Cover Image

Beyond Sharp Minima: Robust LLM Unlearning via Feedback-Guided Multi-Point Optimization

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Wu, Wenhan, Liu, Zheyuan, Gao, Chongyang, Wang, Ren, Ding, Kaize
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

GNN-as-Judge: Unleashing the Power of LLMs for Graph Learning with GNN Feedback
by: Xu, Ruiyao, et al.
Published: (2026)

POLO: Preference-Guided Multi-Turn Reinforcement Learning for Lead Optimization
by: Wang, Ziqing, et al.
Published: (2025)

Certified Signed Graph Unlearning
by: Zhao, Junpeng, et al.
Published: (2025)

Beyond Superficial Unlearning: Sharpness-Aware Robust Erasure of Hallucinations in Multimodal LLMs
by: Fang, Xianya, et al.
Published: (2026)

Tackling Fake Forgetting through Uncertainty Quantification
by: Shi, Yingdan, et al.
Published: (2025)

A Function-Centric Perspective on Flat and Sharp Minima
by: Mason-Williams, Israel, et al.
Published: (2025)

BLUR: A Benchmark for LLM Unlearning Robust to Forget-Retain Overlap
by: Hu, Shengyuan, et al.
Published: (2025)

Distribution Preference Optimization: A Fine-grained Perspective for LLM Unlearning
by: Qin, Kai, et al.
Published: (2025)

Breaking the Trilemma of Privacy, Utility, Efficiency via Controllable Machine Unlearning
by: Liu, Zheyuan, et al.
Published: (2023)

Harmonizing Multi-Objective LLM Unlearning via Unified Domain Representation and Bidirectional Logit Distillation
by: Zhong, Yisheng, et al.
Published: (2026)

A Flat Minima Perspective on Understanding Augmentations and Model Robustness
by: Yoo, Weebum, et al.
Published: (2025)

Machine Unlearning in Generative AI: A Survey
by: Liu, Zheyuan, et al.
Published: (2024)

DP-FedPGN: Finding Global Flat Minima for Differentially Private Federated Learning via Penalizing Gradient Norm
by: Liu, Junkang, et al.
Published: (2025)

Towards Mitigating Excessive Forgetting in LLM Unlearning via Entanglement-Guidance with Proxy Constraint
by: Liu, Zhihao, et al.
Published: (2025)

Robust LLM Unlearning with MUDMAN: Meta-Unlearning with Disruption Masking And Normalization
by: Sondej, Filip, et al.
Published: (2025)

LLM Unlearning via Neural Activation Redirection
by: Shen, William F., et al.
Published: (2025)

Sharpe Ratio-Guided Active Learning for Preference Optimization in RLHF
by: Belakaria, Syrine, et al.
Published: (2025)

Are Flat Minima an Illusion?
by: Bennett, Michael Timothy
Published: (2026)

Population-Aligned Persona Generation for LLM-based Social Simulation
by: Hu, Zhengyu, et al.
Published: (2025)

Towards the Connection between Activation Sparsity and Flat Minima
by: Peng, Ze, et al.
Published: (2026)

MolMem: Memory-Augmented Agentic Reinforcement Learning for Sample-Efficient Molecular Optimization
by: Wang, Ziqing, et al.
Published: (2026)

Zeroth-Order Optimization Finds Flat Minima
by: Zhang, Liang, et al.
Published: (2025)

Efficient Machine Unlearning via Influence Approximation
by: Liu, Jiawei, et al.
Published: (2025)

Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning
by: Fan, Chongyu, et al.
Published: (2024)

Agentic Unlearning: When LLM Agent Meets Machine Unlearning
by: Wang, Bin, et al.
Published: (2026)

Unlearned but Not Forgotten: Data Extraction after Exact Unlearning in LLM
by: Wu, Xiaoyu, et al.
Published: (2025)

Robust LLM Alignment via Distributionally Robust Direct Preference Optimization
by: Xu, Zaiyan, et al.
Published: (2025)

Re-understanding Graph Unlearning through Memorization
by: Ding, Pengfei, et al.
Published: (2026)

On Large Language Model Continual Unlearning
by: Gao, Chongyang, et al.
Published: (2024)

Conformal Feedback Alignment: Quantifying Answer-Level Reliability for Robust LLM Alignment
by: Chen, Tiejin, et al.
Published: (2026)

Robust Guided Diffusion for Offline Black-Box Optimization
by: Chen, Can Sam, et al.
Published: (2024)

LLM Unlearning via Loss Adjustment with Only Forget Data
by: Wang, Yaxuan, et al.
Published: (2024)

TGPO: Tree-Guided Preference Optimization for Robust Web Agent Reinforcement Learning
by: Chen, Ziyuan, et al.
Published: (2025)

Influence-Guided Symbolic Regression: Scientific Discovery via LLM-Driven Equation Search with Granular Feedback
by: Saveliev, Evgeny S., et al.
Published: (2026)

Exploring Criteria of Loss Reweighting to Enhance LLM Unlearning
by: Yang, Puning, et al.
Published: (2025)

Step-by-Step Causality: Transparent Causal Discovery with Multi-Agent Tree-Query and Adversarial Confidence Estimation
by: Ding, Ziyi, et al.
Published: (2026)

Towards Aligned Data Removal via Twin Machine Unlearning
by: Ji, Haoxuan, et al.
Published: (2024)

DRAGON: Guard LLM Unlearning in Context via Negative Detection and Reasoning
by: Wang, Yaxuan, et al.
Published: (2025)

Beyond Superficial Forgetting: Thorough Unlearning through Knowledge Density Estimation and Block Re-insertion
by: Guo, Feng, et al.
Published: (2025)

CluCERT: Certifying LLM Robustness via Clustering-Guided Denoising Smoothing
by: Wang, Zixia, et al.
Published: (2025)