Image de couverture de livre

Safety Pretraining: Toward the Next Generation of Safe AI

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Maini, Pratyush, Goyal, Sachin, Sam, Dylan, Robey, Alex, Savani, Yash, Jiang, Yiding, Zou, Andy, Fredrikson, Matt, Lipton, Zacharcy C., Kolter, J. Zico
Format:	Preprint
Publié:	2025
Sujets:	Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

When Should We Introduce Safety Interventions During Pretraining?
par: Sam, Dylan, et autres
Publié: (2026)

Evaluating Language Model Reasoning about Confidential Information
par: Sam, Dylan, et autres
Publié: (2025)

T-MARS: Improving Visual Representations by Circumventing Text Feature Learning
par: Maini, Pratyush, et autres
Publié: (2023)

Scaling Laws for Data Filtering -- Data Curation cannot be Compute Agnostic
par: Goyal, Sachin, et autres
Publié: (2024)

Understanding Hallucinations in Diffusion Models through Mode Interpolation
par: Aithal, Sumukh K, et autres
Publié: (2024)

Diffusing Differentiable Representations
par: Savani, Yash, et autres
Publié: (2024)

TOFU: A Task of Fictitious Unlearning for LLMs
par: Maini, Pratyush, et autres
Publié: (2024)

Rethinking LLM Memorization through the Lens of Adversarial Compression
par: Schwarzschild, Avi, et autres
Publié: (2024)

Adversarial Attacks on Robotic Vision Language Action Models
par: Jones, Eliot Krzysztof, et autres
Publié: (2025)

Not All Rollouts are Useful: Down-Sampling Rollouts in LLM Reinforcement Learning
par: Xu, Yixuan Even, et autres
Publié: (2025)

Antidistillation Fingerprinting
par: Xu, Yixuan Even, et autres
Publié: (2026)

Antidistillation Sampling
par: Savani, Yash, et autres
Publié: (2025)

OpenUnlearning: Accelerating LLM Unlearning via Unified Benchmarking of Methods and Metrics
par: Dorna, Vineeth, et autres
Publié: (2025)

Predicting the Performance of Black-box LLMs through Follow-up Queries
par: Sam, Dylan, et autres
Publié: (2025)

Context-Parametric Inversion: Why Instruction Finetuning Can Worsen Context Reliance
par: Goyal, Sachin, et autres
Publié: (2024)

D-REX: A Benchmark for Detecting Deceptive Reasoning in Large Language Models
par: Krishna, Satyapriya, et autres
Publié: (2025)

Improving Alignment and Robustness with Circuit Breakers
par: Zou, Andy, et autres
Publié: (2024)

Inference Optimal VLMs Need Fewer Visual Tokens and More Parameters
par: Li, Kevin Y., et autres
Publié: (2024)

Finetuning CLIP to Reason about Pairwise Differences
par: Sam, Dylan, et autres
Publié: (2024)

Toward Understanding the Transferability of Adversarial Suffixes in Large Language Models
par: Ball, Sarah, et autres
Publié: (2025)

Looking beyond the next token
par: Thankaraj, Abitha, et autres
Publié: (2025)

Transferable Adversarial Attacks on Black-Box Vision-Language Models
par: Hu, Kai, et autres
Publié: (2025)

Bayesian Neural Networks with Domain Knowledge Priors
par: Sam, Dylan, et autres
Publié: (2024)

Adaptive Data Optimization: Dynamic Sample Selection with Scaling Laws
par: Jiang, Yiding, et autres
Publié: (2024)

Mimetic Initialization of MLPs
par: Trockman, Asher, et autres
Publié: (2026)

AcceleratedLiNGAM: Learning Causal DAGs at the speed of GPUs
par: Akinwande, Victor, et autres
Publié: (2024)

Peeking Behind Closed Doors: Risks of LLM Evaluation by Private Data Curators
par: Bansal, Hritik, et autres
Publié: (2025)

AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents
par: Andriushchenko, Maksym, et autres
Publié: (2024)

Comparing AI Agents to Cybersecurity Professionals in Real-World Penetration Testing
par: Lin, Justin W., et autres
Publié: (2025)

FUSE-ing Language Models: Zero-Shot Adapter Discovery for Prompt Optimization Across Tokenizers
par: Williams, Joshua Nathaniel, et autres
Publié: (2024)

Equilibrium Reasoners: Learning Attractors Enables Scalable Reasoning
par: Huang, Benhao, et autres
Publié: (2026)

Why is SAM Robust to Label Noise?
par: Baek, Christina, et autres
Publié: (2024)

Automated Black-box Prompt Engineering for Personalized Text-to-Image Generation
par: He, Yutong, et autres
Publié: (2024)

Computing Low-Entropy Couplings for Large-Support Distributions
par: Sokota, Samuel, et autres
Publié: (2024)

Security Challenges in AI Agent Deployment: Insights from a Large Scale Public Competition
par: Zou, Andy, et autres
Publié: (2025)

From Entropy to Epiplexity: Rethinking Information for Computationally Bounded Intelligence
par: Finzi, Marc, et autres
Publié: (2026)

Defining and measuring homicide rates for birth cohorts: Methodological and theoretical challenges and solutions
par: Jason Robey, et autres
Publié: (2025)

One-Step Diffusion Distillation via Deep Equilibrium Models
par: Geng, Zhengyang, et autres
Publié: (2023)

STAMP Your Content: Proving Dataset Membership via Watermarked Rephrasings
par: Rastogi, Saksham, et autres
Publié: (2025)

Existing Large Language Model Unlearning Evaluations Are Inconclusive
par: Feng, Zhili, et autres
Publié: (2025)