Kullback-Leiber divergence has been widely used in Knowledge Distillation (KD) to compress Large Language Models (LLMs). Contrary to prior assertions that reverse Kullback-Leibler (RKL) divergence is mode-seeking and thus preferable over the mean-seeking forward Kullback-Leibler (FKL) divergence, this study empirically and theoretically demonstrates that neither mode-seeking nor mean-seeking properties manifest in KD for LLMs. Instead, RKL and FKL are found to share the same optimization objective and both converge after a sufficient number of epochs. However, due to practical constraints, LLMs are seldom trained for such an extensive number of epochs. Meanwhile, we further find that RKL focuses on the tail part of the distributions, while FKL focuses on the head part at the beginning epochs. Consequently, we propose a simple yet effective Adaptive Kullback-Leiber (AKL) divergence method, which adaptively allocates weights to combine FKL and RKL. Metric-based and GPT-4-based evaluations demonstrate that the proposed AKL outperforms the baselines across various tasks and improves the diversity and quality of generated responses.

The COVID-19 pandemic has disproportionately impacted the lives of minorities, such as members of the LGBTQ community (lesbian, gay, bisexual, transgender, and queer) due to pre-existing social disadvantages and health disparities. Although extensive research has been carried out on the impact of the COVID-19 pandemic on different aspects of the general population's lives, few studies are focused on the LGBTQ population. In this paper, we develop and evaluate two sets of machine learning classifiers using a pre-pandemic and a during-pandemic dataset to identify Twitter posts exhibiting minority stress, which is a unique pressure faced by the members of the LGBTQ population due to their sexual and gender identities. We demonstrate that our best pre- and during-pandemic models show strong and stable performance for detecting posts that contain minority stress. We investigate the linguistic differences in minority stress posts across pre- and during-pandemic periods. We find that anger words are strongly associated with minority stress during the COVID-19 pandemic. We explore the impact of the pandemic on the emotional states of the LGBTQ population by adopting propensity score-based matching to perform a causal analysis. The results show that the LGBTQ population have a greater increase in the usage of cognitive words and worsened observable attribute in the usage of positive emotion words than the group of the general population with similar pre-pandemic behavioral attributes. Our findings have implications for the public health domain and policy-makers to provide adequate support, especially with respect to mental health, to the LGBTQ population during future crises.

Walmart has revealed it is using fully driverless trucks to bring groceries from a fulfillment center to one of its Arkansas supermarkets, in a move that will cut costs and address the ongoing labor shortage affecting retail supply chains. Twelve hours a day, apair of trucks are running on a seven-mile loop of public roads from a fulfillment center to the Walmart on Regional Airport Boulevard in Bentonville, Arkansas, where the mega-retailer is headquartered. From there customers can conveniently pick up their orders. Walmart started driverless deliveries in August using autonomous trucks from Palo Alto, California-based start-up Gatik, but waited to make the announcement until Monday, after two months of incident-free deliveries. The trucking industry has faced a record worker shortage since the pandemic started, Chris Spear, president of the American Trucking Associations, told CNN, with 80,000 drivers still needed.

Walmart will begin testing an autonomous vehicle delivery service this year that will allow customers to place orders online and have their groceries delivered by a driverless car. The pilot program is being launched in Austin, Texas; Miami, Florida: and Washington D.C. It's a partnership between the $560-billion mega-retailer and Ford, which will provide Ford Escape hybrids outfitted with Argo AI technology to make the deliveries. Argo AI, a co-venture between Ford and Volkswagen, will provide the cloud-based infrastructure to schedule drop-offs and safely route orders. Focusing on those three metro areas will show'the potential for autonomous vehicle delivery services at scale,' said Argo AI founder Bryan Salesky. Initial integration testing is expected to begin later this year, the companies said, and the service initially will be limited to specific areas in each city before being expanded over time.