Copertina

Exploring Criteria of Loss Reweighting to Enhance LLM Unlearning

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Yang, Puning, Wang, Qizhou, Huang, Zhuo, Liu, Tongliang, Zhang, Chengqi, Han, Bo
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Distinguishable Deletion: Unifying Knowledge Erasure and Refusal for Large Language Model Unlearning
di: Yang, Puning, et al.
Pubblicazione: (2026)

Towards Effective Evaluations and Comparisons for LLM Unlearning Methods
di: Wang, Qizhou, et al.
Pubblicazione: (2024)

FaLW: A Forgetting-aware Loss Reweighting for Long-tailed Unlearning
di: Yu, Liheng, et al.
Pubblicazione: (2026)

Is Gradient Ascent Really Necessary? Memorize to Forget for Machine Unlearning
di: Huang, Zhuo, et al.
Pubblicazione: (2026)

On the Necessity of Output Distribution Reweighting for Effective Class Unlearning
di: Ebrahimpour-Boroojeny, Ali, et al.
Pubblicazione: (2025)

Explainable LLM Unlearning Through Reasoning
di: Liao, Junfeng, et al.
Pubblicazione: (2026)

A Huber Loss Minimization Approach to Byzantine Robust Federated Learning
di: Zhao, Puning, et al.
Pubblicazione: (2023)

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

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

Enhancing Adversarial Training via Reweighting Optimization Trajectory
di: Huang, Tianjin, et al.
Pubblicazione: (2023)

Post-Training as Reweighting: A Stochastic View of Reasoning Trajectories in Language Models
di: Bu, Dake, et al.
Pubblicazione: (2025)

Dynamic Loss-Based Sample Reweighting for Improved Large Language Model Pretraining
di: Sow, Daouda, et al.
Pubblicazione: (2025)

GEAR: Granularity-Adaptive Advantage Reweighting for LLM Agents via Self-Distillation
di: Li, Sijia, et al.
Pubblicazione: (2026)

IFFair: Influence Function-driven Sample Reweighting for Fair Classification
di: Yang, Jingran, et al.
Pubblicazione: (2025)

Rethinking Loss Reweighting for Imbalance Learning as an Inverse Problem: A Neural Collapse Point of View
di: Wang, Jinping, et al.
Pubblicazione: (2026)

Beyond Losses Reweighting: Empowering Multi-Task Learning via the Generalization Perspective
di: Phan, Hoang, et al.
Pubblicazione: (2022)

Beyond Squared Error: Exploring Loss Design for Enhanced Training of Generative Flow Networks
di: Hu, Rui, et al.
Pubblicazione: (2024)

Unlearners Can Lie: Evaluating and Improving Honesty in LLM Unlearning
di: Gu, Renjie, et al.
Pubblicazione: (2026)

Unlearning Information Bottleneck: Machine Unlearning of Systematic Patterns and Biases
di: Han, Ling, et al.
Pubblicazione: (2024)

MeGU: Machine-Guided Unlearning with Target Feature Disentanglement
di: Wang, Haoyu, et al.
Pubblicazione: (2026)

Efficient Knowledge Graph Unlearning with Zeroth-order Information
di: Xiao, Yang, et al.
Pubblicazione: (2025)

DUET: Distilled LLM Unlearning from an Efficiently Contextualized Teacher
di: Zhong, Yisheng, et al.
Pubblicazione: (2026)

Filter-then-Weight: Online Data Selection and Reweighting for LLM Fine-Tuning
di: Wang, Fangxin, et al.
Pubblicazione: (2026)

AEGIS: Adversarial Target-Guided Retention-Data-Free Robust Concept Erasure from Diffusion Models
di: Li, Fengpeng, et al.
Pubblicazione: (2026)

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

LLM Unlearning with LLM Beliefs
di: Li, Kemou, et al.
Pubblicazione: (2025)

Discovering and Reasoning of Causality in the Hidden World with Large Language Models
di: Liu, Chenxi, et al.
Pubblicazione: (2024)

More is not always better? Enhancing Many-Shot In-Context Learning with Differentiated and Reweighting Objectives
di: Zhang, Xiaoqing, et al.
Pubblicazione: (2025)

DreamPRM: Domain-Reweighted Process Reward Model for Multimodal Reasoning
di: Cao, Qi, et al.
Pubblicazione: (2025)

eCIL-MU: Embedding based Class Incremental Learning and Machine Unlearning
di: Zuo, Zhiwei, et al.
Pubblicazione: (2024)

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

LLM Unlearning Without an Expert Curated Dataset
di: Zhu, Xiaoyuan, et al.
Pubblicazione: (2025)

Residual Reweighted Conformal Prediction for Graph Neural Networks
di: Zhang, Zheng, et al.
Pubblicazione: (2025)

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

Erase then Rectify: A Training-Free Parameter Editing Approach for Cost-Effective Graph Unlearning
di: Yang, Zhe-Rui, et al.
Pubblicazione: (2024)

Enhancing Interpretability Through Loss-Defined Classification Objective in Structured Latent Spaces
di: Geissler, Daniel, et al.
Pubblicazione: (2024)

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

Randomized Antipodal Search Done Right for Data Pareto Improvement of LLM Unlearning
di: Liu, Ziwen, et al.
Pubblicazione: (2026)

Rethinking Data Curation in LLM Training: Online Reweighting Offers Better Generalization than Offline Methods
di: Zhao, Wanru, et al.
Pubblicazione: (2026)

Exploring Incremental Unlearning: Techniques, Challenges, and Future Directions
di: Qureshi, Sadia, et al.
Pubblicazione: (2025)