No article text was recorded for this version.
negative preference optimization
Version 1 of 123 papers25 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Variance-Reduced $(\varepsilon, \delta)-$Unlearning using Forget Set Gradients
- Blog: Graph Unlearning Meets Influence-aware Negative Preference Optimization
- Blog: Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning
- Blog: Alternate Preference Optimization for Unlearning Factual Knowledge in Large Language Models
- Blog: LLM Unlearning Under the Microscope: A Full-Stack View on Methods and Metrics
- Blog: CLEAR: Character Unlearning in Textual and Visual Modalities
- Blog: SafeEraser: Enhancing Safety in Multimodal Large Language Models through Multimodal Machine Unlearning
- Blog: Can Large Language Models Truly Forget Scientific Claims? A Systematic Evaluation of Claim Level Unlearning
- Blog: From Narrow Unlearning to Emergent Misalignment: Causes, Consequences, and Containment in LLMs
- Blog: ReLearn: Unlearning via Learning for Large Language Models
- Blog: Which Retain Set Matters for LLM Unlearning? A Case Study on Entity Unlearning
- Blog: Improving LLM Safety Alignment with Dual-Objective Optimization
- Blog: Not Everything is All You Need: Toward Low-Redundant Optimization for Large Language Model Alignment
- Blog: Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning
- Blog: Opt-Out: Investigating Entity-Level Unlearning for Large Language Models via Optimal Transport
- Blog: RESTOR: Knowledge Recovery in Machine Unlearning
- Blog: Unveiling Entity-Level Unlearning for Large Language Models: A Comprehensive Analysis
- Blog: Towards Safer Large Language Models through Machine Unlearning
- Blog: On Localizing and Deleting Toxic Memories in Large Language Models
- Blog: Redirection for Erasing Memory (REM): Towards a universal unlearning method for corrupted data
- Blog: LLM Unlearning Reveals a Stronger-Than-Expected Coreset Effect in Current Benchmarks
- Blog: An Adversarial Perspective on Machine Unlearning for AI Safety
- Blog: Minimalist Concept Erasure in Generative Models
History & citation
Current · Permanent link
- This direction split from Preference Optimization for Surgical Concept Erasure (78% overlap)
- This direction split from catastrophic forgetting in unlearning (52% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_negative_preference_optimization_d11ec93c,
title = {negative preference optimization},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/negative-preference-optimization/rdr_d11ec93cbe7f4faa8f74126ee266041b},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}This exact version (version 1, frozen text):
@misc{lacuna_negative_preference_optimization_d11ec93c_v1,
title = {negative preference optimization},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/negative-preference-optimization/rdr_d11ec93cbe7f4faa8f74126ee266041b/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}