{"total":50,"rows":[{"id":"2511.04703","title":"Measuring what Matters: Construct Validity in Large Language Model Benchmarks","created":"2025-11-03","cited_by_count":119,"area":null},{"id":"2505.24832","title":"How much do language models memorize?","created":"2025-05-30","cited_by_count":61,"area":null},{"id":"2503.17514","title":"Language Models May Verbatim Complete Text They Were Not Explicitly Trained On","created":"2025-03-21","cited_by_count":21,"area":null},{"id":"2502.00678","title":"How Contaminated Is Your Benchmark? Quantifying Dataset Leakage in Large Language Models with Kernel Divergence","created":"2025-02-02","cited_by_count":15,"area":null},{"id":"2506.14562","title":"AlphaDecay: Module-wise Weight Decay for Heavy-Tailed Balancing in LLMs","created":"2025-06-17","cited_by_count":13,"area":null},{"id":"2407.02596","title":"Towards More Realistic Extraction Attacks: An Adversarial Perspective","created":"2024-07-02","cited_by_count":8,"area":null},{"id":"2410.10880","title":"Fine-tuning can Help Detect Pretraining Data from Large Language Models","created":"2024-10-09","cited_by_count":8,"area":null},{"id":"2503.12072","title":"Information-Guided Identification of Training Data Imprint in (Proprietary) Large Language Models","created":"2025-03-15","cited_by_count":8,"area":null},{"id":"2504.13416","title":"STAMP Your Content: Proving Dataset Membership via Watermarked Rephrasings","created":"2025-04-18","cited_by_count":8,"area":null},{"id":"2507.05622","title":"DATABench: Evaluating Dataset Auditing in Deep Learning from an Adversarial Perspective","created":"2025-07-08","cited_by_count":8,"area":null},{"id":"2510.27055","title":"Detecting Data Contamination in LLMs via In-Context Learning","created":"2025-10-30","cited_by_count":8,"area":null},{"id":"2602.04735","title":"From Data to Behavior: Predicting Unintended Model Behaviors Before Training","created":"2026-02-04","cited_by_count":8,"area":null},{"id":"2501.09653","title":"The Heap: A Contamination-Free Multilingual Code Dataset for Evaluating Large Language Models","created":"2025-01-16","cited_by_count":6,"area":null},{"id":"2506.19399","title":"Automated Detection of Pre-training Text in Black-box LLMs","created":"2025-06-24","cited_by_count":6,"area":null},{"id":"2509.14278","title":"Beyond Data Privacy: New Privacy Risks for Large Language Models","created":"2025-09-16","cited_by_count":6,"area":null},{"id":"2412.02857","title":"Measuring Fingerprints of Web-filtered Text Datasets and Fingerprint Propagation Through Training","created":"2024-12-03","cited_by_count":5,"area":null},{"id":"2505.21636","title":"The Feasibility of Topic-Based Watermarking on Academic Peer Reviews","created":"2025-05-27","cited_by_count":4,"area":null},{"id":"2509.04304","title":"Facts Fade Fast: Evaluating Memorization of Outdated Medical Knowledge in Large Language Models","created":"2025-09-04","cited_by_count":4,"area":null},{"id":"2510.09655","title":"Data Provenance Auditing of Fine-Tuned Large Language Models with a Text-Preserving Technique","created":"2025-10-07","cited_by_count":4,"area":null},{"id":"2605.28591","title":"Models That Know How Evaluations Are Designed Score Safer","created":"2026-05-27","cited_by_count":4,"area":null},{"id":"2506.06057","title":"Hey, That's My Data! Token-Only Dataset Inference in Large Language Models","created":"2025-06-06","cited_by_count":3,"area":null},{"id":"2508.01074","title":"Evading Data Provenance in Deep Neural Networks","created":"2025-08-01","cited_by_count":2,"area":null},{"id":"2508.17767","title":"ISACL: Internal State Analyzer for Copyrighted Training Data Leakage","created":"2025-08-25","cited_by_count":2,"area":null},{"id":"2511.07722","title":"Critical Confabulation: Can LLMs Hallucinate for Social Good?","created":"2025-11-11","cited_by_count":2,"area":null},{"id":"2511.20799","title":"Memories Retrieved from Many Paths: A Multi-Prefix Framework for Robust Detection of Training Data Leakage in Large Language Models","created":"2025-11-25","cited_by_count":2,"area":null},{"id":"2601.04301","title":"Quantifying the Effect of Test Set Contamination on Generative Evaluations","created":"2026-01-07","cited_by_count":2,"area":null},{"id":"2607.09692","title":"Reference-Based Distillation Detection in LLMs","created":"2026-06-19","cited_by_count":2,"area":null},{"id":"2408.02140","title":"VidModEx: Interpretable and Efficient Black Box Model Extraction for High-Dimensional Spaces","created":"2024-08-04","cited_by_count":1,"area":null},{"id":"2501.13983","title":"AdEval: Alignment-based Dynamic Evaluation to Mitigate Data Contamination in Large Language Models","created":"2025-01-23","cited_by_count":1,"area":null},{"id":"2505.18425","title":"Advertising in AI systems: Society must be vigilant","created":"2025-05-23","cited_by_count":1,"area":null},{"id":"2507.06056","title":"Data Compressibility Quantifies LLM Memorization","created":"2025-07-08","cited_by_count":1,"area":null},{"id":"2605.07878","title":"Black-box model classification under the discriminative factorization","created":"2026-05-08","cited_by_count":1,"area":null},{"id":"2606.03305","title":"The Reliability Gap in Benchmark Auditing: Distribution Shift and Scale as Failure Modes of Contamination Detection","created":"2026-06-02","cited_by_count":1,"area":null},{"id":"2506.16460","title":"Black-Box Privacy Attacks on Shared Representations in Multitask Learning","created":"2025-06-19","cited_by_count":0,"area":null},{"id":"2511.14084","title":"Observational Auditing of Label Privacy","created":"2025-11-18","cited_by_count":0,"area":null},{"id":"2512.17075","title":"Perturb Your Data: Paraphrase-Guided Training Data Watermarking","created":"2025-12-18","cited_by_count":0,"area":null},{"id":"2601.13122","title":"Responsible AI for General-Purpose Systems: Overview, Challenges, and A Path Forward","created":"2026-01-19","cited_by_count":0,"area":null},{"id":"2601.18127","title":"The Limits of AI Data Transparency Policy: Three Disclosure Fallacies","created":"2026-01-26","cited_by_count":0,"area":null},{"id":"2602.04572","title":"From Competition to Collaboration: Designing Sustainable Mechanisms Between LLMs and Online Forums","created":"2026-02-04","cited_by_count":0,"area":null},{"id":"2603.22707","title":"Detecting Non-Membership in LLM Training Data via Rank Correlations","created":"2026-03-24","cited_by_count":0,"area":null},{"id":"2605.21543","title":"Provable Joint Decontamination for Benchmarking Multiple Large Language Models","created":"2026-05-20","cited_by_count":0,"area":null},{"id":"2606.21045","title":"OVIG: Optimistic Verification of AI Training Integrity via Gradient Signals","created":"2026-06-19","cited_by_count":0,"area":null},{"id":"2606.21240","title":"DIPBox: A Multi-scale Testing Framework for Tracking Dataset Regeneration","created":"2026-06-19","cited_by_count":0,"area":null},{"id":"2606.24408","title":"Natural Identifiers for Privacy and Data Audits in Large Language Models","created":"2026-06-23","cited_by_count":0,"area":null},{"id":"2606.31991","title":"Anchored or Drifting: What Recursive Self-Generation Reveals About Training Data","created":"2026-06-30","cited_by_count":0,"area":null},{"id":"2607.12649","title":"Extractable Memorization From First Principles","created":"2026-07-14","cited_by_count":0,"area":null},{"id":"2608.07914","title":"When Is Benchmark Contamination Detectable? Information Limits and Power-Calibrated Audits","created":"2026-08-08","cited_by_count":0,"area":null},{"id":"2608.10462","title":"Calibrating Post-Training Feature Shifts for LLM Data Contamination Detection","created":"2026-08-11","cited_by_count":0,"area":null},{"id":"2609.25014","title":"Not All 4-bit Quantizers Are Equal: Deployment-Time Mitigation of PII Leakage in Fine-Tuned Small Language Models","created":"2026-08-02","cited_by_count":0,"area":null},{"id":"2610.00251","title":"The Null Is the Hard Part: Exact Tests for Memorization in Generative Models","created":"2026-09-23","cited_by_count":0,"area":null}],"next_cursor":null}