Cover Image

SafetyRepro: Configuration-Conditional Rank Instability on Alignment Benchmarks

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Li, Yanhang, Fan, Zhichao, Zhuang, Zexin
Format:	Preprint
Published:	2026
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Pre-Registering the Detectable Effect: A Paired-MDE Budget for 4-bit Quantization Benchmarks, with a Pilot Audit
by: Zhuang, Zexin, et al.
Published: (2026)

Auditing Reasoning-Trace Memorization Claims after Unlearning with Head-Conditioned Canaries
by: Li, Yanhang, et al.
Published: (2026)

Auditing and Fixing Economic Validity in Tabular Foundation Models for Discrete Choice
by: Wang, Yingshuo, et al.
Published: (2026)

PID-Guided Partial Alignment for Multimodal Decentralized Federated Learning
by: Shi, Yanhang, et al.
Published: (2026)

SaLoRA: Safety-Alignment Preserved Low-Rank Adaptation
by: Li, Mingjie, et al.
Published: (2025)

Advancing LLM Safe Alignment with Safety Representation Ranking
by: Du, Tianqi, et al.
Published: (2025)

ReproMIA: A Comprehensive Analysis of Model Reprogramming for Proactive Membership Inference Attacks
by: Huang, Chihan, et al.
Published: (2026)

Quantifying Ranking Instability Across Evaluation Protocol Axes in Gene Regulatory Network Benchmarking
by: Kendiukhov, Ihor
Published: (2026)

SoSBench: Benchmarking Safety Alignment on Six Scientific Domains
by: Jiang, Fengqing, et al.
Published: (2025)

Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications
by: Wei, Boyi, et al.
Published: (2024)

Conditional Rank-Rank Regression via Deep Conditional Transformation Models
by: Wang, Xiaoyi, et al.
Published: (2026)

Interpretable Safety Alignment via SAE-Constructed Low-Rank Subspace Adaptation
by: Wang, Dianyun, et al.
Published: (2025)

MSign: An Optimizer Preventing Training Instability in Large Language Models via Stable Rank Restoration
by: Ren, Lianhai, et al.
Published: (2026)

Turning the Spell Around: Lightweight Alignment Amplification via Rank-One Safety Injection
by: Shairah, Harethah Abu, et al.
Published: (2025)

SafeTuneBed: A Toolkit for Benchmarking LLM Safety Alignment in Fine-Tuning
by: Hossain, Saad, et al.
Published: (2025)

Reducing the Safety Tax in LLM Safety Alignment with On-Policy Self-Distillation
by: Fu, Yu, et al.
Published: (2026)

Learning Discrete Latent Variable Structures with Tensor Rank Conditions
by: Chen, Zhengming, et al.
Published: (2024)

Curriculum Learning for Safety Alignment
by: Kumar, Sandeep, et al.
Published: (2026)

Proximal Ranking Policy Optimization for Practical Safety in Counterfactual Learning to Rank
by: Gupta, Shashank, et al.
Published: (2024)

Any-Depth Alignment: Unlocking Innate Safety Alignment of LLMs to Any-Depth
by: Zhang, Jiawei, et al.
Published: (2025)

Mitigating the Safety Alignment Tax with Null-Space Constrained Policy Optimization
by: Niu, Yifan, et al.
Published: (2025)

Regularized Conditional Diffusion Model for Multi-Task Preference Alignment
by: Yu, Xudong, et al.
Published: (2024)

Multilingual Safety Alignment Via Sparse Weight Editing
by: Liang, Jiaming, et al.
Published: (2026)

CausalVerse: Benchmarking Causal Representation Learning with Configurable High-Fidelity Simulations
by: Chen, Guangyi, et al.
Published: (2025)

Safety Alignment Can Be Not Superficial With Explicit Safety Signals
by: Li, Jianwei, et al.
Published: (2025)

The Blessing and Curse of Dimensionality in Safety Alignment
by: Teo, Rachel S. Y., et al.
Published: (2025)

Origin-Conditional Trajectory Encoding: Measuring Urban Configurational Asymmetries through Neural Decomposition
by: Law, Stephen, et al.
Published: (2025)

Guardrails in Logit Space: Safety Token Regularization for LLM Alignment
by: Bach, Thong, et al.
Published: (2026)

Continual Safety Alignment via Gradient-Based Sample Selection
by: Bach, Thong, et al.
Published: (2026)

Dialectics of Alignment: Harnessing Unsafe Knowledge for Dynamic Safety Routing
by: Hashemzadeh, Maryam, et al.
Published: (2026)

Alignment with Preference Optimization Is All You Need for LLM Safety
by: Alami, Reda, et al.
Published: (2024)

Rank Also Matters: Hierarchical Configuration for Mixture of Adapter Experts in LLM Fine-Tuning
by: Cong, Peizhuang, et al.
Published: (2025)

PLAN: Proactive Low-Rank Allocation for Continual Learning
by: Wang, Xiequn, et al.
Published: (2025)

Multi-Conditional Ranking with Large Language Models
by: Pezeshkpour, Pouya, et al.
Published: (2024)

Test-Time Safety Alignment
by: Saglam, Baturay, et al.
Published: (2026)

Alignment and Safety in Large Language Models: Safety Mechanisms, Training Paradigms, and Emerging Challenges
by: Lu, Haoran, et al.
Published: (2025)

BSO: Safety Alignment Is Density Ratio Matching
by: Nguyen, Tien-Phat, et al.
Published: (2026)

Multilingual Safety Alignment via Self-Distillation
by: Qin, Ruiyang, et al.
Published: (2026)

Conformational Rank Conditioned Committees for Machine Learning-Assisted Directed Evolution
by: Adler, Mia, et al.
Published: (2025)

SurvCORN: Survival Analysis with Conditional Ordinal Ranking Neural Network
by: Ridzuan, Muhammad, et al.
Published: (2024)