Differences

This shows you the differences between two versions of the page.

--- ml:model_editing_and_unlearning [2025/07/02 21:43] – [Model Editing and Unlearning] jmflanig
+++ ml:model_editing_and_unlearning [2025/07/07 07:13] (current) – [In NLP or LLMs] jmflanig
@@ Line 1: / Line 1: @@
 ====== Model Editing and Unlearning ======
-//Model editing// is where a model, such as a large language model, is "edited" to change the facts in the model.  //Machine unlearning// is where a trained model is adjusted to "remove" one or more datapoints (or classes of datapoints, such as everything about bioweapons) that were used to train the model, so that it behaves like a model that was trained without those datapoints.
+//Model editing// is where a model, such as a large language model, is "edited" to change the facts in the model.  //Machine unlearning// is where a trained model is adjusted to "remove" one or more datapoints that were used to train the model, so that it behaves like a model that was trained without those datapoints.  The datapoints to remove can either be specific datapoints from the training set, or classes of datapoints, such as all datapoints about bioweapons.
 ===== Model Editing =====
@@ Line 15: / Line 15: @@
   * [[https://arxiv.org/pdf/2306.03558|Xu et al 2023 - Machine Unlearning: A Survey]]
   * [[https://arxiv.org/pdf/2405.07406|Wang et al 2024 - Machine Unlearning: A Comprehensive Survey]]
+  * [[https://arxiv.org/pdf/2407.20516|Liu et al 2024 - Machine Unlearning in Generative AI: A Survey]]
   * **Paper lists**
     * [[https://github.com/chrisliu298/awesome-llm-unlearning|Awesome LLM Unlearning]]
   * ** For NLP or LLMs **
+    * [[https://arxiv.org/pdf/2402.08787.pdf|Liu et al 2024 - Rethinking Machine Unlearning for Large Language Models]] (This is also a survey paper.)
     * [[https://arxiv.org/pdf/2503.01854|Geng et al 2025 - A Comprehensive Survey of Machine Unlearning Techniques for Large Language Models]]
@@ Line 29: / Line 31: @@
   * [[https://arxiv.org/pdf/2402.08787.pdf|Liu et al 2024 - Rethinking Machine Unlearning for Large Language Models]]
   * [[https://arxiv.org/pdf/2403.03329.pdf|Thaker et al 2024 - Guardrail Baselines for Unlearning in LLMs]]
+  * [[https://arxiv.org/pdf/2410.02760|Gandikota et al 2024 - Erasing Conceptual Knowledge from Language Models]]
   * [[https://arxiv.org/pdf/2505.22586|Gur-Arieh et al 2025 - Precise In-Parameter Concept Erasure in Large Language Models]]