The undesired memorization of sensitive information by Large Language Models (LLMs) has emphasized the need for safety mechanisms that can regulate model behavior. This has led to the development of machine unlearning techniques that enable models to precisely forget sensitive and unwanted information. For machine unlearning, first-order and second-order optimizer-based methods have shown significant progress in enabling LLMs to forget targeted information. However, in doing so, these approaches often compromise the model's original capabilities, resulting in unlearned models that struggle to retain their prior knowledge and overall utility (Liu et al., 2024b). To address this, we propose Selective Influence Machine Unlearning (SIMU), a two-step framework that enhances second-order optimizer-based unlearning by selectively updating only the critical neurons responsible for encoding the forget-set. By constraining updates to these targeted neurons, SIMU achieves comparable unlearning efficacy while substantially outperforming current methods in retaining the model's original knowledge.

The Unlearning Sensitive Content from Large Language Models task aims to remove targeted datapoints from trained models while minimally affecting their general knowledge. In our work, we leverage parameter-efficient, gradient-based unlearning using low-rank (LoRA) adaptation and layer-focused fine-tuning. To further enhance unlearning effectiveness, we employ data chunking, splitting forget data into disjoint partitions and merging them with cyclically sampled retain samples at a pre-defined ratio. Our task-agnostic method achieves an outstanding forget-retain balance, ranking first on leaderboards and significantly outperforming baselines and competing systems.

Smartphone manufacturer Azumi Mobile, being new to the U.S. market, wanted to provide a customer experience journey that outshined its competitors. To achieve that goal, the company implemented new artificial intelligence-driven self-service tech that better equipped customers to solve problems on their own. Azumi Mobile joined ranks with DeviceBits, a Columbus, Ohio-based company that merges artificial intelligence, machine learning and predictive analytics to help businesses provide customer service through chatbots and interactive content that enables customers to help themselves. Last year, Apple introduced an iPhone and iPad support app to promote self-service, and the company also uses predictive analytics to improve OS functionality and the customer experience. DeviceBits, though, saw a unique opportunity that was being omitted in the area of customer service; specifically, the customer experience post-purchase, said CEO JC Ramey.