Although large language models (LLMs) are increasingly trained using human feedback for safety and alignment with human values, alignment decisions often overlook human social diversity. This study examines how incorporating pluralistic values affects LLM behavior by systematically evaluating demographic variation and design parameters in the alignment pipeline. We collect alignment data from US and German participants (N = 1,095 participants, 27,375 ratings) who rated LLM responses across five dimensions: Toxicity, Emotional Awareness (EA), Sensitivity, Stereotypical Bias, and Helpfulness. We fine-tuned multiple Large Language Models and Large Reasoning Models using preferences from different social groups while varying rating scales, disagreement handling methods, and optimization techniques. The results revealed systematic demographic effects: male participants rated responses 18% less toxic than female participants; conservative and Black participants rated responses 27.9% and 44% higher on EA than liberal and White participants, respectively. Models fine-tuned on group-specific preferences exhibited distinct behaviors. Technical design choices showed strong effects: the preservation of rater disagreement achieved roughly 53% greater toxicity reduction than majority voting, and 5-point scales yielded about 22% more reduction than binary formats; and Direct Preference Optimization (DPO) consistently outperformed Group Relative Policy Optimization (GRPO) in multi-value optimization. These findings represent a preliminary step in answering a critical question: How should alignment balance expert-driven and user-driven signals to ensure both safety and fair representation?

Social media platforms increasingly employ proactive moderation techniques, such as detecting and curbing toxic and uncivil comments, to prevent the spread of harmful content. Despite these efforts, such approaches are often criticized for creating a climate of censorship and failing to address the underlying causes of uncivil behavior. Our work makes both theoretical and practical contributions by proposing and evaluating two types of emotion monitoring dashboards to users' emotional awareness and mitigate hate speech. In a study involving 211 participants, we evaluate the effects of the two mechanisms on user commenting behavior and emotional experiences. The results reveal that these interventions effectively increase users' awareness of their emotional states and reduce hate speech. However, our findings also indicate potential unintended effects, including increased expression of negative emotions (Angry, Fear, and Sad) when discussing sensitive issues. These insights provide a basis for further research on integrating proactive emotion regulation tools into social media platforms to foster healthier digital interactions.

Large language models (LLMs), exemplified by ChatGPT, have gained considerable attention for their excellent natural language processing capabilities. Nonetheless, these LLMs present many challenges, particularly in the realm of trustworthiness. Therefore, ensuring the trustworthiness of LLMs emerges as an important topic. This paper introduces TrustLLM, a comprehensive study of trustworthiness in LLMs, including principles for different dimensions of trustworthiness, established benchmark, evaluation, and analysis of trustworthiness for mainstream LLMs, and discussion of open challenges and future directions. Specifically, we first propose a set of principles for trustworthy LLMs that span eight different dimensions. Based on these principles, we further establish a benchmark across six dimensions including truthfulness, safety, fairness, robustness, privacy, and machine ethics. We then present a study evaluating 16 mainstream LLMs in TrustLLM, consisting of over 30 datasets. Our findings firstly show that in general trustworthiness and utility (i.e., functional effectiveness) are positively related. Secondly, our observations reveal that proprietary LLMs generally outperform most open-source counterparts in terms of trustworthiness, raising concerns about the potential risks of widely accessible open-source LLMs. However, a few open-source LLMs come very close to proprietary ones. Thirdly, it is important to note that some LLMs may be overly calibrated towards exhibiting trustworthiness, to the extent that they compromise their utility by mistakenly treating benign prompts as harmful and consequently not responding. Finally, we emphasize the importance of ensuring transparency not only in the models themselves but also in the technologies that underpin trustworthiness. Knowing the specific trustworthy technologies that have been employed is crucial for analyzing their effectiveness.

According to real estate data firm Co-Star, over 90 million square feet of retail space is slated to close this year, leading observers to point to an obvious truth: empathy matters in customer service. Getting it right is another story. When businesses are out of touch with consumer needs, consumers stop buying and stores start dying. Enter "affective computing," an area of research involving machines that can read and display emotional intelligence, with applications as far ranging from preventative medicine to music lessons and every commercial sector in between. The retail industry isn't the only one eying "emotion AI" as a potential savior from digital disruption, but the physical spaces that characterize the retail experience ares providing innovators with a ripe venue to demonstrate the power that capturing and understanding customer sentiment can have.

Dogs have a toddler's level of emotional awareness and understanding, new research suggests. The study shows that dogs display a level of emotional intelligence comparable with that of a human two to three-year-old. Evidence from recent brain scan studies strengthens the case for banning scientific tests carried out on man's best friend, the researchers claim. Dogs have a toddler's level of emotional awareness and understanding, new research suggests. The team conducted a review of research involving functional magnetic resonance imaging (MRI) brain scans of dogs.

Artificial intelligence that reads and responds to our emotions is the killer app of the digital economy. It will make customers and employees happier--as long as it learns to respect our boundaries. When psychologist Dr. Paul Ekman visited the Fore tribe in the highlands of Papua New Guinea in 1967, he probably didn't imagine that his work would become the foundation for some of the latest developments in artificial intelligence (AI). After studying the tribe, which was still living in the preliterate state it had been in since the Stone Age, Ekman believed he had found the blueprint for a set of universal human emotions and related expressions that crossed cultures and were present in all humans. A decade later he created the Facial Action Coding System, a comprehensive tool for objectively measuring facial movement.

When psychologist Dr. Paul Ekman visited the Fore tribe in the highlands of Papua New Guinea in 1967, he probably didn't imagine that his work would become the foundation for some of the latest developments in artificial intelligence (AI). After studying the tribe, which was still living in the preliterate state it had been in since the Stone Age, Ekman believed he had found the blueprint for a set of universal human emotions and related expressions that crossed cultures and were present in all humans. A decade later he created the Facial Action Coding System, a comprehensive tool for objectively measuring facial movement. Ekman's work has been used by the FBI and police departments to identify the seeds of violent behavior in nonverbal expressions of sentiment. He has also developed the online Atlas of Emotions at the behest of the Dalai Lama.

After studying the tribe, which was still living in the preliterate state it had been in since the Stone Age, Ekman believed he had found the blueprint for a set of universal human emotions and related expressions that crossed cultures and were present in all humans. A decade later he created the Facial Action Coding System, a comprehensive tool for objectively measuring facial movement. Ekman's work has been used by the FBI and police departments to identify the seeds of violent behavior in nonverbal expressions of sentiment. He has also developed the online Atlas of Emotions at the behest of the Dalai Lama. And today his research is being used to teach computer systems how to feel.