PAPER EVIDENCE RANKING
基于PaperMiner证据链动态评分。缺失证据保持未知，不等同于零分；不同任务论文不宜只凭总分直接比较。
| 排名 | 学校 | 得分 | 论文数 | 覆盖率 |
|---|---|---|---|---|
| 1 | Zhejiang University RealHD: A High-Quality Dataset for Robust Detection of State-of-the-Art AI-Generated Images · NeurIDA: Dynamic Modeling for Effective In-Database Analytics · Generative Video Matting | 64.72 | 246 | 100% |
| 2 | University of Science and Technology of China The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · AIM 2025 challenge on Inverse Tone Mapping Report: Methods and Results · InstructTime $^{++}$ : Time Series Classification with Multimodal Language Modeling via Implicit Feature Enhancement | 64.64 | 183 | 100% |
| 3 | National University of Singapore Automated high speed 3D imaging of organoid cultures with efficient multi-scale phenotypic quantification · Factorized Learning for Temporally Grounded Video-Language Models · NeurIDA: Dynamic Modeling for Effective In-Database Analytics | 64.64 | 147 | 100% |
| 4 | 上海交通大学 The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · AnchorSync: Global Consistency Optimization for Long Video Editing · GraphDLG: Exploring Deep Leakage from Gradients in Federated Graph Learning | 64.54 | 274 | 100% |
| 5 | Peking University Efficient Content-based Recommendation Model Training via Noise-aware Coreset Selection · FilDeep: Learning Large Deformations of Elastic-Plastic Solids with Multi-Fidelity Data · Graph Retrieval-Augmented Generation: A Survey | 64.42 | 227 | 99% |
| 6 | Tsinghua University The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · FuzzySQL: Uncovering Hidden Vulnerabilities in DBMS Special Features with LLM-Driven Fuzzing · PCIE Pose Solution for EgoExo4D Pose and Proficiency Estimation Challenge | 64.25 | 321 | 100% |
| 7 | The Hong Kong University of Science and Technology Enhancing AIGC Service Efficiency with Adaptive Multi-Edge Collaboration in A Distributed System · Reflexa: Uncovering How LLM-Supported Reflection Scaffolding Reshapes Creativity in Creative Coding · Relit-LiVE: Relight Video by Jointly Learning Environment Video | 64.11 | 162 | 98% |
| 8 | University of California DeltaDorsal: Enhancing Hand Pose Estimation with Dorsal Features in Egocentric Views · Adaptive Wavelet Thresholding for Image Denoising and Compression · FAIRCAUSESYN: TOWARDS CAUSALLY FAIR LLM-AUGMENTED SYNTHETIC DATA GENERATION | 64.06 | 223 | 98% |
| 9 | Harbin Institute of Technology NTIRE 2026 Challenge on Bitstream-Corrupted Video Restoration: Methods and Results · Dependency-Guided Repository-Level C-to-Rust Translation with Reinforcement Alignment · Beyond Degradation Redundancy: Contrastive Prompt Learning for All-in-One Image Restoration | 63.83 | 119 | 97% |
| 10 | Indian Institute of Technology The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · NTIRE 2026 Challenge on Bitstream-Corrupted Video Restoration: Methods and Results · Melting upon cooling in a quantum magnet | 63.67 | 76 | 96% |
| 11 | Carnegie Mellon University Hearsay: Vision-Language Medical Diagnoses Without an Image · Non-local Neural Networks · Hearsay: Vision-Language Medical Diagnoses Without an Image | 63.66 | 105 | 97% |
| 12 | University of Chinese Academy of Sciences PCIE Pose Solution for EgoExo4D Pose and Proficiency Estimation Challenge · Relit-LiVE: Relight Video by Jointly Learning Environment Video · PerlAD: Towards Enhanced Closed-loop End-to-end Autonomous Driving with Pseudo-simulation-based Reinforcement Learning | 63.54 | 139 | 97% |
| 13 | University of Wurzburg The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · AIM 2025 Low-light RAW Video Denoising Challenge: Dataset, Methods and Results · Efficient and Accurate Quantized Image Super-Resolution on Mobile NPUs, Mobile AI & AIM 2022 challenge: Report | 63.53 | 18 | 97% |
| 14 | University of Electronic Science and Technology of China DRNet: All-in-One Image Restoration via Prior-Guided Dynamic Reparameterization · The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · AIM 2025 challenge on Inverse Tone Mapping Report: Methods and Results | 63.49 | 67 | 96% |
| 15 | Wuhan University Universally Unfiltered and Unseen: Input-Agnostic Multimodal Jailbreaks against Text-to-Image Model Safeguards · C2|Q⟩: A Robust Framework for Bridging Classical and Quantum Software Development — RCR Report · Towards Perfection: Building Inter-component Mutual Correction for Retinex-based Low-light Image Enhancement | 63.47 | 78 | 96% |
| 16 | Northeastern University The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · Methodological Frontiers in 21-cm Intensity Mapping: the Treatment of Systematics and Foreground Contamination · Predicting Conflict Impact on Performance in O-RAN | 63.43 | 69 | 96% |
| 17 | Technical University of Munich VariViT: A Vision Transformer for Variable Image Sizes · MT4G: A Tool for Reliable Auto-Discovery of NVIDIA and AMD GPU Compute and Memory Topologies · Disentangling Progress in Medical Image Registration: Beyond Trend-Driven Architectures towards Domain-Specific Strategies | 63.42 | 86 | 96% |
| 18 | Nanyang Technological University Temporal Restoration and Spatial Rewiring for Source-Free Multivariate Time Series Domain Adaptation · NTIRE 2026 Challenge on Short-form UGC Video Restoration in the Wild with Generative Models: Datasets, Methods and Results · Inclusion 2024 $\cdot$ Global Multimedia Deepfake Detection Challenge: Towards Multi-dimensional Face Forgery Detection | 63.4 | 165 | 96% |
| 19 | City University of Hong Kong EHPE: Skeleton Cues-based Gaussian Coordinate Encoding for Efficient Human Pose Estimation · DALD-PCAC: Density-Adaptive Learning Descriptor for Point Cloud Lossless Attribute Compression · InterFlow: Designing Unobtrusive AI to Empower Interviewers in Semi-Structured Interviews | 63.36 | 70 | 96% |
| 20 | Fudan University The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · Mimir: Hierarchical Goal-Driven Diffusion with Uncertainty Propagation for End-to-End Autonomous Driving · SEW: Strengthening Robustness of Black-box DNN Watermarking via Specificity Enhancement | 63.35 | 154 | 98% |
| 21 | Nanjing University The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · Relit-LiVE: Relight Video by Jointly Learning Environment Video · Towards Perfection: Building Inter-component Mutual Correction for Retinex-based Low-light Image Enhancement | 63.23 | 89 | 96% |
| 22 | The Hong Kong Polytechnic University High-resolution Photo Enhancement in Real-time: A Laplacian Pyramid Network · The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · CheatAgent: Attacking LLM-Empowered Recommender Systems via LLM Agent | 63.23 | 66 | 95% |
| 23 | The Chinese University of Hong Kong Relit-LiVE: Relight Video by Jointly Learning Environment Video · Unifying Physically-Informed Weather Priors in A Single Model for Image Restoration Across Multiple Adverse Weather Conditions · Dependency-Guided Repository-Level C-to-Rust Translation with Reinforcement Alignment | 63.21 | 145 | 96% |
| 24 | Johns Hopkins University Feedback by Design: Understanding and Overcoming User Feedback Barriers in Conversational Agents · Asymmetric Aerosol Distribution on the Terminators of the Warm Saturn WASP-69 b Revealed by JWST NIRISS/SOSS · Privy: From Fine Print to Fair Practice in Privacy Rights Exercise | 63.2 | 73 | 95% |
| 25 | Georgia Institute of Technology TerraFormer: Automated Infrastructure-as-Code with LLMs Fine-Tuned via Policy-Guided Verifier Feedback · Explainable Model Routing for Agentic Workflows · What Must Generalist Agents Remember? | 63.18 | 57 | 96% |
| 26 | Beihang University GraphDLG: Exploring Deep Leakage from Gradients in Federated Graph Learning · NTIRE 2026 Challenge on Short-form UGC Video Restoration in the Wild with Generative Models: Datasets, Methods and Results · Low Light Image Enhancement Challenge at NTIRE 2026 | 63.15 | 84 | 96% |
| 27 | Hefei University of Technology The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · PSQE: A Theoretical-Practical Approach to Pseudo Seed Quality Enhancement for Unsupervised Multimodal Entity Alignment · Dynamic Spectral Denoising with Global-Context Attention for Multi-Behavior Recommendation | 63.12 | 31 | 95% |
| 28 | Nanjing University of Science and Technology The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · AIM 2025 challenge on Inverse Tone Mapping Report: Methods and Results · LoBoFit: Flexible Garment Refitting via Local Bone Mapping Blending | 63.09 | 44 | 94% |
| 29 | Massachusetts Institute of Technology Gesturing Toward Abstraction: Multimodal Convention Formation in Collaborative Physical Tasks · Dark matter searches with a 13 meV threshold superconducting sensor array · The limits of fair medical imaging AI in real-world generalization | 63.06 | 138 | 94% |
| 30 | University of Illinois Urbana-Champaign NTIRE 2026 Challenge on Bitstream-Corrupted Video Restoration: Methods and Results · "I Don’t Trust Any Professional Research Tool": A Re-Imagination of Knowledge Production Workflows by, with, and for Blind and Low-Vision Researchers · Job Anxiety in Post-Secondary Computer Science Students Caused by Artificial Intelligence | 63.06 | 65 | 94% |
| 31 | University of Toronto Detection and Evolution of Linear Polarization of the Galactic Center Transient MAXI J1744-294 · Job Anxiety in Post-Secondary Computer Science Students Caused by Artificial Intelligence · LoRAFusion: Efficient LoRA Fine-Tuning for LLMs | 63.05 | 70 | 94% |
| 32 | Beijing Institute of Technology Hi-LOAM: Hierarchical Implicit Neural Fields for LiDAR Odometry and Mapping · Single-shot 3D imaging with point cloud projection based on metadevice · AIM 2025 Low-light RAW Video Denoising Challenge: Dataset, Methods and Results | 62.99 | 63 | 94% |
| 33 | Xidian University NTIRE 2026 Challenge on Bitstream-Corrupted Video Restoration: Methods and Results · ANCHOR: Agentic Noise Creation Framework for Human Simulation and Denoising Recommendation · Group-of-Latents: Perceptual Video Compression at Extreme Bitrates via Masked Latent Generative Modeling | 62.99 | 50 | 94% |
| 34 | Stanford University Gesturing Toward Abstraction: Multimodal Convention Formation in Collaborative Physical Tasks · Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank · Physical Object Understanding with a Physically Controllable World Model | 62.96 | 142 | 95% |
| 35 | Sun Yat-sen University CoCo-BERT: Improving Video-Language Pre-training with Contrastive Cross-modal Matching and Denoising · NTIRE 2026 Challenge on Short-form UGC Video Restoration in the Wild with Generative Models: Datasets, Methods and Results · Inclusion 2024 $\cdot$ Global Multimedia Deepfake Detection Challenge: Towards Multi-dimensional Face Forgery Detection | 62.92 | 76 | 94% |
| 36 | University of Michigan PRIVATEEDIT: A Privacy-Preserving Pipeline for Face-Centric Generative Image Editing · L3GS: Layered 3D Gaussian Splats for Efficient 3D Scene Delivery · Auditorily Embodied Conversational Agents: Effects of Spatialization and Situated Audio Cues on Presence and Social Perception | 62.92 | 75 | 95% |
| 37 | University College AI Content Moderation, Racism and (de)Coloniality · Compositional Segmentation of Cardiac Images Leveraging Metadata · From Research to Practice: An Interactive Rapid Review of Autonomous Driving System Testing in Industry | 62.9 | 80 | 94% |
| 38 | Macao Polytechnic University MoGen: A Unified Collaborative Framework for Controllable Multi-Object Image Generation · Decoding the Sequence Determinants of Locus-Specific DNA Methylation Across Human Tissues · The Tenth NTIRE 2025 Image Denoising Challenge Report | 62.89 | 10 | 96% |
| 39 | Huazhong University of Science and Technology High-resolution Photo Enhancement in Real-time: A Laplacian Pyramid Network · Source-Free Domain Adaptation (SFDA) for Privacy-Preserving Seizure Subtype Classification · Relit-LiVE: Relight Video by Jointly Learning Environment Video | 62.88 | 65 | 93% |
| 40 | Sichuan University MoGen: A Unified Collaborative Framework for Controllable Multi-Object Image Generation · LoViF 2026 Challenge on Real-World All-in-One Image Restoration: Methods and Results · Dual-Representation Interaction Driven Image Quality Assessment with Restoration Assistance | 62.83 | 38 | 93% |
| 41 | University of Trento The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · AIM 2025 challenge on Inverse Tone Mapping Report: Methods and Results · Feature Slice Matching for Precise Bug Detection | 62.83 | 27 | 95% |
| 42 | Tongji University Bridging Cognitive Neuroscience and Graph Intelligence: Hippocampus-Inspired Multi-View Hypergraph Learning for Web Finance Fraud · RAW Image Reconstruction from RGB on Smartphones. NTIRE 2025 Challenge Report · NTIRE 2026 The 3rd Restore Any Image Model (RAIM) Challenge: Professional Image Quality Assessment (Track 1) | 62.82 | 61 | 94% |
| 43 | The University of Tokyo A roadmap for AI in robotics · GazeLLM: Multimodal LLMs incorporating Human Visual Attention · Realtime Multimodal Emotion Estimation using Behavioral and Neurophysiological Data | 62.77 | 69 | 94% |
| 44 | Monash University Relightable and Dynamic Gaussian Avatar Reconstruction from Monocular Video · Reporting LLM Prompting in Automated Software Engineering: A Guideline Based on Current Practices and Expectations · Human Factors in Immersive Analytics | 62.77 | 48 | 94% |
| 45 | Southeast University Dependency-Guided Repository-Level C-to-Rust Translation with Reinforcement Alignment · Cognitive Disentanglement for Referring Multi-Object Tracking · Towards Perfection: Building Inter-component Mutual Correction for Retinex-based Low-light Image Enhancement | 62.72 | 67 | 94% |
| 46 | University of Oxford Asymmetric Aerosol Distribution on the Terminators of the Warm Saturn WASP-69 b Revealed by JWST NIRISS/SOSS · Red Teaming Quantum-Resistant Cryptographic Standards: A Penetration Testing Framework Integrating AI and Quantum Security · The second H.E.S.S. gamma-ray burst catalogue: 15 years of observations with the H.E.S.S. telescopes | 62.7 | 84 | 93% |
| 47 | Seoul National University AIM 2025 Low-light RAW Video Denoising Challenge: Dataset, Methods and Results · Efficient and Accurate Quantized Image Super-Resolution on Mobile NPUs, Mobile AI & AIM 2022 challenge: Report · Low Light Image Enhancement Challenge at NTIRE 2026 | 62.65 | 62 | 94% |
| 48 | University of Pisa The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report · AIM 2025 challenge on Inverse Tone Mapping Report: Methods and Results · Evaluating LLM-Based Mobile App Recommendations: An Empirical Study | 62.61 | 17 | 94% |
| 49 | University of Maryland Detection and Evolution of Linear Polarization of the Galactic Center Transient MAXI J1744-294 · Designing Multi-Robot Ground Video Sensemaking with Public Safety Professionals · Designing Multi-Robot Ground Video Sensemaking with Public Safety Professionals | 62.6 | 60 | 94% |
| 50 | University of Cambridge Detection and Evolution of Linear Polarization of the Galactic Center Transient MAXI J1744-294 · Methodological Frontiers in 21-cm Intensity Mapping: the Treatment of Systematics and Foreground Contamination · ALPaCA: Adapting Llama for Pathology Context Analysis to enable slide-level question answering | 62.58 | 88 | 93% |
排名用于发现值得进一步核查的论文，不替代同行评审。评分会随新增引用、复现、实验验证、新闻评价和落地证据持续更新。
PaperMiner MCP已开放结构化论文反馈。可提交复现成功或失败、代码可运行或失效、Benchmark纠错、链接纠错和部署结果。
MCP接口：submit_paper_feedback、get_paper_community_feedback
权限：有效PaperMiner账号获配的个人MCP密钥可提交；公共代理和共享只读密钥只能查询。每个账号每天最多提交5条新反馈。