Skip to content
AI Atlas
Research

Papers

Publications linked to models, labs and benchmarks. Authors, venues and abstracts come from arXiv and publisher pages.

1,050 total

Reset
Papers
TitleAuthorsPublishedCategoriesOrganization / venueQuality
When Auditors Fabricate: Batch-Size Degradation and Confident Hallucination in LLM Detection of Planted Document ContaminationarXiv:2609.09696Karan Parekh, Sanjana Pendyala Ravinder, Sana Mhapsekar +111 Sept 2026cs.CL—89
Kernel-Complexity Edge Sanitization for Training-Free Defense against Structural Graph AttacksarXiv:2609.09698Yaning Jia, Shenyang Deng, Yaoqing Yang +211 Sept 2026cs.LG—89
Distilling Image Prototypes for Guided Test-Time AdaptationarXiv:2609.09737Liwen Wang, Xingbo Dong, Iman Yi Liao +211 Sept 2026cs.CV—89
HiRAD: A Flexible Large-Scale AGV Routing SystemarXiv:2609.09752Yunjie Huang, Ruizhong Wu, Mengxuan Zhang +211 Sept 2026cs.RO—89
Fine-Tuning a KV Cache Concatenation-Aware Model or Recomputing KV Caches? Why Not Both?arXiv:2609.09768Fumihiko Tachibana, Daisuke Miyashita, Jun Deguchi11 Sept 2026cs.LG—89
BRACE: Anchored Bellman-Residual Correction for Stale Critics in Asynchronous RLarXiv:2609.09783Guanqun Zhao, Zijun Xie, Binbin Zheng +211 Sept 2026cs.LG—89
Pairit: A Platform for Live Experiments on Human-AI CollaborationarXiv:2609.09789Harang Ju, Sinan Aral11 Sept 2026cs.HC—89
LogiScope-VQA: Benchmarking Vision-Language Models for Logistics Hazard Identification in Industrial ScenariosarXiv:2609.09790Hanjing Zhou, Mingze Yin, Ying Lian +211 Sept 2026cs.CV—89
How Fragile Is Safety Alignment at Frontier Scale? A Single-Direction Attack on a 320B MoEarXiv:2609.09793Yi Shi, Tanyu Chen, Kai Shen11 Sept 2026cs.CR—89
CS-Guard: Benchmarking LLM Guardrails for Code Generation SecurityarXiv:2609.09798Jinyang Li, Mingyu Guo, Hung X. Nguyen11 Sept 2026cs.CR—89
uFlowCSP: Crystal Structure Prediction using Mean flow generative modelsarXiv:2609.09799Sourin Dey, Dipannoy Das Gupta, Lai Wei +211 Sept 2026cond-mat.mtrl-sci—89
Subgroup Membership Inference Audits of Differentially Private Synthetic TextarXiv:2609.09848Yidan Sun, Viktor Schlegel, Srinivasan Nandakumar +211 Sept 2026cs.CR—89
Can AI Agents Detect and Repair Artifact Drift in Network Experiments?arXiv:2609.09849Tianzhu Zhang, Weichen Tao, Changgang Zheng +211 Sept 2026cs.NI—89
With a Thermomix You Lose the Ability to Cook: A Kitchen Machine Analogy for Applications of Generative AI in EducationarXiv:2609.09856Nikol Rummel, Valentina Nachtigall, Ernesto Panadero11 Sept 2026cs.CY—89
Forward-Free LLM Depth Pruning via Weight RedundancyarXiv:2609.09883Vincent-Daniel Yun, Woosang Lim11 Sept 2026cs.LG—89
Albedo Estimation via Latent Bridge MatchingarXiv:2609.09884Carme Corbi, David Serrano-Lozano, Javier Vazquez-Corral +111 Sept 2026cs.CV—89
Strangers to Themselves: What Language Models Say About Themselves Is GenericarXiv:2609.09899Phil Blandfort, Urja Pawar11 Sept 2026cs.LG—89
FlowCPO: A Unified Divergence View of Preference Alignment for Flow ModelsarXiv:2609.09905Yansen Han, Shengyi Liao, Peng Sun +211 Sept 2026stat.ML—89
Improving Cross-Lingual Token Representations by Adding a Pinch of SALTarXiv:2609.09953Guillem Ram\'irez11 Sept 2026cs.CL—89
Fidelity-Aware Scheduling of Quantum Circuits on Multi-QPU SystemsarXiv:2609.09980Innocenzo Fulginiti, Antonio Tudisco, Salvatore Zammuto +211 Sept 2026quant-ph—89
What Makes Adversarial Examples Transfer Across Deepfake Detectors?arXiv:2609.10002Rafael M. Mamede, Pedro C. Neto, Ana F. Sequeira11 Sept 2026cs.CV—89
MetroLLM-Bench: Evaluating Language Models as Transit Kiosk RuntimesarXiv:2609.10016Remco Hendriks (Continker)11 Sept 2026cs.LG—79
Elastoformer: Enabling Dynamic Adaptivity via Elastic Model TransformationarXiv:2609.10018Sudaksh Kalra, Dolly Sapra11 Sept 2026cs.CV—89
Direct Diversity Optimization for Diverse Successful Trajectories in Preference Post-TrainingarXiv:2609.10052Junwon Ko, Dong-Jae Lee, Minchan Kwon +211 Sept 2026cs.CL—89
NOPE-HYPE: A Structured Simulation Workflow for Robust Speech-to-Text Across Diverse Acoustic EnvironmentsarXiv:2609.10058Niramay M. Patel, Bibek Behera, Raksha Sharma11 Sept 2026cs.SD—89
A statistical approach to bias in zero-shot learning: the lens of handwriting recognitionarXiv:2609.10084Clarence Chew, Gim Siang Chia, Sukalpa Chanda +211 Sept 2026stat.ML—89
Beyond Training: A Feasibility Taxonomy for Inference-Time AI GovernancearXiv:2609.10105Samar Ansari11 Sept 2026cs.CY—89
A Trust-Network-Based Federated Learning Framework for Multi-Center Aging Clock PredictionarXiv:2609.10108Chunxu Zhang, Bo Li, Wenliang Wang +211 Sept 2026cs.LG—89
SA-Profile: Automated Sulcus Angle Profiling from Super-Resolution MRIarXiv:2609.10125Michael Wehrli, Leo Widmer, Edwin Li +211 Sept 2026cs.CV—89
Context operations to architecture modelling output from large language models and evaluation criteria for their use in systems engineering designarXiv:2609.10132Vinicius Kaster Marini, Petter Krus11 Sept 2026eess.SY—89
Active Adaptation, Not Static Defense: Temporal Dynamics of Preventative Steering in Adversarial Fine-TuningarXiv:2609.10142Jing Guan, Yachao Yang, Zhaoliang Liu +211 Sept 2026cs.CL—89
Can AI Agents Deliver Verifiable Network-Wide Outcomes Across Authority Boundaries?arXiv:2609.10181Tianzhu Zhang, Chih-Kai Huang, Meikang Qiu11 Sept 2026cs.NI—89
Hierarchical and Permutation-Invariant Feature Transformation Learning via Policy-Guided Embedding SearcharXiv:2609.10225Rui Liu, Tao Zhe, Yanyong Huang +211 Sept 2026cs.LG—89
LiteRAG: Cost-Efficient Graph-Based Retrieval-Augmented GenerationarXiv:2609.10239Daniel Alejandro Coll Tejeda, Pedro Garc\'ia L\'opez, Daniel Barcelona-Pons11 Sept 2026cs.IR—89
A-JIT: Agentic Just-In-Time Software ConstructionarXiv:2609.10248Mark Marron, Earl T. Barr11 Sept 2026cs.SE—89
DiSCo: A Distribution-First Steering and Cultural Prior Evaluation Framework for Measuring Cultural Preference Bias in LLMsarXiv:2609.10253Bhuvan Arora, Devesh Saraogi, Sravya Varada +111 Sept 2026cs.CL—89
GANDR: Claim Auditing for Verifiable Legal Answer GenerationarXiv:2609.10293Chen Qian, Yimeng Wang, Yu Chen +211 Sept 2026cs.CL—89
Learning Intrusion Response Strategies for OT SystemsarXiv:2609.10298Duc Huy Le, Rolf Stadler11 Sept 2026cs.CR—89
RiLM: Parameter-Efficient Language Modeling via Geodesic DecodingarXiv:2609.10305Fang Li11 Sept 2026cs.CL—89
One Loop, Two Gains: Can Active Learning win the Lottery for Free?arXiv:2609.10311Benedikt Tscheschner, Eduardo Veas, Marc Masana11 Sept 2026cs.LG—89

Author lists and categories are copied from the paper's own metadata (arXiv, publisher). Each paper page shows the abstract, related models and every source snapshot.