Imagen de Portada

The Alignment Auditor: A Bayesian Framework for Verifying and Refining LLM Objectives

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Bou, Matthieu, Patel, Nyal, Jagota, Arjun, Krishna, Satyapriya, Parbhoo, Sonali
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Learning from Failures: Understanding LLM Alignment through Failure-Aware Inverse RL
por: Patel, Nyal, et al.
Publicado: (2025)

Insights from the Inverse: Reconstructing LLM Training Goals Through Inverse Reinforcement Learning
por: Joselowitz, Jared, et al.
Publicado: (2024)

On Almost Surely Safe Alignment of Large Language Models at Inference-Time
por: Ji, Xiaotong, et al.
Publicado: (2025)

References Improve LLM Alignment in Non-Verifiable Domains
por: Shi, Kejian, et al.
Publicado: (2026)

Tree-Based Leakage Inspection and Control in Concept Bottleneck Models
por: Ragkousis, Angelos, et al.
Publicado: (2024)

Uncovering Cross-Objective Interference in Multi-Objective Alignment
por: Lu, Yining, et al.
Publicado: (2026)

In-Context Explainers: Harnessing LLMs for Explaining Black Box Models
por: Kroeger, Nicholas, et al.
Publicado: (2023)

Improving LLM Safety Alignment with Dual-Objective Optimization
por: Zhao, Xuandong, et al.
Publicado: (2025)

Why LLMs Fail at Causal Discovery and How Interventional Agents Escape
por: Roy, Amartya, et al.
Publicado: (2026)

Causal Bayesian Optimization with Unknown Graphs
por: Durand, Jean, et al.
Publicado: (2025)

Drifting Objectives for Refining Discrete Diffusion Language Models
por: Oba, Daisuke, et al.
Publicado: (2026)

Importing Phantoms: Measuring LLM Package Hallucination Vulnerabilities
por: Krishna, Arjun, et al.
Publicado: (2025)

Compress the Context, Keep the Commitments: A Formal Framework for Verifiable LLM Context Compression
por: Trukhina, Natalia, et al.
Publicado: (2026)

Robust Multi-Objective Preference Alignment with Online DPO
por: Gupta, Raghav, et al.
Publicado: (2025)

Multi-Objective Alignment of Language Models for Personalized Psychotherapy
por: Beikzadeh, Mehrab, et al.
Publicado: (2026)

Discovering Implicit Large Language Model Alignment Objectives
por: Chen, Edward, et al.
Publicado: (2026)

VerifierQ: Enhancing LLM Test Time Compute with Q-Learning-based Verifiers
por: Qi, Jianing, et al.
Publicado: (2024)

Solving the Inverse Alignment Problem for Efficient RLHF
por: Krishna, Shambhavi, et al.
Publicado: (2024)

DualDiffusion: A Speculative Decoding Strategy for Masked Diffusion Models
por: Goyal, Satyam, et al.
Publicado: (2026)

Blockwise Advantage Estimation for Multi-Objective RL with Verifiable Rewards
por: Pavlenko, Kirill, et al.
Publicado: (2026)

UNA: A Unified Supervised Framework for Efficient LLM Alignment Across Feedback Types
por: Wang, Zhichao, et al.
Publicado: (2024)

Reward-free Alignment for Conflicting Objectives
por: Chen, Peter, et al.
Publicado: (2026)

Learning to Optimize Multi-Objective Alignment Through Dynamic Reward Weighting
por: Lu, Yining, et al.
Publicado: (2025)

De Jure: Iterative LLM Self-Refinement for Structured Extraction of Regulatory Rules
por: Guliani, Keerat, et al.
Publicado: (2026)

Multi-Task GRPO: Reliable LLM Reasoning Across Tasks
por: Ramesh, Shyam Sundhar, et al.
Publicado: (2026)

Do regularization methods for shortcut mitigation work as intended?
por: Hong, Haoyang, et al.
Publicado: (2025)

Rewarding Progress: Scaling Automated Process Verifiers for LLM Reasoning
por: Setlur, Amrith, et al.
Publicado: (2024)

Refined Direct Preference Optimization with Synthetic Data for Behavioral Alignment of LLMs
por: Gallego, Víctor
Publicado: (2024)

OrthAlign: Orthogonal Subspace Decomposition for Non-Interfering Multi-Objective Alignment
por: Lin, Liang, et al.
Publicado: (2025)

More Than a Score: Probing the Impact of Prompt Specificity on LLM Code Generation
por: Zi, Yangtian, et al.
Publicado: (2025)

Pareto Multi-Objective Alignment for Language Models
por: He, Qiang, et al.
Publicado: (2025)

Mixture of Attentions For Speculative Decoding
por: Zimmer, Matthieu, et al.
Publicado: (2024)

Bayesian Inverse Transition Learning: Learning Dynamics From Near-Optimal Trajectories
por: Benac, Leo, et al.
Publicado: (2024)

Self-Improvement Towards Pareto Optimality: Mitigating Preference Conflicts in Multi-Objective Alignment
por: Li, Moxin, et al.
Publicado: (2025)

Implicit Multimodal Alignment: On the Generalization of Frozen LLMs to Multimodal Inputs
por: Shukor, Mustafa, et al.
Publicado: (2024)

Refining Input Guardrails: Enhancing LLM-as-a-Judge Efficiency Through Chain-of-Thought Fine-Tuning and Alignment
por: Rad, Melissa Kazemi, et al.
Publicado: (2025)

SparsePO: Controlling Preference Alignment of LLMs via Sparse Token Masks
por: Christopoulou, Fenia, et al.
Publicado: (2024)

Concept-driven Off Policy Evaluation
por: Majumdar, Ritam, et al.
Publicado: (2024)

Simultaneous Multi-objective Alignment Across Verifiable and Non-verifiable Rewards
por: Shen, Yiran, et al.
Publicado: (2025)

Guarantee Regions for Local Explanations
por: Havasi, Marton, et al.
Publicado: (2024)