Reinforcement Learning on Benign Facts Amplifies Leakage of Memorized Private Data
cs.LG, cs.AI
Submitted: 2026-08-22
Updated: 2026-08-30
Terminology
Sources
- The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality
- DeepSeek-V3 Technical Report
- Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMs
- Alignment Whack-a-Mole : Finetuning Activates Verbatim Recall of Copyrighted Books in Large Language Models
- Reinforcement Learning Improves Traversal of Parametric Knowledge in LLMs
- Natural Emergent Misalignment from Reward Hacking in Production RL
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- The Invisible Leash: Why RLVR May or May Not Escape Its Origin
- Qwen3 Technical Report
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks