: PTQ and QAT Techniques","papers":1433,"papersByYear":[53,77,102,99,138,129,187,247,401],"rel":0.1753,"reviewed":true,"share":[1.424,1.5181,1.5712,1.2946,1.5837,1.1592,1.3365,1.3783,1.6736],"size":2703,"slope":-4e-06},{"claims":[102,203,214,276,333,370,380,370,343],"deltaPp":-0.4142,"id":32,"keywords":["few-shot","zero-shot","classification","learning","generalization"],"label":"Few-Shot & Zero-Shot Learning","labelFull":"Few-Shot and Zero-Shot Learning for Image Classification","papers":1737,"papersByYear":[60,105,120,165,203,256,262,279,287],"rel":-0.2569,"reviewed":true,"share":[1.612,2.0702,1.8484,2.1577,2.3296,2.3005,1.8725,1.5569,1.1978],"size":2591,"slope":-0.000501},{"claims":[6,12,22,23,26,86,189,645,1581],"deltaPp":4.283,"id":33,"keywords":["reasoning","mathematical","math","chain-of-thought","benchmarks"],"label":"Efficient Chain-of-Thought Reasoning","labelFull":"Efficient Chain-of-Thought Reasoning in Large Language Models","papers":1716,"papersByYear":[4,6,16,18,18,58,134,410,1052],"rel":39.8419,"reviewed":true,"share":[0.1075,0.1183,0.2465,0.2354,0.2066,0.5212,0.9577,2.2879,4.3905],"size":2590,"slope":0.004225},{"claims":[78,140,172,238,279,406,370,396,469],"deltaPp":0.0399,"id":34,"keywords":["privacy","private","differential","guarantees","utility"],"label":"Differentially Private Machine Learning","labelFull":"Differential Privacy Guarantees for Machine Learning","papers":1131,"papersByYear":[33,59,64,97,112,175,169,200,222],"rel":0.045,"reviewed":true,"share":[0.8866,1.1632,0.9858,1.2685,1.2853,1.5726,1.2078,1.1161,0.9265],"size":2548,"slope":0.000128},{"claims":[5,7,25,32,71,146,291,645,1290],"deltaPp":3.7971,"id":35,"keywords":["token","attention","prediction","information","inference"],"label":"Token Pruning","labelFull":"Dynamic Token Pruning for Efficient Transformer Inference","papers":1841,"papersByYear":[5,6,18,30,58,111,211,460,942],"rel":28.2733,"reviewed":true,"share":[0.1343,0.1183,0.2773,0.3923,0.6656,0.9975,1.508,2.567,3.9314],"size":2512,"slope":0.004267},{"claims":[34,58,103,138,225,317,365,567,625],"deltaPp":1.1863,"id":36,"keywords":["continual","forgetting","unlearning","learning","catastrophic"],"label":"Continual Learning","labelFull":"Continual Learning with Catastrophic Forgetting Mitigation","papers":1427,"papersByYear":[20,36,67,75,113,176,203,324,413],"rel":2.2079,"reviewed":true,"share":[0.5373,0.7098,1.032,0.9808,1.2968,1.5816,1.4508,1.808,1.7236],"size":2432,"slope":0.00158},{"claims":[78,156,223,309,305,345,311,301,177],"deltaPp":-0.6151,"id":37,"keywords":["point","cloud","registration","3d","completion"],"label":"3D Point Cloud Deep Learning","labelFull":"Deep Learning Methods for 3D Point Cloud Analysis","papers":1253,"papersByYear":[42,80,120,150,164,210,186,178,123],"rel":-0.5451,"reviewed":true,"share":[1.1284,1.5773,1.8484,1.9616,1.882,1.8871,1.3293,0.9933,0.5133],"size":2205,"slope":-0.000888},{"claims":[39,56,90,102,106,179,234,453,880],"deltaPp":1.9959,"id":38,"keywords":["memory","gpu","latency","benchmarks","overhead"],"label":"Efficient Inference","labelFull":"Efficient Inference and Model Acceleration for Deep Neural Networks","papers":1799,"papersByYear":[36,48,76,87,94,156,208,384,710],"rel":2.0636,"reviewed":true,"share":[0.9672,0.9464,1.1707,1.1377,1.0787,1.4019,1.4866,2.1429,2.9631],"size":2139,"slope":0.002078},{"claims":[81,146,161,155,189,167,335,469,422],"deltaPp":-0.0861,"id":39,"keywords":["driving","autonomous","traffic","lane","vehicle"],"label":"End-to-End Autonomous Driving","labelFull":"End-to-End Autonomous Driving Model Training and Evaluation","papers":1083,"papersByYear":[38,60,84,79,96,95,168,239,224],"rel":-0.0843,"reviewed":true,"share":[1.021,1.183,1.2939,1.0331,1.1017,0.8537,1.2007,1.3337,0.9349],"size":2125,"slope":-4.3e-05},{"claims":[26,49,111,129,181,250,304,437,606],"deltaPp":1.1406,"id":40,"keywords":["pde","equations","neural","differential","physics-informed"],"label":"Neural PDE Solvers","labelFull":"Neural Networks for Solving Partial Differential Equations","papers":1142,"papersByYear":[13,29,58,70,94,129,165,227,357],"rel":3.2654,"reviewed":true,"share":[0.3493,0.5718,0.8934,0.9154,1.0787,1.1592,1.1792,1.2667,1.4899],"size":2093,"slope":0.001244},{"claims":[114,142,163,260,266,224,201,323,315],"deltaPp":-1.112,"id":41,"keywords":["video","frames","temporal","prediction","features"],"label":"Efficient Video Modeling","labelFull":"Efficient Video Modeling and Representation Learning","papers":1460,"papersByYear":[81,94,127,173,177,171,151,231,255],"rel":-0.511,"reviewed":true,"share":[2.1762,1.8533,1.9563,2.2623,2.0312,1.5367,1.0792,1.2891,1.0642],"size":2008,"slope":-0.001437},{"claims":[0,8,57,141,266,397,437,372,312],"deltaPp":0.7721,"id":42,"keywords":["federated","learning","client","personalized","communication"],"label":"Heterogeneous Federated Learning","labelFull":"Federated Learning for Heterogeneous Data and Clients","papers":977,"papersByYear":[0,3,25,65,117,192,205,185,185],"rel":null,"reviewed":true,"share":[0.0,0.0591,0.3851,0.85,1.3427,1.7254,1.4651,1.0324,0.7721],"size":1990,"slope":0.001507},{"claims":[180,211,238,271,277,225,227,183,140],"deltaPp":-2.0753,"id":43,"keywords":["action","recognition","temporal","localization","activity"],"label":"Video Action Recognition","labelFull":"Video Action Recognition and Temporal Action Detection","papers":1039,"papersByYear":[92,106,116,134,139,126,119,112,95],"rel":-0.8396,"reviewed":true,"share":[2.4718,2.0899,1.7868,1.7523,1.5951,1.1323,0.8505,0.625,0.3965],"size":1952,"slope":-0.002531},{"claims":[101,72,83,94,78,126,223,425,719],"deltaPp":0.7467,"id":44,"keywords":["question","answering","retrieval-augmented","summarization","vqa"],"label":"Retrieval-Augmented QA","labelFull":"Retrieval-Augmented Generation for Question Answering","papers":1334,"papersByYear":[55,46,52,60,51,80,155,302,533],"rel":0.5053,"reviewed":true,"share":[1.4777,0.9069,0.801,0.7846,0.5853,0.7189,1.1078,1.6853,2.2244],"size":1921,"slope":0.000978},{"claims":[16,24,27,32,89,161,411,515,567],"deltaPp":1.3039,"id":45,"keywords":["editing","generation","text-to-image","quality","video"],"label":"Text-to-Image Generation and Editing","labelFull":"Unified Text-to-Image Generation and Editing Models","papers":1365,"papersByYear":[15,20,23,23,56,117,302,400,409],"rel":3.2355,"reviewed":true,"share":[0.403,0.3943,0.3543,0.3008,0.6426,1.0514,2.1584,2.2321,1.7069],"size":1842,"slope":0.002515},{"claims":[53,100,162,182,216,254,254,256,344],"deltaPp":0.0847,"id":46,"keywords":["regret","bound","algorithm","sqrt","online"],"label":"Online Regret Minimization","labelFull":"Online Regret Minimization with Sublinear Regret Bounds","papers":1129,"papersByYear":[29,64,105,116,134,167,155,152,207],"rel":0.1087,"reviewed":true,"share":[0.7792,1.2618,1.6174,1.5169,1.5378,1.5007,1.1078,0.8482,0.8639],"size":1821,"slope":-0.000323},{"claims":[109,166,233,278,251,229,195,130,160],"deltaPp":-1.7633,"id":47,"keywords":["imagenet","top-1","accuracy","resnet","classification"],"label":"ImageNet Classification","labelFull":"ImageNet Classification: State‑of‑the‑Art Top‑1 Accuracy","papers":1480,"papersByYear":[88,136,185,228,213,191,171,124,144],"rel":-0.7458,"reviewed":true,"share":[2.3643,2.6814,2.8497,2.9816,2.4443,1.7164,1.2221,0.692,0.601],"size":1751,"slope":-0.002924},{"claims":[75,128,182,236,220,287,239,211,152],"deltaPp":-0.7207,"id":48,"keywords":["domain","adaptation","source","generalization","unsupervised"],"label":"Unsupervised Domain Adaptation","labelFull":"Unsupervised Multi-Source Domain Adaptation","papers":1114,"papersByYear":[45,75,106,132,138,185,169,147,117],"rel":-0.5961,"reviewed":true,"share":[1.209,1.4787,1.6328,1.7262,1.5837,1.6625,1.2078,0.8203,0.4883],"size":1730,"slope":-0.000962},{"claims":[64,125,164,214,220,252,223,268,192],"deltaPp":-0.4998,"id":49,"keywords":["bandits","regret","contextual","multi-armed","linear"],"label":"Bandit Regret Minimization","labelFull":"Regret Minimization for Stochastic and Contextual Bandits","papers":908,"papersByYear":[36,66,81,108,116,136,111,142,112],"rel":-0.5167,"reviewed":true,"share":[0.9672,1.3013,1.2477,1.4123,1.3312,1.2221,0.7933,0.7924,0.4674],"size":1722,"slope":-0.000771},{"claims":[115,123,109,130,89,130,178,234,370],"deltaPp":-0.5718,"id":50,"keywords":["low-rank","matrix","matrices","sensing","factorization"],"label":"Low-Rank Adaptation","labelFull":"Low-Rank Matrix Adaptation for Efficient Model Fine-Tuning","papers":951,"papersByYear":[64,66,70,70,60,75,111,160,275],"rel":-0.3325,"reviewed":true,"share":[1.7195,1.3013,1.0783,0.9154,0.6885,0.674,0.7933,0.8929,1.1477],"size":1478,"slope":-0.000721},{"claims":[163,196,195,190,205,190,112,100,126],"deltaPp":-1.6418,"id":51,"keywords":["saliency","video object","object segmentation","salient","video instance"],"label":"Video Salient Object Detection","labelFull":"Video Salient Object Detection and Segmentation","papers":771,"papersByYear":[74,84,90,94,105,106,66,69,83],"rel":-0.8258,"reviewed":true,"share":[1.9882,1.6562,1.3863,1.2292,1.205,0.9526,0.4717,0.385,0.3464],"size":1477,"slope":-0.002081},{"claims":[130,113,144,171,170,177,173,210,183],"deltaPp":-1.1368,"id":52,"keywords":["tracking","trackers","multi-object","visual","state-of-the-art"],"label":"Multi-Object Tracking","labelFull":"Multi-Object Tracking in Video Sequences","papers":686,"papersByYear":[58,47,65,71,71,84,86,103,101],"rel":-0.7295,"reviewed":true,"share":[1.5583,0.9267,1.0012,0.9285,0.8148,0.7549,0.6146,0.5748,0.4215],"size":1471,"slope":-0.001092},{"claims":[55,86,97,115,130,164,176,259,328],"deltaPp":-0.1234,"id":53,"keywords":["generative","modeling","latent","deep","distribution"],"label":"Plug-and-Play Generative Modeling","labelFull":"Plug-and-Play Adaptation for Deep Generative Models","papers":1095,"papersByYear":[47,62,71,95,96,121,132,198,273],"rel":-0.0977,"reviewed":true,"share":[1.2628,1.2224,1.0937,1.2423,1.1017,1.0873,0.9434,1.1049,1.1394],"size":1410,"slope":-0.000217},{"claims":[22,52,33,40,76,115,180,316,527],"deltaPp":0.5713,"id":54,"keywords":["eeg","brain","fmri","activity","recordings"],"label":"EEG Foundation Models","labelFull":"Pretrained EEG Foundation Models for Neural Decoding","papers":645,"papersByYear":[14,27,20,27,39,62,85,144,227],"rel":1.519,"reviewed":true,"share":[0.3761,0.5323,0.3081,0.3531,0.4476,0.5572,0.6075,0.8036,0.9474],"size":1361,"slope":0.00065},{"claims":[26,25,40,64,121,137,207,266,473],"deltaPp":0.9007,"id":55,"keywords":["query","retrieval","attention","information","accuracy"],"label":"Query-Aware Retrieval","labelFull":"Query-Aware Retrieval: Decomposition, Expansion, and Efficiency","papers":1034,"papersByYear":[21,21,35,49,93,113,153,198,351],"rel":1.5964,"reviewed":true,"share":[0.5642,0.414,0.5391,0.6408,1.0672,1.0155,1.0935,1.1049,1.4649],"size":1359,"slope":0.001193},{"claims":[16,53,103,134,196,271,251,192,125],"deltaPp":0.0217,"id":56,"keywords":["lidar","3d object","object detection","monocular 3d","kitti"],"label":"LiDAR 3D Object Detection","labelFull":"LiDAR and Pseudo‑LiDAR 3D Object Detection in Point Clouds","papers":743,"papersByYear":[11,30,58,73,106,150,133,106,76],"rel":0.0734,"reviewed":true,"share":[0.2955,0.5915,0.8934,0.9546,1.2164,1.348,0.9505,0.5915,0.3172],"size":1341,"slope":9.9e-05},{"claims":[45,55,60,78,40,68,155,238,599],"deltaPp":1.1396,"id":57,"keywords":["code","verification","program","tool","execution"],"label":"Verified Code Generation","labelFull":"Execution-Based Verification for LLM Code Generation","papers":856,"papersByYear":[25,31,34,32,22,41,93,144,434],"rel":1.6966,"reviewed":true,"share":[0.6717,0.6112,0.5237,0.4185,0.2525,0.3684,0.6647,0.8036,1.8113],"size":1338,"slope":0.000895},{"claims":[24,43,63,80,103,159,212,338,315],"deltaPp":0.3143,"id":58,"keywords":["anomaly","detection","normal","industrial","video"],"label":"Unsupervised Anomaly Detection","labelFull":"Unsupervised Anomaly Detection for Visual and Tabular Data","papers":594,"papersByYear":[13,20,35,36,41,69,88,133,159],"rel":0.8998,"reviewed":true,"share":[0.3493,0.3943,0.5391,0.4708,0.4705,0.6201,0.6289,0.7422,0.6636],"size":1337,"slope":0.000438},{"claims":[30,20,75,73,126,135,220,239,355],"deltaPp":0.2994,"id":59,"keywords":["spiking","neural networks","neurons","membrane","neuromorphic"],"label":"Spiking Neural Network Training","labelFull":"Training Methods for Spiking Neural Networks","papers":534,"papersByYear":[12,12,28,35,48,58,91,101,149],"rel":0.9287,"reviewed":true,"share":[0.3224,0.2366,0.4313,0.4577,0.5508,0.5212,0.6504,0.5636,0.6218],"size":1273,"slope":0.000447},{"claims":[84,125,162,179,171,184,138,101,127],"deltaPp":-1.4749,"id":60,"keywords":["cifar-10","cifar-100","imagenet","mnist","cifar10"],"label":"Image Classification Benchmarks","labelFull":"Image Classification on CIFAR-10/100, ImageNet, and MNIST","papers":1180,"papersByYear":[74,115,153,158,161,167,131,98,123],"rel":-0.7418,"reviewed":true,"share":[1.9882,2.2674,2.3567,2.0662,1.8476,1.5007,0.9362,0.5469,0.5133],"size":1271,"slope":-0.002411},{"claims":[61,105,89,159,158,156,148,176,167],"deltaPp":-0.3832,"id":61,"keywords":["style","stylization","transfer","content","font"],"label":"Arbitrary Style Transfer","labelFull":"Arbitrary Image Style Transfer with Style Disentanglement","papers":595,"papersByYear":[27,42,47,75,87,77,78,80,82],"rel":-0.5283,"reviewed":true,"share":[0.7254,0.8281,0.724,0.9808,0.9984,0.6919,0.5575,0.4464,0.3422],"size":1219,"slope":-0.00055},{"claims":[1,0,0,5,35,172,289,339,370],"deltaPp":1.2293,"id":62,"keywords":["prompt","tuning","text","learning"],"label":"Prompt Learning","labelFull":"Prompt Learning and Tuning for Large Language Models","papers":881,"papersByYear":[1,0,0,3,22,100,204,250,301],"rel":45.6989,"reviewed":true,"share":[0.0269,0.0,0.0,0.0392,0.2525,0.8986,1.458,1.3951,1.2562],"size":1211,"slope":0.002146},{"claims":[74,84,53,78,71,117,156,266,297],"deltaPp":-0.1453,"id":63,"keywords":["question","answering","visual","video","pairs"],"label":"Visual Question Answering","labelFull":"Visual Question Answering for Images, Diagrams, and Videos","papers":822,"papersByYear":[43,43,29,45,45,65,109,201,242],"rel":-0.1258,"reviewed":true,"share":[1.1553,0.8478,0.4467,0.5885,0.5164,0.5841,0.779,1.1217,1.01],"size":1196,"slope":0.00015},{"claims":[187,242,202,208,158,91,46,26,25],"deltaPp":-2.254,"id":64,"keywords":["gan","generator","discriminator","wasserstein","conditional"],"label":"GAN Training Stabilization","labelFull":"Techniques for Stabilizing and Improving GAN Training","papers":628,"papersByYear":[87,117,103,109,91,54,25,22,20],"rel":-0.9643,"reviewed":true,"share":[2.3375,2.3068,1.5866,1.4254,1.0443,0.4853,0.1787,0.1228,0.0835],"size":1185,"slope":-0.003221},{"claims":[96,151,100,81,90,121,174,198,172],"deltaPp":-0.6634,"id":65,"keywords":["captioning","image","video","dense","visual"],"label":"Image and Video Captioning","labelFull":"Image and Video Captioning: Diverse, Controllable, Zero-Shot","papers":600,"papersByYear":[41,59,39,39,43,58,100,116,105],"rel":-0.6022,"reviewed":true,"share":[1.1016,1.1632,0.6007,0.51,0.4935,0.5212,0.7147,0.6473,0.4382],"size":1183,"slope":-0.00066},{"claims":[1,0,1,1,8,14,121,422,612],"deltaPp":1.8219,"id":66,"keywords":["multimodal","reasoning","visual","understanding","capabilities"],"label":"Multimodal LLM Reasoning","labelFull":"Multimodal Large Language Model Reasoning and Benchmarking","papers":839,"papersByYear":[1,0,1,1,5,11,91,286,443],"rel":67.7286,"reviewed":true,"share":[0.0269,0.0,0.0154,0.0131,0.0574,0.0988,0.6504,1.596,1.8488],"size":1180,"slope":0.002239},{"claims":[66,76,93,86,98,193,166,194,196],"deltaPp":-0.17,"id":67,"keywords":["hand","pose","tactile","3d","contact"],"label":"Hand-Object Interaction Pose Estimation","labelFull":"3D Hand-Object Interaction Pose Estimation, Reconstruction","papers":469,"papersByYear":[20,25,34,34,42,68,72,86,88],"rel":-0.3164,"reviewed":true,"share":[0.5373,0.4929,0.5237,0.4446,0.482,0.6111,0.5146,0.4799,0.3673],"size":1168,"slope":-9.5e-05},{"claims":[63,95,117,124,127,158,171,148,159],"deltaPp":-0.8629,"id":68,"keywords":["kernel","ridge","regression","loss","surrogate"],"label":"Kernel Surrogate Loss","labelFull":"Kernel Methods with Flexible Surrogate Loss Functions","papers":821,"papersByYear":[52,66,76,95,95,100,106,103,128],"rel":-0.6176,"reviewed":true,"share":[1.3971,1.3013,1.1707,1.2423,1.0902,0.8986,0.7576,0.5748,0.5342],"size":1162,"slope":-0.001133},{"claims":[15,20,9,16,21,53,112,240,608],"deltaPp":1.2766,"id":69,"keywords":["human","review","evaluation","llm","judgments"],"label":"LLM Evaluation","labelFull":"Human and Model-Based Evaluation of Large Language Models","papers":775,"papersByYear":[14,18,9,15,21,46,83,173,396],"rel":3.3943,"reviewed":true,"share":[0.3761,0.3549,0.1386,0.1962,0.241,0.4134,0.5932,0.9654,1.6527],"size":1094,"slope":0.001344},{"claims":[78,96,100,113,114,164,103,155,157],"deltaPp":-0.6621,"id":70,"keywords":["clustering","k-means","algorithm","approximation","fair"],"label":"Efficient Large-Scale Clustering","labelFull":"Efficient Large-Scale Clustering Algorithms","papers":532,"papersByYear":[38,48,42,63,55,73,51,76,86],"rel":-0.6485,"reviewed":true,"share":[1.021,0.9464,0.647,0.8239,0.6312,0.656,0.3645,0.4241,0.3589],"size":1080,"slope":-0.000825},{"claims":[71,99,109,113,137,132,111,157,101],"deltaPp":-0.8734,"id":71,"keywords":["flow","optical","scene","motion","estimation"],"label":"Optical Flow Estimation","labelFull":"Optical Flow and Scene Flow Estimation","papers":599,"papersByYear":[44,62,62,56,74,70,66,91,74],"rel":-0.7388,"reviewed":true,"share":[1.1822,1.2224,0.955,0.7323,0.8492,0.629,0.4717,0.5078,0.3088],"size":1030,"slope":-0.001118},{"claims":[18,26,66,77,96,103,124,203,316],"deltaPp":0.7775,"id":72,"keywords":["trajectory","prediction","future","forecasting","state-of-the-art"],"label":"Trajectory Prediction","labelFull":"Trajectory Prediction for Autonomous Driving","papers":737,"papersByYear":[13,19,39,48,55,69,88,136,270],"rel":2.2259,"reviewed":true,"share":[0.3493,0.3746,0.6007,0.6277,0.6312,0.6201,0.6289,0.7589,1.1268],"size":1029,"slope":0.000719},{"claims":[2,1,10,24,38,61,209,293,383],"deltaPp":0.5156,"id":73,"keywords":["watermark","robustness","embeds","image","attacks"],"label":"Robust AI Watermarking","labelFull":"Robust Watermarking for AI-Generated Images, Video, and Text","papers":349,"papersByYear":[1,1,4,9,13,22,68,101,130],"rel":19.1673,"reviewed":true,"share":[0.0269,0.0197,0.0616,0.1177,0.1492,0.1977,0.486,0.5636,0.5425],"size":1021,"slope":0.00077},{"claims":[117,126,139,141,96,68,88,121,111],"deltaPp":-1.1652,"id":74,"keywords":["person","re-identification","identity","visible-infrared","unsupervised"],"label":"Person Re-Identification","labelFull":"Person Re-Identification (Image, Video, Cross-Modality)","papers":507,"papersByYear":[53,62,65,70,44,36,50,65,62],"rel":-0.8183,"reviewed":true,"share":[1.424,1.2224,1.0012,0.9154,0.5049,0.3235,0.3573,0.3627,0.2588],"size":1007,"slope":-0.00152},{"claims":[53,108,96,122,91,128,146,151,110],"deltaPp":-0.5055,"id":75,"keywords":["illumination","light","reflection","imaging","polarization"],"label":"Intrinsic Image Decomposition","labelFull":"Intrinsic Image Decomposition and Relighting","papers":592,"papersByYear":[30,52,46,65,53,77,97,100,72],"rel":-0.6272,"reviewed":true,"share":[0.806,1.0252,0.7086,0.85,0.6082,0.6919,0.6933,0.558,0.3005],"size":1005,"slope":-0.000602},{"claims":[7,21,37,42,46,41,66,214,490],"deltaPp":1.2244,"id":76,"keywords":["flow","matching","velocity","normalizing","rectified"],"label":"Flow Matching","labelFull":"Flow Matching for Training Normalizing Flows","papers":639,"papersByYear":[6,15,28,24,36,28,41,129,332],"rel":7.5955,"reviewed":true,"share":[0.1612,0.2957,0.4313,0.3138,0.4131,0.2516,0.293,0.7199,1.3856],"size":964,"slope":0.000972},{"claims":[40,50,88,110,127,117,135,126,163],"deltaPp":-0.1497,"id":77,"keywords":["fairness","group","demographic","parity","constraints"],"label":"Fairness-Accuracy Trade-Offs","labelFull":"Fairness-Accuracy Trade-Off Analysis and Pareto Frontiers","papers":472,"papersByYear":[18,26,36,52,62,58,79,61,80],"rel":-0.3096,"reviewed":true,"share":[0.4836,0.5126,0.5545,0.68,0.7115,0.5212,0.5646,0.3404,0.3339],"size":956,"slope":-0.000209},{"claims":[16,40,79,147,162,170,123,115,92],"deltaPp":-0.0029,"id":78,"keywords":["contrastive","augmentation","prototypes","learning","loss"],"label":"Augmented Contrastive Learning","labelFull":"Data Augmentation for Contrastive Self‑Supervised Learning","papers":730,"papersByYear":[13,31,47,115,122,123,102,94,83],"rel":-0.0083,"reviewed":true,"share":[0.3493,0.6112,0.724,1.5039,1.4,1.1053,0.729,0.5246,0.3464],"size":944,"slope":-0.00011},{"claims":[25,44,57,71,99,143,130,154,197],"deltaPp":0.0631,"id":79,"keywords":["equivariant","group","symmetries","lie","invariant"],"label":"Group Equivariant Neural Networks","labelFull":"Group Equivariant Neural Networks and Steerable CNNs","papers":571,"papersByYear":[18,25,31,42,62,86,81,95,131],"rel":0.1305,"reviewed":true,"share":[0.4836,0.4929,0.4775,0.5492,0.7115,0.7728,0.5789,0.5301,0.5467],"size":920,"slope":0.000132},{"claims":[29,46,93,89,121,143,99,121,133],"deltaPp":-0.152,"id":80,"keywords":["games","nash","equilibrium","zero-sum","equilibria"],"label":"Equilibrium Computation in Games","labelFull":"Algorithms for Nash and Correlated Equilibria in Games","papers":447,"papersByYear":[17,23,42,49,58,66,58,61,73],"rel":-0.3328,"reviewed":true,"share":[0.4567,0.4535,0.647,0.6408,0.6656,0.5931,0.4145,0.3404,0.3047],"size":874,"slope":-0.000243},{"claims":[20,41,71,61,81,113,105,134,242],"deltaPp":0.2702,"id":81,"keywords":["transport","optimal","unbalanced","entropic","plans"],"label":"Scalable Optimal Transport","labelFull":"Scalable Optimal Transport Algorithms for Machine Learning","papers":491,"papersByYear":[12,21,38,35,43,63,61,76,142],"rel":0.8381,"reviewed":true,"share":[0.3224,0.414,0.5853,0.4577,0.4935,0.5661,0.436,0.4241,0.5926],"size":868,"slope":0.000153},{"claims":[3,7,16,36,58,86,174,198,247],"deltaPp":0.6456,"id":82,"keywords":["transformers","trained","architectures","layers","linear"],"label":"Efficient Transformers","labelFull":"Efficient Transformers: Sparse, Lightweight, Fast Training","papers":582,"papersByYear":[3,6,11,29,47,60,115,137,174],"rel":8.0099,"reviewed":true,"share":[0.0806,0.1183,0.1694,0.3792,0.5394,0.5392,0.8219,0.7645,0.7262],"size":825,"slope":0.000998},{"claims":[24,70,51,65,83,95,139,200,86],"deltaPp":-0.072,"id":83,"keywords":["inpainting","outpainting","image","video","regions"],"label":"Image Inpainting","labelFull":"Image, Video, and 3D Inpainting (including Outpainting)","papers":455,"papersByYear":[12,35,27,34,45,56,79,107,60],"rel":-0.2233,"reviewed":true,"share":[0.3224,0.6901,0.4159,0.4446,0.5164,0.5032,0.5646,0.5971,0.2504],"size":813,"slope":-3.5e-05},{"claims":[36,51,67,79,104,101,104,163,101],"deltaPp":-0.1048,"id":84,"keywords":["compression","rate-distortion","image","jpeg","video"],"label":"Learned Image Compression","labelFull":"Learned Image and Video Compression","papers":402,"papersByYear":[14,24,32,37,45,51,53,81,65],"rel":-0.2786,"reviewed":true,"share":[0.3761,0.4732,0.4929,0.4838,0.5164,0.4583,0.3788,0.452,0.2713],"size":806,"slope":-0.000123},{"claims":[51,44,77,66,96,103,109,116,140],"deltaPp":-0.4563,"id":85,"keywords":["multi-task","transfer learning","multitask","task-specific","generalization"],"label":"Parameter-Efficient Multi-Task Learning","labelFull":"Parameter-Efficient Multi-Task Learning for Scalable Transfer","papers":627,"papersByYear":[35,37,56,54,76,83,83,87,116],"rel":-0.4852,"reviewed":true,"share":[0.9404,0.7295,0.8626,0.7062,0.8722,0.7459,0.5932,0.4855,0.4841],"size":802,"slope":-0.000509},{"claims":[0,2,2,4,7,5,108,272,401],"deltaPp":0.8889,"id":86,"keywords":["hallucination","vision-language","reduces","multimodal","decoding"],"label":"Vision-Language Hallucination Mitigation","labelFull":"Vision-Language Hallucination Detection and Mitigation","papers":420,"papersByYear":[0,2,2,3,5,5,56,134,213],"rel":null,"reviewed":true,"share":[0.0,0.0394,0.0308,0.0392,0.0574,0.0449,0.4002,0.7478,0.8889],"size":801,"slope":0.001071},{"claims":[9,33,52,57,53,97,117,116,254],"deltaPp":0.6114,"id":87,"keywords":["uncertainty","calibration","epistemic","quantification","aleatoric"],"label":"Uncertainty Quantification","labelFull":"Predictive Uncertainty Quantification and Calibration","papers":613,"papersByYear":[8,20,42,46,42,81,89,87,198],"rel":2.845,"reviewed":true,"share":[0.2149,0.3943,0.647,0.6015,0.482,0.7279,0.6361,0.4855,0.8263],"size":788,"slope":0.000471},{"claims":[8,6,13,12,11,16,120,279,322],"deltaPp":0.7922,"id":88,"keywords":["preference","optimization","direct","user","dpo"],"label":"Direct Preference Optimization","labelFull":"Direct Preference Optimization and Variants for LLM Alignment","papers":527,"papersByYear":[5,4,10,8,9,10,74,185,222],"rel":5.8987,"reviewed":true,"share":[0.1343,0.0789,0.154,0.1046,0.1033,0.0899,0.5289,1.0324,0.9265],"size":787,"slope":0.001127},{"claims":[0,0,0,21,33,49,123,234,320],"deltaPp":0.9933,"id":89,"keywords":["tokens","visual","vision","image","pruning"],"label":"Vision Token Pruning","labelFull":"Vision Token Pruning for Efficient Vision Transformers","papers":569,"papersByYear":[0,0,0,15,28,40,89,159,238],"rel":null,"reviewed":true,"share":[0.0,0.0,0.0,0.1962,0.3213,0.3595,0.6361,0.8873,0.9933],"size":780,"slope":0.001345},{"claims":[17,19,46,57,89,125,168,138,93],"deltaPp":-0.0016,"id":90,"keywords":["out-of-distribution","ood","detection","outlier","in-distribution"],"label":"Out-of-Distribution Detection","labelFull":"Out-of-Distribution Detection for Deep Neural Networks","papers":425,"papersByYear":[10,13,28,37,44,70,86,73,64],"rel":-0.006,"reviewed":true,"share":[0.2687,0.2563,0.4313,0.4838,0.5049,0.629,0.6146,0.4074,0.2671],"size":752,"slope":0.00016},{"claims":[20,23,28,51,50,87,93,176,221],"deltaPp":0.216,"id":91,"keywords":["multi-view","clustering","cross-view","incomplete","contrastive"],"label":"Deep Multi-View Clustering","labelFull":"Deep Multi-View Clustering With Contrastive And Missing Views","papers":428,"papersByYear":[12,15,19,31,33,49,50,90,129],"rel":0.67,"reviewed":true,"share":[0.3224,0.2957,0.2927,0.4054,0.3787,0.4403,0.3573,0.5022,0.5384],"size":749,"slope":0.000275},{"claims":[15,20,34,30,43,57,87,193,257],"deltaPp":0.5986,"id":92,"keywords":["inference","speed","speedup","latency","faster"],"label":"Efficient Inference","labelFull":"Efficient Inference: Speed, Latency, and Memory Optimization","papers":691,"papersByYear":[15,17,33,29,41,51,83,182,240],"rel":1.4854,"reviewed":true,"share":[0.403,0.3352,0.5083,0.3792,0.4705,0.4583,0.5932,1.0156,1.0016],"size":736,"slope":0.000781},{"claims":[29,53,78,96,77,114,120,103,65],"deltaPp":-0.1411,"id":93,"keywords":["garment","clothing","mesh","human","body"],"label":"Clothed Human 3D Reconstruction","labelFull":"3D Reconstruction and Modeling of Clothed Humans","papers":383,"papersByYear":[11,29,32,45,42,62,66,59,37],"rel":-0.4775,"reviewed":true,"share":[0.2955,0.5718,0.4929,0.5885,0.482,0.5572,0.4717,0.3292,0.1544],"size":735,"slope":-0.000228}],"totals":{"claims":[17801,24336,31263,37099,42476,54300,68471,89898,124244],"papers":[3722,5072,6492,7647,8714,11128,13992,17920,23961]},"years":[2018,2019,2020,2021,2022,2023,2024,2025,2026]}