Image de couverture de livre

No Certificate for Alignment: Two Independent Impossibilities and the Pareto Frontier of Achievable Safety Guarantees

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteur principal:	Agarwal, Ayushi
Format:	Preprint
Publié:	2026
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

Towards Realistic Guarantees: A Probabilistic Certificate for SmoothLLM
par: Kumarappan, Adarsh, et autres
Publié: (2025)

Achievable Fairness on Your Data With Utility Guarantees
par: Taufiq, Muhammad Faaiz, et autres
Publié: (2024)

What, Indeed, is an Achievable Provable Guarantee for Learning-Enabled Safety Critical Systems
par: Bensalem, Saddek, et autres
Publié: (2023)

LLMs Will Always Hallucinate, and We Need to Live With This
par: Banerjee, Sourav, et autres
Publié: (2024)

Separation-Utility Pareto Frontier: An Information-Theoretic Characterization
par: Xu, Shizhou
Publié: (2026)

Federated Multi-Objective Learning with Controlled Pareto Frontiers
par: Rao, Jiansheng, et autres
Publié: (2025)

Navigating the Alignment-Calibration Trade-off: A Pareto-Superior Frontier via Model Merging
par: Hu, Tiancheng, et autres
Publié: (2025)

The Vulnerability of Language Model Benchmarks: Do They Accurately Reflect True LLM Performance?
par: Banerjee, Sourav, et autres
Publié: (2024)

BigMac: Breaking the Pareto Frontier of Compute and Memory in Multimodal LLM Training
par: Zhang, Zili, et autres
Publié: (2026)

The Pareto Frontier of Resilient Jet Tagging
par: Gambhir, Rikab, et autres
Publié: (2025)

Pushing Forward Pareto Frontiers of Proactive Agents with Behavioral Agentic Optimization
par: Yao, Yihang, et autres
Publié: (2026)

Pareto-Optimal Energy Alignment for Designing Nature-Like Antibodies
par: Wen, Yibo, et autres
Publié: (2024)

What Is The Performance Ceiling of My Classifier? Utilizing Category-Wise Influence Functions for Pareto Frontier Analysis
par: Nahin, Shahriar Kabir, et autres
Publié: (2025)

Fairness vs Performance: Characterizing the Pareto Frontier of Algorithmic Decision Systems
par: Wilms, Mieke, et autres
Publié: (2026)

General Identifiability and Achievability for Causal Representation Learning
par: Varıcı, Burak, et autres
Publié: (2023)

COM-BOM: Bayesian Exemplar Search for Efficiently Exploring the Accuracy-Calibration Pareto Frontier
par: Luo, Gaoxiang, et autres
Publié: (2025)

Pareto Multi-Objective Alignment for Language Models
par: He, Qiang, et autres
Publié: (2025)

Reinforcement Learning with Ensemble Model Predictive Safety Certification
par: Gronauer, Sven, et autres
Publié: (2024)

Causal Bandits: The Pareto Optimal Frontier of Adaptivity, a Reduction to Linear Bandits, and Limitations around Unknown Marginals
par: Liu, Ziyi, et autres
Publié: (2024)

ParetoHqD: Fast Offline Multiobjective Alignment of Large Language Models using Pareto High-quality Data
par: Gu, Haoran, et autres
Publié: (2025)

Hair-Trigger Alignment: Black-Box Evaluation Cannot Guarantee Post-Update Alignment
par: Bakman, Yavuz, et autres
Publié: (2026)

Approximating Pareto Frontiers in Stochastic Multi-Objective Optimization via Hashing and Randomization
par: Li, Jinzhao, et autres
Publié: (2026)

MPM-LLM4DSE: Reaching the Pareto Frontier in HLS with Multimodal Learning and LLM-Driven Exploration
par: Xu, Lei, et autres
Publié: (2026)

Towards Distillation Guarantees under Algorithmic Alignment for Combinatorial Optimization
par: Le, Thien, et autres
Publié: (2026)

On the Impossibility of Retrain Equivalence in Machine Unlearning
par: Yu, Jiatong, et autres
Publié: (2025)

Equivariant Frames and the Impossibility of Continuous Canonicalization
par: Dym, Nadav, et autres
Publié: (2024)

Proactive Routing to Interpretable Surrogates with Distribution-Free Safety Guarantees
par: Uddin, Iqtedar, et autres
Publié: (2026)

Lost in the Shuffle: Testing Power in the Presence of Errorful Network Vertex Labels
par: Saxena, Ayushi, et autres
Publié: (2022)

Survey of Data-driven Newsvendor: Unified Analysis and Spectrum of Achievable Regrets
par: Chen, Zhuoxin, et autres
Publié: (2024)

Achievable distributional robustness when the robust risk is only partially identified
par: Kostin, Julia, et autres
Publié: (2025)

Multi-Objective Bayesian Optimization with Independent Tanimoto Kernel Gaussian Processes for Diverse Pareto Front Exploration
par: Yong, Anabel
Publié: (2025)

Rigorous Error Certification for Neural PDE Solvers: From Empirical Residuals to Solution Guarantees
par: Mukherjee, Amartya, et autres
Publié: (2026)

Beyond the Independence Assumption: Finite-Sample Guarantees for Deep Q-Learning under $τ$-Mixing
par: Halgryn, Leon, et autres
Publié: (2026)

Impossibility Theorems for Feature Attribution
par: Bilodeau, Blair, et autres
Publié: (2022)

An Analysis of Safety Guarantees in Multi-Task Bayesian Optimization
par: Luebsen, Jannis O., et autres
Publié: (2025)

Self-Improvement Towards Pareto Optimality: Mitigating Preference Conflicts in Multi-Objective Alignment
par: Li, Moxin, et autres
Publié: (2025)

Mechanistic Interpretability of Brain-to-Speech Models Across Speech Modes
par: Maghsoudi, Maryam, et autres
Publié: (2026)

Error Bounds For Gaussian Process Regression Under Bounded Support Noise With Applications To Safety Certification
par: Reed, Robert, et autres
Publié: (2024)

Calibrate-Then-Delegate: Safety Monitoring with Risk and Budget Guarantees via Model Cascades
par: Pona, Edoardo, et autres
Publié: (2026)

Adversarial Robustness in Two-Stage Learning-to-Defer: Algorithms and Guarantees
par: Montreuil, Yannis, et autres
Publié: (2025)