Distillation Defenses Easily Break After Reinforcement Learning
Paper • 2609.35699 • Published • 3
Alexander Panfilov
kotekjedi
AI & ML interests
None yet
Recent Activity
updated a dataset about 2 hours ago
kotekjedi/stolen-thoughts published a dataset about 2 hours ago
kotekjedi/stolen-thoughts upvoted a paper 24 days ago
Distillation Defenses Easily Break After Reinforcement LearningOrganizations
upvoted a paper 24 days ago
upvoted a paper 2 months ago
upvoted a paper 11 months ago
upvoted 5
papers about 1 year ago
Adaptive Attacks on Trusted Monitors Subvert AI Control Protocols
Paper • 2510.09462 • Published • 6
DISCO: Diversifying Sample Condensation for Efficient Model Evaluation
Paper • 2510.07959 • Published • 15
D-REX: A Benchmark for Detecting Deceptive Reasoning in Large Language Models
Paper • 2509.17938 • Published • 4
Strategic Dishonesty Can Undermine AI Safety Evaluations of Frontier LLM
Paper • 2509.18058 • Published • 12
Video models are zero-shot learners and reasoners
Paper • 2509.20328 • Published • 101
upvoted an article about 1 year ago
Article
Welcome GPT OSS, the new open-source model family from OpenAI!
- +10
reach-vb, pcuenq, lewtun, clem, Rocketknight1, clefourrier, celinah, Wauplin, marcsun13, pagezyhf, ahadnagy, joaogante
• • 514 upvoted an article over 1 year ago
Article
Reachy Mini - The Open-Source Robot for Today's and Tomorrow's AI Builders
thomwolf, matthieu-lapeyre
• • 817 upvoted 3
papers over 1 year ago
GPTailor: Large Language Model Pruning Through Layer Cutting and Stitching
Paper • 2506.20480 • Published • 7
Watermarking Degrades Alignment in Language Models: Analysis and Mitigation
Paper • 2506.04462 • Published • 2
Capability-Based Scaling Laws for LLM Red-Teaming
Paper • 2505.20162 • Published • 4