pair_id string | source string | ai_paper_id string | human_paper_id string | ai_title string | human_title string | human_arxiv string | human_venue string | human_v1_year int64 | human_iclr_rating float64 | contribution_type string | human_contribution_type string | same_contribution_type bool | pool string | sim_tier int64 | relation string | tier_reason string | H1_human string | H1_by string | H2_top_tier string | H3_type_match string | H4_source string | specter2_cosine float64 | tfidf_similarity float64 | match_rank int64 | human_pangram_ai_fraction float64 | ai_body_words int64 | human_body_words int64 | body_length_ratio float64 | topic string | secondary_topic string | a4s_submission_id string | a4s_openreview_url string | a4s_decision string | a4s_reviewer_scores string | a4s_ai_involvement string | ai_tex_source string |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
A4S0224 | agents4science | AI_A4S0224 | HU_2601.19204 | From Monoliths to Pharmacists-at-Scale: Patient-Aware Multi-Agent Reasoning Tames Million-Dimensional Discovery | MATA: A Trainable Hierarchical Automaton System for Multi-Agent Visual Reasoning | 2601.19204 | ICLR 2026 poster | null | 4 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | hierarchical multi-agent system that decides when to collaborate, the same orchestration contribution | pass | pangram | pass | pass | pass | null | 0.101 | null | 0 | 2,299 | 3,750 | 1.46 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 88lrbJ8PTe | https://openreview.net/forum?id=88lrbJ8PTe | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0225 | agents4science | AI_A4S0225 | HU_2506.22095 | MosquitoSwarm: Bio-Inspired Collective Intelligence for Multi-Objective Optimization in Computational Sciences | Beyond Simple Graphs: Neural Multi-Objective Routing on Multigraphs | 2506.22095 | ICLR 2026 poster | null | 5.333 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | neural multi-objective optimization for routing, the same multi-objective search problem | pass | pangram | pass | pass | pass | null | 0.079 | null | 0 | 1,941 | 4,228 | 2.11 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 1g8lpVSLZ1 | https://openreview.net/forum?id=1g8lpVSLZ1 | rejected | {"AIRev1": "2.0", "AIRev2": "3.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0226 | agents4science | AI_A4S0226 | HU_2305.10601 | Beyond Chain-of-Thought: Theory-Grounded Approaches to Elicit Deep Reasoning in LLMs | Tree of Thoughts: Deliberate Problem Solving with Large Language Models | 2305.10601 | NeurIPS 2023 | 2,023 | null | method | method | true | cited | 2 | cited | structured reasoning prompting, the paradigm the paper argues against | pass | year<=2024 | pass | pass | pass | null | null | null | null | 3,060 | 4,666 | 1.49 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | Jfx32wFUXq | https://openreview.net/forum?id=Jfx32wFUXq | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "A", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0227 | agents4science | AI_A4S0227 | HU_1805.01070 | On the Failure of a Universal Linear Representation Hypothesis in Deep Neural Networks | What you can cram into a single vector: Probing sentence embeddings for linguistic properties | 1805.01070 | ACL 2018 | 2,018 | null | analysis | analysis | true | cited | 2 | cited | probes what linear classifiers can read off embeddings, the same object measured empirically | pass | year<=2024 | pass | pass | pass | null | null | null | null | 5,386 | 4,882 | 1.66 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | pmfF7wwX6W | https://openreview.net/forum?id=pmfF7wwX6W | rejected | {"AIRev1": "1.0", "AIRev2": "4.0", "AIRev3": "4.0", "Human": ""} | {"hypothesis_development": "A", "experimental_design": "D", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0228 | agents4science | AI_A4S0228 | HU_2506.14245 | The Impact of Training Data Composition on Reinforcement Learning with Verifiable Rewards: Theoretical Analysis and Empirical Investigation | Reinforcement Learning with Verifiable Rewards Implicitly Incentivizes Correct Reasoning in Base LLMs | 2506.14245 | ICLR 2026 poster | null | 5.333 | analysis | analysis | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | theoretical and empirical study of what RLVR actually incentivizes, the same problem and contribution kind | pass | pangram | pass | pass | pass | null | 0.139 | null | 0.031638 | 2,637 | 3,761 | 1.41 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | GitkwMobgt | https://openreview.net/forum?id=GitkwMobgt | rejected | {"AIRev1": "2.0", "AIRev2": "4.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "A", "experimental_design": "D", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0229 | agents4science | AI_A4S0229 | HU_2506.07003 | Distribution Enforcement via Random Probe: Active Distributional Constraints for Robust Deep Learning | End-to-End Probabilistic Framework for Learning with Hard Constraints | 2506.07003 | ICLR 2026 poster | null | 6.5 | method | method | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | enforces hard constraints inside training through a differentiable projection layer, the same problem and contribution kind | pass | pangram | pass | pass | pass | null | 0.115 | null | 0 | 1,949 | 3,988 | 1.88 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | j24USL1xKd | https://openreview.net/forum?id=j24USL1xKd | rejected | {"AIRev1": "2.0", "AIRev2": "2.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "D", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0230 | agents4science | AI_A4S0230 | HU_2510.26121 | Physics-Informed Discrepancy Decomposition and Robust Astrophysical Inference for GW231123 | Uncertainty-Aware Diagnostics for Physics-Informed Machine Learning | 2510.26121 | ICLR 2026 poster | null | 6 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | uncertainty-aware diagnostics for physics-informed inference, same discrepancy-diagnosis problem | pass | pangram | pass | pass | pass | null | 0.169 | null | 0 | 4,091 | 4,069 | 0.97 | Natural Sciences | Astronomy | 8DzD2zITNi | https://openreview.net/forum?id=8DzD2zITNi | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0231 | agents4science | AI_A4S0231 | HU_1510.00149 | Next-Gen Model Compression: Tensor Program Synthesis for Beyond Pruning/Quantization Optimization | Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding | 1510.00149 | ICLR 2016 | 2,015 | null | method | method | true | cited | 3 | cited | pruning plus quantization compression pipeline, the same problem and contribution kind | pass | year<=2024 | pass | pass | pass | null | null | null | null | 1,057 | 4,168 | 3.78 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 3oCWrOf4Gj | https://openreview.net/forum?id=3oCWrOf4Gj | rejected | {"AIRev1": "2.0", "AIRev2": "2.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0232 | agents4science | AI_A4S0232 | HU_2602.02165 | Quantum Circuit Synthesis via Reinforcement Learning: Automated Design of Efficient Quantum Algorithms | AQER: A Scalable and Efficient Data Loader for Digital Quantum Computers | 2602.02165 | ICLR 2026 poster | null | 5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | efficient construction of quantum circuits under resource limits, the same synthesis-cost problem | pass | pangram | pass | pass | pass | null | 0.185 | null | 0.03043 | 681 | 4,345 | 5.96 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | Hp3rUCPI98 | https://openreview.net/forum?id=Hp3rUCPI98 | rejected | {"AIRev1": "2.0", "AIRev2": "1.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0233 | agents4science | AI_A4S0233 | HU_1609.02907 | Hypergraph Neural Networks for Complex Relational Data: Capturing Higher-Order Dependencies in Real-World Systems | Semi-Supervised Classification with Graph Convolutional Networks | 1609.02907 | ICLR 2017 | 2,016 | null | method | method | true | cited | 2 | cited | graph convolution, the architecture HyperGNN extends to hyperedges | pass | year<=2024 | pass | pass | pass | null | null | null | null | 525 | 3,425 | 6.03 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | hv0EdHgQJP | https://openreview.net/forum?id=hv0EdHgQJP | rejected | {"AIRev1": "2.0", "AIRev2": "1.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0234 | agents4science | AI_A4S0234 | HU_2511.02101 | Explicit vs Implicit Representations: A Systematic Comparison of GA-Planes, K-Planes, and NeRF for 2D Matrix Reconstruction | Measuring the Intrinsic Dimension of Earth Representations | 2511.02101 | ICLR 2026 poster | null | 5.5 | analysis | analysis | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | measures properties of implicit neural representations, the same measurement kind on a different signal | pass | pangram | pass | pass | pass | null | 0.094 | null | 0 | 2,209 | 3,844 | 1.64 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 0peaDMOMk8 | https://openreview.net/forum?id=0peaDMOMk8 | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "C", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0236 | agents4science | AI_A4S0236 | HU_2306.02561 | AI Passes Humanity’s Last Exam and Generates Video Tutorials | LLM-Blender: Ensembling Large Language Models with Pairwise Ranking and Generative Fusion | 2306.02561 | ACL 2023 | 2,023 | null | method | method | true | cited | 3 | cited | ensembling several LLMs and selecting the best output, the same problem and contribution kind | pass | year<=2024 | pass | pass | pass | null | null | null | null | 2,653 | 4,279 | 1.36 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | EJ8cQFi5cU | https://openreview.net/forum?id=EJ8cQFi5cU | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "B", "data_analysis": "B", "writing": "B"} | rebuilt_from_pdf |
A4S0237 | agents4science | AI_A4S0237 | HU_2510.08630 | Accelerating NLP for Health Equity: Fine-Tuning Binary and Multi-Class Stigma Classifiers in 48 Hours | ExPO-HM: Learning to Explain-then-Detect for Hateful Meme Detection | 2510.08630 | ICLR 2026 poster | null | 6.667 | framework | framework | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | explain-then-detect for hateful content, same harmful-language detection problem | pass | pangram | pass | pass | pass | null | 0.083 | null | 0 | 3,476 | 4,938 | 1.34 | Interdisciplinary | Computational Social Science | zP2eARUsK3 | https://openreview.net/forum?id=zP2eARUsK3 | rejected | {"AIRev1": "2.0", "AIRev2": "5.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "A", "data_analysis": "B", "writing": "C"} | rebuilt_from_pdf |
A4S0238 | agents4science | AI_A4S0238 | HU_2510.12273 | Overcoming Combinatorial Explosion in Alloy Design via Hierarchical Multi-Agent Systems | Multi-Action Self-Improvement For Neural Combinatorial Optimization | 2510.12273 | ICLR 2026 poster | null | 5.5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | self-improving search over a combinatorial space, same combinatorial explosion problem | pass | pangram | pass | pass | pass | null | 0.052 | null | 0.044302 | 2,622 | 3,873 | 1.4 | Engineering & Technology | Materials Science | kreJgMPdtf | https://openreview.net/forum?id=kreJgMPdtf | rejected | {"AIRev1": "1.0", "AIRev2": "4.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0239 | agents4science | AI_A4S0239 | HU_2504.19223 | Predictive Modeling of Grapevine Red Blotch Disease Using Multi-Temporal Remote Sensing and Spatial Epidemiology | CARL: Camera-Agnostic Representation Learning for Spectral Image Analysis | 2504.19223 | ICLR 2026 poster | null | 5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | spectral image analysis for prediction, same spectral-imaging prediction problem | pass | pangram | pass | pass | pass | null | 0.111 | null | 0 | 1,680 | 3,642 | 2.06 | Life & Health Sciences | Agriculture & Ecology | 5HqxSctSmB | https://openreview.net/forum?id=5HqxSctSmB | rejected | {"AIRev1": "2.0", "AIRev2": "4.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "C"} | rebuilt_from_pdf |
A4S0242 | agents4science | AI_A4S0242 | HU_2208.10264 | Simulating Two-Sided Job Marketplaces with AI Agents | Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studies | 2208.10264 | Accepted for oral presentation at Intern | 2,022 | null | method | method | true | cited | 3 | cited | LLMs standing in for human subjects in economic and social experiments, the same problem and contribution kind | pass | year<=2024 | pass | pass | pass | null | null | null | null | 3,730 | 5,116 | 1.32 | Social Sciences | Economics | pjpkEHH5YS | https://openreview.net/forum?id=pjpkEHH5YS | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "5.0", "Human": "5.0"} | {"hypothesis_development": "D", "experimental_design": "C", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0243 | agents4science | AI_A4S0243 | HU_2603.01630 | A Decision Matrix for Optimal Matching of Biological Systems to Microgravity Simulation Platforms | SEED-SET: Scalable Evolving Experimental Design for System-level Ethical Testing | 2603.01630 | ICLR 2026 poster | null | 4 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | scalable evolving experimental design, same problem of choosing an experimental setup systematically | pass | pangram | pass | pass | pass | null | 0.058 | null | 0.040247 | 1,707 | 4,514 | 2.52 | Life & Health Sciences | Biology | vXVQbDoYbP | https://openreview.net/forum?id=vXVQbDoYbP | rejected | {"AIRev1": "1.0", "AIRev2": "2.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "A", "data_analysis": "A", "writing": "B"} | rebuilt_from_pdf |
A4S0244 | agents4science | AI_A4S0244 | HU_2504.18579 | Information-Efficient Transformers via Adaptive Token Pruning | Sparsity Forcing: Reinforcing Token Sparsity of MLLMs | 2504.18579 | ICLR 2026 poster | null | 5 | method | method | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | token sparsity for cheaper attention, the same problem and contribution kind | pass | pangram | pass | pass | pass | null | 0.101 | null | 0 | 2,850 | 3,852 | 1.22 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | TX4BUsNGsA | https://openreview.net/forum?id=TX4BUsNGsA | rejected | {"AIRev1": "2.0", "AIRev2": "3.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "B", "data_analysis": "B", "writing": "C"} | shipped_tex |
A4S0246 | agents4science | AI_A4S0246 | HU_2511.00177 | AI-Driven Discovery of Temporal-Demographic Interactions in Emergency Department Care Delivery: AMulti-Agent Collaborative Analysis of Healthcare Equity Patterns | Can SAEs reveal and mitigate racial biases of LLMs in healthcare? | 2511.00177 | ICLR 2026 poster | null | 5.6 | method | analysis | false | filled_iclr26 | 2 | filled (nearest admissible anchor) | racial bias of LLMs in healthcare, same demographic-disparity question | pass | pangram | pass | relaxed | pass | null | 0.076 | 3 | 0 | 2,808 | 3,155 | 1.06 | Life & Health Sciences | Medicine & Health | x8R9TaMNv3 | https://openreview.net/forum?id=x8R9TaMNv3 | rejected | {"AIRev1": "2.0", "AIRev2": "1.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "C"} | rebuilt_from_pdf |
A4S0248 | agents4science | AI_A4S0248 | HU_2510.04855 | Synthetic Medical Imaging with Pathology-Aware Variational Autoencoders | Synthesising Counterfactual Explanations via Label-Conditional Gaussian Mixture Variational Autoencoders | 2510.04855 | ICLR 2026 poster | null | 4 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | class-conditional VAE synthesis with structured latents, the same generator family for a different aim | pass | pangram | pass | pass | pass | null | 0.101 | null | 0 | 859 | 4,559 | 4.25 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 7zAuPR82do | https://openreview.net/forum?id=7zAuPR82do | rejected | {"AIRev1": "2.0", "AIRev2": "2.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "B", "data_analysis": "B", "writing": "B"} | shipped_tex |
A4S0251 | agents4science | AI_A4S0251 | HU_2505.12299 | Restoring Hover on Touchscreens Using a Mouse Pointer | MobileIPL: Enhancing Mobile Agents Thinking Process via Iterative Preference Learning | 2505.12299 | ICLR 2026 poster | null | 5 | method | method | true | filled_iclr26 | 1 | filled (nearest admissible anchor) | mobile agent reasoning, mobile area but not an input technique | pass | pangram | pass | pass | pass | null | 0.061 | 2 | 0 | 3,693 | 3,757 | 0.97 | Computer & Data Sciences | Human-Computer Interaction | I9plFigPxl | https://openreview.net/forum?id=I9plFigPxl | rejected | {"AIRev1": "2.0", "AIRev2": "3.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "A", "experimental_design": "B", "data_analysis": "B", "writing": "B"} | rebuilt_from_pdf |
A4S0252 | agents4science | AI_A4S0252 | HU_2603.03511 | Ab Initio Calculation and Theoretical Construction of the First Excitation Energy of Lithium Atom | Orbital Transformers for Predicting Wavefunctions in Time-Dependent Density Functional Theory | 2603.03511 | ICLR 2026 poster | null | 6 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | predicting TDDFT wavefunctions, same electronic-excitation computation problem | pass | pangram | pass | pass | pass | null | 0.096 | null | 0 | 3,921 | 4,359 | 1.06 | Natural Sciences | Physics | 8V0TGsJHg4 | https://openreview.net/forum?id=8V0TGsJHg4 | rejected | {"AIRev1": "2.0", "AIRev2": "2.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0254 | agents4science | AI_A4S0254 | HU_2509.19877 | Ground-State Energy Calculation of Metallic Hydrogen Using Local Density Approximation (LDA) | Advancing Universal Deep Learning for Electronic-Structure Hamiltonian Prediction of Materials | 2509.19877 | ICLR 2026 poster | null | 4.5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | electronic-structure Hamiltonian prediction, same DFT energy computation problem | pass | pangram | pass | pass | pass | null | 0.096 | null | 0 | 3,744 | 4,052 | 1.05 | Natural Sciences | Physics | ZHFdsEbBVu | https://openreview.net/forum?id=ZHFdsEbBVu | rejected | {"AIRev1": "2.0", "AIRev2": "1.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0255 | agents4science | AI_A4S0255 | HU_2509.16825 | Neural Reaction-Diffusion Operators for Spatially Heterogeneous Tumor Modeling | KANO: Kolmogorov-Arnold Neural Operator | 2509.16825 | ICLR 2026 poster | null | 6 | method | method | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | neural operator architecture that fixes a spectral limitation of FNO, the same problem and contribution kind | pass | pangram | pass | pass | pass | null | 0.173 | null | 0 | 2,192 | 4,034 | 1.7 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | Jxp7TqMCAi | https://openreview.net/forum?id=Jxp7TqMCAi | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "2.0"} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0256 | agents4science | AI_A4S0256 | HU_2509.24895 | Reasoning Models Outperform Standard Language Models in De Novo Protein Design | Towards Understanding the Shape of Representations in Protein Language Models | 2509.24895 | ICLR 2026 poster | null | 6 | analysis | analysis | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | studies what protein language models encode, the same models measured on a different property | pass | pangram | pass | pass | pass | null | 0.124 | null | 0 | 1,852 | 4,094 | 1.98 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | yXYEbPQp8x | https://openreview.net/forum?id=yXYEbPQp8x | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": "4.0"} | {"hypothesis_development": "B", "experimental_design": "A", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0257 | agents4science | AI_A4S0257 | HU_2509.22341 | The Digital Inbreeding Crisis: Empirical Evidence of LLM Capability Degradation under Multi-Generational Synthetic Training | Preventing Model Collapse Under Overparametrization: Optimal Mixing Ratios for Interpolation Learning and Ridge Regression | 2509.22341 | ICLR 2026 poster | null | 5.5 | analysis | analysis | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | model collapse from repeated training on own outputs, the same hypothesis analysed | pass | pangram | pass | pass | pass | null | 0.066 | null | 0 | 2,306 | 4,252 | 1.69 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | X26I2fwt3f | https://openreview.net/forum?id=X26I2fwt3f | rejected | {"AIRev1": "2.0", "AIRev2": "5.0", "AIRev3": "4.0", "Human": ""} | {"hypothesis_development": "C", "experimental_design": "C", "data_analysis": "C", "writing": "C"} | shipped_tex |
A4S0258 | agents4science | AI_A4S0258 | HU_2508.21107 | Green by Design: Energy-Guided Reranking of LLM-Generated Programs | Learning to Generate Unit Test via Adversarial Reinforcement Learning | 2508.21107 | ICLR 2026 poster | null | 4.5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | training-time selection of generated code against an executable signal, the same measure-and-select loop | pass | pangram | pass | pass | pass | null | 0.087 | null | 0 | 3,593 | 4,272 | 1.11 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | BxscqmB9Rs | https://openreview.net/forum?id=BxscqmB9Rs | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": "4.0"} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0260 | agents4science | AI_A4S0260 | HU_1704.01701 | Interpretable by Design: Boosting Neural Network Performance with Rule-Augmented Features | Learning Certifiably Optimal Rule Lists for Categorical Data | 1704.01701 | KDD '17 | 2,017 | null | method | method | true | cited | 2 | cited | optimal rule lists, the interpretable half of the hybrid | pass | year<=2024 | pass | pass | pass | null | null | null | null | 1,759 | 16,666 | 8.55 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 73EbNDKArc | https://openreview.net/forum?id=73EbNDKArc | rejected | {"AIRev1": "2.0", "AIRev2": "5.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0262 | agents4science | AI_A4S0262 | HU_2504.15964 | Computational Analysis of Cryptographic Hash Function Performance and Security | Quantum machine learning advantages beyond hardness of evaluation | 2504.15964 | ICLR 2026 poster | null | 6 | analysis | analysis | true | filled_iclr26 | 0 | filled (nearest admissible anchor) | quantum learning theory with cryptographic labels, unrelated problem | pass | pangram | pass | relaxed | pass | null | 0.066 | 0 | 0 | 1,480 | 5,404 | 3.48 | Computer & Data Sciences | Software & Systems | Y3Mn3UkPcz | https://openreview.net/forum?id=Y3Mn3UkPcz | rejected | {"AIRev1": "1.0", "AIRev2": "2.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "B", "data_analysis": "B", "writing": "B"} | shipped_tex |
A4S0263 | agents4science | AI_A4S0263 | HU_2509.23050 | Discovering Domain-Adaptive Multimodal Design Principles Through Computational Systematic Review | Understanding Language Prior of LVLMs by Contrasting Chain-of-Embedding | 2509.23050 | ICLR 2026 poster | null | 5 | survey | analysis | false | filled_iclr26 | 1 | filled (nearest admissible anchor) | language prior in LVLMs, multimodal area only | pass | pangram | pass | relaxed | pass | null | 0.068 | 1 | 0.037798 | 2,018 | 4,616 | 2.16 | Social Sciences | Education | FYYCrBCTgk | https://openreview.net/forum?id=FYYCrBCTgk | rejected | {"AIRev1": "2.0", "AIRev2": "3.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "C"} | rebuilt_from_pdf |
A4S0264 | agents4science | AI_A4S0264 | HU_2602.08467 | Robust Time-Series Anomaly Detection for AGI System Monitoring: A Hybrid Neural-Statistical Approach | Low Rank Transformer for Multivariate Time Series Anomaly Detection and Localization | 2602.08467 | ICLR 2026 poster | null | 5.5 | method | method | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | multivariate time-series anomaly detection with theory-backed design, the same problem and contribution kind | pass | pangram | pass | pass | pass | null | 0.153 | null | 0 | 2,313 | 3,907 | 1.53 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | u4i2vyJVqe | https://openreview.net/forum?id=u4i2vyJVqe | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0265 | agents4science | AI_A4S0265 | HU_2510.22620 | The Double Helix of Productivity: A Comprehensive Framework for Understanding and Mitigating Technostress from Generative AI Agents in the Scientific Workplace | Breaking Agent Backbones: Evaluating the Security of Backbone LLMs in AI Agents | 2510.22620 | ICLR 2026 poster | null | 6 | analysis | analysis | true | filled_iclr26 | 1 | filled (nearest admissible anchor) | security of agent backbones, AI-agent area only | pass | pangram | pass | pass | pass | null | 0.087 | 2 | 0 | 3,279 | 5,655 | 1.69 | Social Sciences | Psychology | HeVPOpanJT | https://openreview.net/forum?id=HeVPOpanJT | rejected | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": "1.0"} | {"hypothesis_development": "B", "experimental_design": "A", "data_analysis": "B", "writing": "B"} | rebuilt_from_pdf |
A4S0268 | agents4science | AI_A4S0268 | HU_2509.25481 | Fairness-Aware Classification with Synthetic Tabular Data | Fair Classification by Direct Intervention on Operating Characteristics | 2509.25481 | ICLR 2026 poster | null | 5.5 | method | method | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | fair classifiers under group fairness constraints, the same problem and contribution kind | pass | pangram | pass | pass | pass | null | 0.142 | null | 0 | 1,397 | 3,448 | 2.17 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | W9TBkcaRLJ | https://openreview.net/forum?id=W9TBkcaRLJ | rejected | {"AIRev1": "2.0", "AIRev2": "5.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | shipped_tex |
A4S0269 | agents4science | AI_A4S0269 | HU_2310.08491 | Towards Automatic Evaluation and Selection of PHI De-identification Models via Multi-Agent Collaboration | Prometheus: Inducing Fine-grained Evaluation Capability in Language Models | 2310.08491 | ICLR 2024 | 2,023 | null | method | method | true | cited | 2 | cited | a language model trained to evaluate against fine-grained criteria, the same judge machinery | pass | year<=2024 | pass | pass | pass | null | null | null | null | 3,254 | 5,957 | 1.65 | Computer & Data Sciences | Human-Computer Interaction | MIjY6VNtY0 | https://openreview.net/forum?id=MIjY6VNtY0 | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "5.0"} | {"hypothesis_development": "A", "experimental_design": "C", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0272 | agents4science | AI_A4S0272 | HU_1606.06357 | Uncertainty-Guided Agents for Rare-Disease Hypothesis Discovery on Knowledge Graphs | Complex Embeddings for Simple Link Prediction | 1606.06357 | ICML 2016 | 2,016 | null | method | method | true | cited | 2 | cited | knowledge graph link prediction, the scorer inside the agent | pass | year<=2024 | pass | pass | pass | null | null | null | null | 2,273 | 4,252 | 1.74 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 2qm4sB6BbS | https://openreview.net/forum?id=2qm4sB6BbS | rejected | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": "2.0"} | {"hypothesis_development": "A", "experimental_design": "B", "data_analysis": "A", "writing": "B"} | rebuilt_from_pdf |
A4S0275 | agents4science | AI_A4S0275 | HU_1803.01498 | ChainML: Byzantine-Resilient Decentralized AI Training with Blockchain-Orchestrated Federated Learning | Byzantine-Robust Distributed Learning: Towards Optimal Statistical Rates | 1803.01498 | ICML 2018 | 2,018 | null | method | method | true | cited | 3 | cited | Byzantine-robust distributed learning with guarantees, the same problem and contribution kind | pass | year<=2024 | pass | pass | pass | null | null | null | null | 1,815 | 4,923 | 2.58 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 5qYADa1lbX | https://openreview.net/forum?id=5qYADa1lbX | rejected | {"AIRev1": "2.0", "AIRev2": "2.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0276 | agents4science | AI_A4S0276 | HU_2510.16756 | AGIArch: A Unified Hierarchical Architecture for Artificial General Intelligence | End-to-end Listen, Look, Speak and Act | 2510.16756 | ICLR 2026 poster | null | 5.5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | unified architecture integrating perception, speech and action, the closest integrated-architecture contribution | pass | pangram | pass | pass | pass | null | 0.116 | null | 0 | 564 | 3,880 | 6.57 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | aMep22Wzw7 | https://openreview.net/forum?id=aMep22Wzw7 | rejected | {"AIRev1": "1.0", "AIRev2": "1.0", "AIRev3": "1.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0277 | agents4science | AI_A4S0277 | HU_2007.12223 | Breaking Points: How Transformer Vulnerabilities Reveal Paths to Faster Inference | The Lottery Ticket Hypothesis for Pre-trained BERT Networks | 2007.12223 | NeurIPS 2020 | 2,020 | null | analysis | analysis | true | cited | 2 | cited | subnetwork analysis of BERT, same layer-level dissection of encoders | pass | year<=2024 | pass | pass | pass | null | null | null | null | 2,892 | 4,233 | 1.39 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | U4q6HXFvKn | https://openreview.net/forum?id=U4q6HXFvKn | rejected | {"AIRev1": "1.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": ""} | {"hypothesis_development": "C", "experimental_design": "D", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0278 | agents4science | AI_A4S0278 | HU_2603.07343 | MechSci: Scaling Clinical Science via Mechanistic Interpretability of Multimodal Medical Foundation Models | Learning Concept Bottleneck Models from Mechanistic Explanations | 2603.07343 | ICLR 2026 poster | null | 5.5 | method | method | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | turns mechanistic internals into human-interpretable concepts, the same problem and contribution kind | pass | pangram | pass | pass | pass | null | 0.123 | null | 0 | 3,148 | 4,718 | 1.14 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | rgpgukbeVf | https://openreview.net/forum?id=rgpgukbeVf | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": "5.0"} | {"hypothesis_development": "C", "experimental_design": "B", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0279 | agents4science | AI_A4S0279 | HU_1908.09791 | TensorCompress: Next-Generation Model Compression via Tensor Program Synthesis Beyond Quantization | Once-for-All: Train One Network and Specialize it for Efficient Deployment | 1908.09791 | ICLR 2020 | 2,019 | null | method | method | true | cited | 3 | cited | specializing one network per hardware target, the same hardware-aware compression problem | pass | year<=2024 | pass | pass | pass | null | null | null | null | 2,474 | 3,918 | 1.52 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | aSlhaqLbL1 | https://openreview.net/forum?id=aSlhaqLbL1 | rejected | {"AIRev1": "1.0", "AIRev2": "1.0", "AIRev3": "1.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0280 | agents4science | AI_A4S0280 | HU_2602.23008 | Modular and Hybrid Frameworks for LLM-Based Agents in Complex Strategy Games: An Empirical Study in Slay the Spire | Exploratory Memory-Augmented LLM Agent via Hybrid On- and Off-Policy Optimization | 2602.23008 | ICLR 2026 poster | null | 7 | analysis | method | false | filled_iclr26 | 2 | filled (nearest admissible anchor) | memory-augmented LLM agent for exploration, one of the architectures under comparison | pass | pangram | pass | relaxed | pass | null | 0.123 | 3 | 0 | 3,948 | 4,047 | 0.98 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | gC3D2ESSyK | https://openreview.net/forum?id=gC3D2ESSyK | rejected | {"AIRev1": "2.0", "AIRev2": "4.0", "AIRev3": "4.0", "Human": ""} | {"hypothesis_development": "C", "experimental_design": "C", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0281 | agents4science | AI_A4S0281 | HU_2502.08416 | Synthetic Realism in Finance: A High-Fidelity Market Simulation with Adaptive LLM Agents | Multifidelity Simulation-based Inference for Computationally Expensive Simulators | 2502.08416 | ICLR 2026 poster | null | 6.5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | inference for expensive simulators at several fidelities, the same simulation-fidelity problem | pass | pangram | pass | pass | pass | null | 0.076 | null | 0 | 3,454 | 4,562 | 1.24 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | qSmh66KJR4 | https://openreview.net/forum?id=qSmh66KJR4 | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "A", "experimental_design": "B", "data_analysis": "B", "writing": "B"} | rebuilt_from_pdf |
A4S0283 | agents4science | AI_A4S0283 | HU_1010.0793 | Computation of Sixth-Order Strong Coupling Expansion Coefficient for the Fermi-Hubbard Model on Honeycomb Lattice | Quantum gravitational contributions to quantum electrodynamics | 1010.0793 | published in Nature | 2,010 | null | method | method | true | cited | 1 | parent | perturbative expansion in a physical theory, same technique family different model | pass | year<=2024 | pass | pass | pass | null | null | null | null | 3,969 | 2,771 | 0.69 | Natural Sciences | Physics | nRK2lFl77R | https://openreview.net/forum?id=nRK2lFl77R | rejected | {"AIRev1": "2.0", "AIRev2": "1.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0284 | agents4science | AI_A4S0284 | HU_2510.02386 | BadScientist: Can a Research Agent Write Convincing but Unsound Papers that Fool LLM Reviewers? | On The Fragility of Benchmark Contamination Detection in Reasoning Models | 2510.02386 | ICLR 2026 poster | null | 6.667 | framework | analysis | false | filled_iclr26 | 2 | filled (nearest admissible anchor) | gaming of an evaluation system and the fragility of its detection, same question of a fooled evaluator | pass | pangram | pass | relaxed | pass | null | 0.076 | 3 | 0 | 3,197 | 3,668 | 1.05 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 7MPstNz66e | https://openreview.net/forum?id=7MPstNz66e | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "5.0", "Human": "5.0"} | {"hypothesis_development": "B", "experimental_design": "C", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0287 | agents4science | AI_A4S0287 | HU_2407.07018 | Echo: A multi-agent AI system for patient-centered pharmacovigilance | End-To-End Causal Effect Estimation from Unstructured Natural Language Data | 2407.07018 | NeurIPS 2024 | 2,024 | null | method | method | true | cited | 3 | cited | LLM pipeline that turns unstructured text into causal effect estimates, the same problem and contribution kind | pass | year<=2024 | pass | pass | pass | null | null | null | null | 3,883 | 4,820 | 1.14 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 4nrWtE6oZ9 | https://openreview.net/forum?id=4nrWtE6oZ9 | accepted | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "5.0"} | {"hypothesis_development": "B", "experimental_design": "C", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0288 | agents4science | AI_A4S0288 | HU_2201.03545 | DinoV3–LSTM for Early-Stage Classification of Paroxysmal Atrial Fibrillation in ECGs Toward Prediction | A ConvNet for the 2020s | 2201.03545 | CVPR 2022 | 2,022 | null | method | method | true | cited | 1 | cited | convolutional backbone, a component | pass | year<=2024 | pass | pass | pass | null | null | null | null | 3,229 | 4,904 | 1.46 | Interdisciplinary | Biomedical Engineering | udPANfLJBW | https://openreview.net/forum?id=udPANfLJBW | rejected | {"AIRev1": "2.0", "AIRev2": "3.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "C", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0289 | agents4science | AI_A4S0289 | HU_2503.16397 | Hardware-Conditioned Generative Channel Modeling: A Diffusion-Based Approach for Location and Hardware-Aware Wireless Dataset Synthesis | Scale-wise Distillation of Diffusion Models | 2503.16397 | ICLR 2026 poster | null | 5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | conditional diffusion generation efficiency, the same generator family | pass | pangram | pass | pass | pass | null | 0.086 | null | 0 | 3,535 | 3,553 | 0.96 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | ExGHHgTM2p | https://openreview.net/forum?id=ExGHHgTM2p | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "4.0"} | {"hypothesis_development": "C", "experimental_design": "D", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0290 | agents4science | AI_A4S0290 | HU_2407.07311 | AgentAdapter-TimesFM: Agentic Residual Adapters for Scientific Time-Series Forecasting | ViTime: Foundation Model for Time Series Forecasting Powered by Vision Intelligence | 2407.07311 | Transactions on Machine Learning Researc | 2,024 | null | method | method | true | cited | 3 | cited | time-series foundation model for forecasting, the same problem and contribution kind | pass | year<=2024 | pass | pass | pass | null | null | null | null | 2,425 | 7,715 | 3.02 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | ud9cRk9yBM | https://openreview.net/forum?id=ud9cRk9yBM | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "C", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0291 | agents4science | AI_A4S0291 | HU_2508.05606 | Mentor-Mind: Risk-Aware, Constraint-Grounded Advice Agents Beyond Chain-of-Thought | Uni-CoT: Towards Unified Chain-of-Thought Reasoning Across Text and Vision | 2508.05606 | ICLR 2026 poster | null | 5.6 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | structured reasoning beyond plain chain of thought, the same going-beyond-CoT contribution | pass | pangram | pass | pass | pass | null | 0.114 | null | 0.014952 | 4,572 | 3,552 | 0.75 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | LqukdleDgU | https://openreview.net/forum?id=LqukdleDgU | accepted | {"AIRev1": "4.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "4.0"} | {"hypothesis_development": "A", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0292 | agents4science | AI_A4S0292 | HU_2602.08031 | SHARP: Cascaded Regex-LLM Architecture for Phishing Detection | Beyond Raw Detection Scores: Markov-Informed Calibration for Boosting Machine-Generated Text Detection | 2602.08031 | ICLR 2026 poster | null | 4.5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | calibrated cascade of statistical detection scores, the same detector-architecture idea | pass | pangram | pass | pass | pass | null | 0.076 | null | 0 | 2,789 | 3,908 | 1.32 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | EwXxI9mF4d | https://openreview.net/forum?id=EwXxI9mF4d | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "2.0"} | {"hypothesis_development": "C", "experimental_design": "C", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0293 | agents4science | AI_A4S0293 | HU_2510.04899 | PsySpace: Simulating Emergent Psychological Dynamics in Long-Duration Space Missions using Multi-Agent LLMs | Human Behavior Atlas: Benchmarking Unified Psychological And Social Behavior Understanding | 2510.04899 | ICLR 2026 poster | null | 5.2 | method | benchmark | false | filled_iclr26 | 2 | filled (nearest admissible anchor) | benchmark of psychological and social behaviour, same target phenomenon | pass | pangram | pass | relaxed | pass | null | 0.092 | 3 | 0 | 4,150 | 4,709 | 1.08 | Computer & Data Sciences | Human-Computer Interaction | WAbHXkmBIn | https://openreview.net/forum?id=WAbHXkmBIn | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "5.0"} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0295 | agents4science | AI_A4S0295 | HU_2005.11401 | LLM-Driven Discovery of High-Entropy Catalysts via Retrieval-Augmented Generation | Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks | 2005.11401 | NeurIPS 2020 | 2,020 | null | method | method | true | cited | 1 | cited | retrieval-augmented generation, the machinery the paper reuses but on knowledge-intensive NLP rather than catalyst discovery | pass | year<=2024 | pass | pass | pass | null | null | null | null | 2,214 | 4,303 | 1.44 | Engineering & Technology | Materials Science | Ka1WYEwLdC | https://openreview.net/forum?id=Ka1WYEwLdC | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "5.0", "Human": "4.0"} | {"hypothesis_development": "Not found", "experimental_design": "Not found", "data_analysis": "Not found", "writing": "Not found"} | shipped_tex |
A4S0296 | agents4science | AI_A4S0296 | HU_2509.24869 | Intelligent_Document_Processing_for_Graduate_Admissions__An_End_to_End_Pipeline_with_Calibrated_Abstention | Retro*: Optimizing LLMs for Reasoning-Intensive Document Retrieval | 2509.24869 | ICLR 2026 poster | null | 4.667 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | retrieval over documents to serve a downstream decision, same document pipeline problem | pass | pangram | pass | pass | pass | null | 0.061 | null | 0 | 1,651 | 3,850 | 2.21 | Computer & Data Sciences | Human-Computer Interaction | w7c1rotw9I | https://openreview.net/forum?id=w7c1rotw9I | rejected | {"AIRev1": "1.0", "AIRev2": "1.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "B", "data_analysis": "B", "writing": "B"} | shipped_tex |
A4S0297 | agents4science | AI_A4S0297 | HU_2203.06486 | Breaking the Chart Barrier: A Comprehensive Analysis Reveals Why AI Excels at Code but Fails at Visual Scientific Diagrams | Chart-to-Text: A Large-Scale Benchmark for Chart Summarization | 2203.06486 | ACL 2022 | 2,022 | null | benchmark | benchmark | true | cited | 3 | cited | large-scale chart benchmark with data construction and metrics, the same problem and contribution kind | pass | year<=2024 | pass | pass | pass | null | null | null | null | 2,900 | 4,581 | 1.51 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 51ri0E84gG | https://openreview.net/forum?id=51ri0E84gG | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "C", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0298 | agents4science | AI_A4S0298 | HU_2505.19619 | Calculation of the n=1 Critical Point in the Bose-Hubbard Model on the Isotropic Union Jack Lattice via Quantum Monte Carlo (QMC) | SESaMo: Symmetry-Enforcing Stochastic Modulation for Normalizing Flows | 2505.19619 | ICLR 2026 poster | null | 4.5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | sampling from unnormalised lattice physics distributions, same lattice-model estimation problem | pass | pangram | pass | pass | pass | null | 0.068 | null | 0 | 1,413 | 4,150 | 2.69 | Natural Sciences | Physics | KjkhwoXbYK | https://openreview.net/forum?id=KjkhwoXbYK | rejected | {"AIRev1": "1.0", "AIRev2": "1.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0299 | agents4science | AI_A4S0299 | HU_2509.26325 | EndoNet: Content-Aware Linear Attention for Endoscopic Video Super-Resolution | Continuous Space-Time Video Super-Resolution with 3D Fourier Fields | 2509.26325 | ICLR 2026 poster | null | 6 | method | method | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | space-time video super-resolution with a new temporal representation, the same problem and contribution kind | pass | pangram | pass | pass | pass | null | 0.162 | null | 0 | 3,651 | 3,883 | 1.02 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | si22iakiB5 | https://openreview.net/forum?id=si22iakiB5 | rejected | {"AIRev1": "2.0", "AIRev2": "3.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "C", "experimental_design": "C", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0300 | agents4science | AI_A4S0300 | HU_2002.08910 | Decontextualization, Everywhere: A Systematic Audit on PeerQA | How Much Knowledge Can You Pack Into the Parameters of a Language Model? | 2002.08910 | Camera-ready version for EMNLP | 2,020 | null | analysis | analysis | true | cited | 1 | cited | parametric knowledge measurement, different question | pass | year<=2024 | pass | pass | pass | null | null | null | null | 3,229 | 2,586 | 0.71 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | cK8YYMc65B | https://openreview.net/forum?id=cK8YYMc65B | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "3.0"} | {"hypothesis_development": "C", "experimental_design": "C", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0301 | agents4science | AI_A4S0301 | HU_2509.21798 | A Simulation Study on the Impact of Technology, Resources, and Culture on Human Desire | Evaluating and Improving Cultural Awareness of Reward Models for LLM Alignment | 2509.21798 | ICLR 2026 poster | null | 5.5 | method | analysis | false | filled_iclr26 | 1 | filled (nearest admissible anchor) | cultural awareness of reward models, culture-variable area only | pass | pangram | pass | relaxed | pass | null | 0.074 | 1 | 0 | 1,556 | 3,915 | 2.45 | Interdisciplinary | Computational Social Science | KnkFyU8BmD | https://openreview.net/forum?id=KnkFyU8BmD | rejected | {"AIRev1": "1.0", "AIRev2": "1.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "B", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0302 | agents4science | AI_A4S0302 | HU_2410.09083 | Exploring the Potential for AI Intervention in Value Judgment Through Free Discussion Among Large Language Models: Focusing on Multi-Stage Trolley Dilemma Discussion Analysis | Can LLMs Reason Soundly in Law? Auditing Inference Patterns for Legal Judgment | 2410.09083 | ICLR 2026 poster | null | 4.5 | analysis | analysis | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | auditing LLM judgment patterns, same question of LLM normative judgment | pass | pangram | pass | pass | pass | null | 0.081 | null | 0 | 8,352 | 4,100 | 0.49 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | diwnmHfFNE | https://openreview.net/forum?id=diwnmHfFNE | rejected | {"AIRev1": "2.0", "AIRev2": "3.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "A", "experimental_design": "C", "data_analysis": "B", "writing": "B"} | rebuilt_from_pdf |
A4S0303 | agents4science | AI_A4S0303 | HU_1709.06560 | The Overfitting Crisis in LLM Workflows: Learning from Machine Learning’s Past Mistakes | Deep Reinforcement Learning that Matters | 1709.06560 | Accepted to the Thirthy-Second AAAI | 2,017 | null | analysis | analysis | true | cited | 3 | cited | field-level critique of reporting practice with controlled reruns, the same problem and contribution kind one field earlier | pass | year<=2024 | pass | pass | pass | null | null | null | null | 3,209 | 4,090 | 1.22 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | maMnVCHl8J | https://openreview.net/forum?id=maMnVCHl8J | rejected | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "3.0"} | {"hypothesis_development": "B", "experimental_design": "NA", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0304 | agents4science | AI_A4S0304 | HU_2504.09389 | Backtest to the Future: Can Large Language Models Generate Publishable AI Research Ideas? | Measuring LLM Novelty As The Frontier Of Original And High-Quality Output | 2504.09389 | ICLR 2026 poster | null | 6 | framework | framework | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | protocol for measuring novelty of LLM-generated research ideas, same evaluation question | pass | pangram | pass | pass | pass | null | 0.079 | null | 0 | 3,095 | 3,837 | 1.2 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | aK9JwuE29c | https://openreview.net/forum?id=aK9JwuE29c | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "C", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0305 | agents4science | AI_A4S0305 | HU_2505.13995 | Moral Elevation, Empathy, and Group Cohesion: Predicting Immediate Prosocial Intentions in an Online Survey Context | ELEPHANT: Measuring and understanding social sycophancy in LLMs | 2505.13995 | ICLR 2026 poster | null | 6 | survey | analysis | false | filled_iclr26 | 1 | filled (nearest admissible anchor) | social sycophancy in LLMs, social-emotion area only | pass | pangram | pass | relaxed | pass | null | 0.064 | 1 | 0.035218 | 3,877 | 3,670 | 0.91 | Social Sciences | Psychology | PYqvzvr5F7 | https://openreview.net/forum?id=PYqvzvr5F7 | rejected | {"AIRev1": "1.0", "AIRev2": "1.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "B", "data_analysis": "B", "writing": "D"} | rebuilt_from_pdf |
A4S0306 | agents4science | AI_A4S0306 | HU_2512.18114 | Automated Mining of Hinge-like Protein Modules from AlphaFold PAE: BCR-Parts | Greater than the Sum of Its Parts: Building Substructure into Protein Encoding Models | 2512.18114 | ICLR 2026 poster | null | 5.5 | method | method | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | builds recurrent substructure units into protein models, the same modular-parts problem and contribution kind | pass | pangram | pass | pass | pass | null | 0.062 | null | 0 | 2,400 | 3,980 | 1.46 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 3PWDmzgjbb | https://openreview.net/forum?id=3PWDmzgjbb | rejected | {"AIRev1": "3.0", "AIRev2": "4.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0307 | agents4science | AI_A4S0307 | HU_2106.07594 | Temporal Motif-Enhanced Contrastive Learning for Adaptive Anomaly Detection in Dynamic Networks | Graph Contrastive Learning Automated | 2106.07594 | ICML 2021 | 2,021 | null | method | method | true | cited | 2 | cited | automated graph contrastive learning, the mechanism on a different task | pass | year<=2024 | pass | pass | pass | null | null | null | null | 1,860 | 3,838 | 1.82 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | AL5YBk5JXP | https://openreview.net/forum?id=AL5YBk5JXP | rejected | {"AIRev1": "2.0", "AIRev2": "2.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "C", "experimental_design": "D", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0308 | agents4science | AI_A4S0308 | HU_2105.14491 | GATv2-NS3 Hybrid IDS: Self-Focusing Simulations for Network Intrusion Detection | How Attentive are Graph Attention Networks? | 2105.14491 | ICLR 2022 | 2,021 | null | method | method | true | cited | 2 | cited | GATv2, the attention architecture the system is built on | pass | year<=2024 | pass | pass | pass | null | null | null | null | 1,968 | 2,515 | 1.13 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | S8uzdZvStA | https://openreview.net/forum?id=S8uzdZvStA | rejected | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "2.0"} | {"hypothesis_development": "C", "experimental_design": "C", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0309 | agents4science | AI_A4S0309 | HU_2510.01279 | PST-Auto-Agent: A Multi-Agent Ensemble Framework for Paper Source Tracing | TUMIX: Multi-Agent Test-Time Scaling with Tool-Use Mixture | 2510.01279 | ICLR 2026 poster | null | 5.5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | multi-agent ensemble with mixed tools at test time, the same ensembling contribution on a different task | pass | pangram | pass | pass | pass | null | 0.109 | null | 0 | 2,729 | 4,130 | 1.47 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 6b5kZtfwIH | https://openreview.net/forum?id=6b5kZtfwIH | rejected | {"AIRev1": "2.0", "AIRev2": "1.0", "AIRev3": "4.0", "Human": ""} | {"hypothesis_development": "A", "experimental_design": "C", "data_analysis": "B", "writing": "C"} | rebuilt_from_pdf |
A4S0312 | agents4science | AI_A4S0312 | HU_2502.15483 | Superior Energy Storage Performance in High-Entropy Dielectric Ceramics Discovered by AI Materials Scientist | MoMa: A Simple Modular Learning Framework for Material Property Prediction | 2502.15483 | ICLR 2026 poster | null | 5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | modular learning for material property prediction, same materials-discovery prediction problem | pass | pangram | pass | pass | pass | null | 0.084 | null | 0 | 2,951 | 4,010 | 1.17 | Engineering & Technology | Materials Science | gArBKNBVyN | https://openreview.net/forum?id=gArBKNBVyN | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "C", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0314 | agents4science | AI_A4S0314 | HU_2510.07195 | Quantum Semantic Dynamics: A Unified Framework for N-Dimensional Semantic Hilbert Spaces with Experimental Validation | Accelerating Inference for Multilayer Neural Networks with Quantum Computers | 2510.07195 | ICLR 2026 poster | null | 5.5 | method | method | true | filled_iclr26 | 1 | filled (nearest admissible anchor) | quantum computing for neural inference, quantum area only | pass | pangram | pass | pass | pass | null | 0.126 | 2 | 0 | 2,927 | 5,144 | 1.6 | Interdisciplinary | Cognitive Science | 814MdsPtZq | https://openreview.net/forum?id=814MdsPtZq | rejected | {"AIRev1": "2.0", "AIRev2": "1.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "C", "experimental_design": "C", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0315 | agents4science | AI_A4S0315 | HU_2410.09403 | Magellan: Guided MCTS for Latent Space Exploration and Novelty Generation | Many Heads Are Better Than One: Improved Scientific Idea Generation by A LLM-Based Multi-Agent System | 2410.09403 | ACL 2025 | 2,024 | null | method | method | true | cited | 3 | cited | multi-agent LLM system for generating novel scientific ideas, the same problem and contribution kind | pass | year<=2024 | pass | pass | pass | null | null | null | null | 3,616 | 4,307 | 1.15 | Computer & Data Sciences | Human-Computer Interaction | wVn3uPvm9W | https://openreview.net/forum?id=wVn3uPvm9W | accepted | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "5.0", "Human": "4.0"} | {"hypothesis_development": "D", "experimental_design": "C", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0317 | agents4science | AI_A4S0317 | HU_2412.14555 | Adaptive Federated Q-Learning with Importance Averaging: Near-Optimal Sample Complexity and $K$-Independent Communication | Single-Loop Federated Actor-Critic across Heterogeneous Environments | 2412.14555 | AAAI'25 | 2,024 | null | method | method | true | cited | 3 | cited | federated RL with heterogeneous agents and communication analysis, the same problem and contribution kind | pass | year<=2024 | pass | pass | pass | null | null | null | null | 2,787 | 4,061 | 1.41 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | wLhQMbyF3y | https://openreview.net/forum?id=wLhQMbyF3y | rejected | {"AIRev1": "3.0", "AIRev2": "5.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0318 | agents4science | AI_A4S0318 | HU_2509.06426 | CXCL13 as a Prognostic Biomarker of Survival Outcomes in Muscle-Invasive Bladder Cancer | Musculoskeletal simulation of limb movement biomechanics in Drosophila melanogaster | 2509.06426 | ICLR 2026 poster | null | 6 | method | method | true | filled_iclr26 | 1 | filled (nearest admissible anchor) | biomechanical simulation, biology area only | pass | pangram | pass | pass | pass | null | 0.069 | 2 | 0.029753 | 1,802 | 3,386 | 1.67 | Life & Health Sciences | Medicine & Health | sXgKlXfaBK | https://openreview.net/forum?id=sXgKlXfaBK | rejected | {"AIRev1": "3.0", "AIRev2": "5.0", "AIRev3": "4.0", "Human": "2.0"} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0320 | agents4science | AI_A4S0320 | HU_2401.06855 | HALT: A Framework for Hallucination Detection in Large Language Models | Fine-grained Hallucination Detection and Editing for Language Models | 2401.06855 | COLM 2024 | 2,024 | null | method | method | true | cited | 3 | cited | fine-grained hallucination detection with its own benchmark, the same problem | pass | year<=2024 | pass | pass | pass | null | null | null | null | 2,822 | 3,770 | 1.32 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | YysFSiPQf7 | https://openreview.net/forum?id=YysFSiPQf7 | rejected | {"AIRev1": "2.0", "AIRev2": "4.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "C", "experimental_design": "D", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0321 | agents4science | AI_A4S0321 | HU_2604.23134 | Integrating Segmented Cell Imaging and Molecular Networks for Drug-Specific Analysis in CM4AI | h-MINT: Modeling Pocket-Ligand Binding with Hierarchical Molecular Interaction Network | 2604.23134 | ICLR 2026 poster | null | 4.5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | hierarchical molecular interaction modelling, the molecular half of the integration | pass | pangram | pass | pass | pass | null | 0.081 | null | 0 | 2,979 | 3,616 | 1.16 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | zTiOPZnpWv | https://openreview.net/forum?id=zTiOPZnpWv | rejected | {"AIRev1": "2.0", "AIRev2": "3.0", "AIRev3": "4.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "C", "data_analysis": "B", "writing": "B"} | rebuilt_from_pdf |
A4S0322 | agents4science | AI_A4S0322 | HU_2603.04730 | CRISPR Screen Design for T Cell Exhaustion Regulators: A Systematic Approach to Identify 32 Target Genes | Count Bridges enable Modeling and Deconvolving Transcriptomics | 2603.04730 | ICLR 2026 poster | null | 5.6 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | modelling and deconvolving transcriptomic counts, same readout of a perturbation screen | pass | pangram | pass | pass | pass | null | 0.089 | null | 0 | 2,475 | 4,534 | 1.69 | Life & Health Sciences | Biology | 98iucZYjap | https://openreview.net/forum?id=98iucZYjap | rejected | {"AIRev1": "2.0", "AIRev2": "5.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0323 | agents4science | AI_A4S0323 | HU_2305.16934 | Exploring Vision-Language Alignment under Subtle Contradictions | On Evaluating Adversarial Robustness of Large Vision-Language Models | 2305.16934 | NeurIPS 2023 | 2,023 | null | analysis | analysis | true | cited | 3 | cited | systematic evaluation of adversarial robustness in large VLMs, the same problem and contribution kind | pass | year<=2024 | pass | pass | pass | null | null | null | null | 1,591 | 3,173 | 1.76 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | bNP4BRONCE | https://openreview.net/forum?id=bNP4BRONCE | rejected | {"AIRev1": "3.0", "AIRev2": "2.0", "AIRev3": "4.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "C"} | rebuilt_from_pdf |
A4S0325 | agents4science | AI_A4S0325 | HU_2505.14886 | Scalable Oversight in Multi-Agent Systems: Provable Alignment via Delegated Debate and Hierarchical Verification | Strategic Planning and Rationalizing on Trees Make LLMs Better Debaters | 2505.14886 | ICLR 2026 poster | null | 4.5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | structured debate with planning for LLMs, the delegated-debate mechanism without the oversight guarantees | pass | pangram | pass | pass | pass | null | 0.109 | null | 0 | 2,659 | 4,928 | 1.43 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | l5Wrcgyobp | https://openreview.net/forum?id=l5Wrcgyobp | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "5.0"} | {"hypothesis_development": "B", "experimental_design": "B", "data_analysis": "B", "writing": "C"} | rebuilt_from_pdf |
A4S0326 | agents4science | AI_A4S0326 | HU_2603.06748 | A Multi-Agent LLM System for Protein Sequence Design and Structure-Oriented Ranking | Property-Driven Protein Inverse Folding with Multi-Objective Preference Alignment | 2603.06748 | ICLR 2026 poster | null | 6 | method | method | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | protein sequence design balancing several competing properties with ranking, the same problem and contribution kind | pass | pangram | pass | pass | pass | null | 0.158 | null | 0.037132 | 2,365 | 3,441 | 1.38 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 7qIGNFpf1C | https://openreview.net/forum?id=7qIGNFpf1C | rejected | {"AIRev1": "2.0", "AIRev2": "4.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0328 | agents4science | AI_A4S0328 | HU_2505.15420 | Indirect Prompt Injection in AI-Native Peer Review: Risks, Detection, and Defenses | Silent Leaks: Implicit Knowledge Extraction Attack on RAG Systems | 2505.15420 | ICLR 2026 poster | null | 4.5 | framework | framework | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | attack and defense on RAG pipelines, same LLM-pipeline attack surface question | pass | pangram | pass | pass | pass | null | 0.075 | null | 0 | 3,240 | 3,701 | 0.72 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | q15GkFTXbM | https://openreview.net/forum?id=q15GkFTXbM | rejected | {"AIRev1": "3.0", "AIRev2": "5.0", "AIRev3": "4.0", "Human": "1.0"} | {"hypothesis_development": "C", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0329 | agents4science | AI_A4S0329 | HU_2505.19558 | The Verifiability Gateway: A Governance Agent’s Discovery of SAI Non-Identifiability | PoliCon: Evaluating LLMs on Achieving Diverse Political Consensus Objectives | 2505.19558 | ICLR 2026 poster | null | 6 | method | benchmark | false | filled_iclr26 | 1 | filled (nearest admissible anchor) | political consensus evaluation, governance area only | pass | pangram | pass | relaxed | pass | null | 0.077 | 1 | 0 | 4,333 | 5,270 | 1.4 | Interdisciplinary | Science & Technology Studies | VUTIDJhvD3 | https://openreview.net/forum?id=VUTIDJhvD3 | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "C", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0331 | agents4science | AI_A4S0331 | HU_2512.06749 | Diagnostic Failure Paradigm: Transforming AI System Validation Through Systematic Analysis of Classical Model Failures | DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systems | 2512.06749 | ICLR 2026 poster | null | 5 | framework | framework | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | intervention-driven debugging of an AI system's failures, same failure-diagnosis validation | pass | pangram | pass | pass | pass | null | 0.084 | null | 0 | 3,297 | 5,378 | 1.53 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 7k3i6JYvFo | https://openreview.net/forum?id=7k3i6JYvFo | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0332 | agents4science | AI_A4S0332 | HU_2603.15483 | The Self-Limiting Nature of QBO-Dependent SAI: An Optimization Agent’s Discovery of Intervention-Variability Feedback | Talk, Evaluate, Diagnose: User-aware Agent Evaluation with Automated Error Analysis | 2603.15483 | ICLR 2026 poster | null | 5.333 | analysis | analysis | true | filled_iclr26 | 1 | filled (nearest admissible anchor) | automated error analysis of agents, agent evaluation area but not a climate finding | pass | pangram | pass | pass | pass | null | 0.054 | 2 | 0 | 3,590 | 5,840 | 1.74 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | eSlhdv8zUL | https://openreview.net/forum?id=eSlhdv8zUL | rejected | {"AIRev1": "2.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0333 | agents4science | AI_A4S0333 | HU_2509.02642 | Multi-target Parallel Drug Discovery with Multi-agent Orchestration | BioMD: All-atom Generative Model for Biomolecular Dynamics Simulation | 2509.02642 | ICLR 2026 poster | null | 5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | generative model inside a drug discovery pipeline, one stage of the same workflow | pass | pangram | pass | pass | pass | null | 0.116 | null | 0 | 3,372 | 3,264 | 0.89 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | AhFsKmuaCb | https://openreview.net/forum?id=AhFsKmuaCb | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "4.0"} | {"hypothesis_development": "", "experimental_design": "", "data_analysis": "", "writing": ""} | rebuilt_from_pdf |
A4S0334 | agents4science | AI_A4S0334 | HU_2511.04854 | Initiation of Programmed Cell Death in Cancer Stem Cells: In Silico Mutagenesis for Optimized TRAIL–DR5 Binding with Perplexity AI | SigmaDock: Untwisting Molecular Docking with Fragment-Based SE(3) Diffusion | 2511.04854 | ICLR 2026 poster | null | 6 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | diffusion molecular docking, same computational binding modelling problem | pass | pangram | pass | pass | pass | null | 0.11 | null | 0 | 1,051 | 4,888 | 3.61 | Life & Health Sciences | Medicine & Health | 2w5HypzUpN | https://openreview.net/forum?id=2w5HypzUpN | rejected | {"AIRev1": "2.0", "AIRev2": "2.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "A", "experimental_design": "A", "data_analysis": "A", "writing": "A"} | rebuilt_from_pdf |
A4S0337 | agents4science | AI_A4S0337 | HU_2405.19697 | Single-Loop Penalty Methods for Bilevel Reinforcement Learning | Bilevel reinforcement learning via the development of hyper-gradient without lower-level convexity | 2405.19697 | AISTATS 2025 | 2,024 | null | method | method | true | cited | 3 | cited | first-order bilevel RL without lower-level convexity, the same problem and contribution kind | pass | year<=2024 | pass | pass | pass | null | null | null | null | 3,233 | 3,709 | 1.09 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | NRaWmVVltg | https://openreview.net/forum?id=NRaWmVVltg | rejected | {"AIRev1": "4.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "3.0"} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0338 | agents4science | AI_A4S0338 | HU_2505.12506 | The Shape of Risk: Dynamic Regime Shifts in Factor Models for Market Portfolio Optimization | Unsupervised Invariant Risk Minimization | 2505.12506 | ICLR 2026 poster | null | 5.333 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | invariance of a model across environments without labels, same regime-stability question | pass | pangram | pass | pass | pass | null | 0.062 | null | 0.043175 | 2,675 | 4,376 | 1.58 | Social Sciences | Economics | Wu0W5BJ7UR | https://openreview.net/forum?id=Wu0W5BJ7UR | rejected | {"AIRev1": "2.0", "AIRev2": "2.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "C", "experimental_design": "C", "data_analysis": "C", "writing": "C"} | rebuilt_from_pdf |
A4S0340 | agents4science | AI_A4S0340 | HU_2601.20882 | The Architectural Immune System: A Framework for Correcting Synthetic Fallacies in AI-Driven Science | DevOps-Gym: Benchmarking AI Agents in Software DevOps Cycle | 2601.20882 | ICLR 2026 poster | null | 6 | framework | benchmark | false | filled_iclr26 | 1 | filled (nearest admissible anchor) | agent lifecycle benchmark, agent area only | pass | pangram | pass | relaxed | pass | null | 0.064 | 1 | 0.04234 | 3,367 | 5,916 | 1.3 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | ShWjvhAZGs | https://openreview.net/forum?id=ShWjvhAZGs | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "3.0", "Human": "5.0"} | {"hypothesis_development": "D", "experimental_design": "C", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
A4S0342 | agents4science | AI_A4S0342 | HU_2508.06361 | Scaling Laws of Deception in AI Scientist Agents: World-Model Manipulation in LLMs | Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts | 2508.06361 | ICLR 2026 oral | null | 6.667 | analysis | analysis | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | investigates deliberate LLM deception under controlled prompts, the same problem and contribution kind | pass | pangram | pass | pass | pass | null | 0.135 | null | 0 | 1,798 | 4,696 | 2.22 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | Pqjg14gnAV | https://openreview.net/forum?id=Pqjg14gnAV | rejected | {"AIRev1": "2.0", "AIRev2": "2.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "B", "experimental_design": "B", "data_analysis": "B", "writing": "B"} | rebuilt_from_pdf |
A4S0343 | agents4science | AI_A4S0343 | HU_2602.02356 | Joint Material Reconstruction for Sparse Dual-Energy CT | NAB: Neural Adaptive Binning for Sparse-View CT reconstruction | 2602.02356 | ICLR 2026 poster | null | 4.5 | method | method | true | widened_iclr26 | 3 | widened (not cited; topic neighbour) | neural binning for sparse-view CT reconstruction, same sparse CT reconstruction problem and same contribution kind | pass | pangram | pass | pass | pass | null | 0.159 | null | 0 | 2,947 | 3,213 | 1.03 | Engineering & Technology | Biomedical Engineering | mEbFdPy2oy | https://openreview.net/forum?id=mEbFdPy2oy | rejected | {"AIRev1": "2.0", "AIRev2": "4.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "C", "experimental_design": "C", "data_analysis": "C", "writing": "D"} | rebuilt_from_pdf |
A4S0344 | agents4science | AI_A4S0344 | HU_2510.03154 | Stylistic Contrastive Learning for Human-Like AI Text Generation | EditLens: Quantifying the Extent of AI Editing in Text | 2510.03154 | ICLR 2026 poster | null | 4.5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | quantifies how far text has drifted from human writing, the gap the method tries to close | pass | pangram | pass | pass | pass | null | 0.093 | null | 0 | 3,227 | 3,961 | 1.21 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 0fo0d9Tbey | https://openreview.net/forum?id=0fo0d9Tbey | accepted | {"AIRev1": "3.0", "AIRev2": "6.0", "AIRev3": "4.0", "Human": "3.0"} | {"hypothesis_development": "A", "experimental_design": "B", "data_analysis": "A", "writing": "B"} | rebuilt_from_pdf |
A4S0345 | agents4science | AI_A4S0345 | HU_2604.23072 | Multi-Agent AI System for Pharmaceutical Commercial Forecasting: GPT-5 Orchestration with Specialized Agent Architecture | Analytica: Soft Propositional Reasoning for Robust and Scalable LLM-Driven Analysis | 2604.23072 | ICLR 2026 poster | null | 5 | method | method | true | widened_iclr26 | 2 | widened (not cited; topic neighbour) | LLM agents doing real-world forecasting and analysis, same multi-agent forecasting system | pass | pangram | pass | pass | pass | null | 0.073 | null | 0 | 2,530 | 4,444 | 1.66 | Computer & Data Sciences | Artificial Intelligence & Machine Learning | 3Sv9wfwXPW | https://openreview.net/forum?id=3Sv9wfwXPW | rejected | {"AIRev1": "2.0", "AIRev2": "1.0", "AIRev3": "2.0", "Human": ""} | {"hypothesis_development": "D", "experimental_design": "D", "data_analysis": "D", "writing": "D"} | rebuilt_from_pdf |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.