Personalization in Question Answering (QA) requires answers that are both accurate and aligned with users' background, preferences, and historical context. Existing state-of-the-art methods primarily rely on retrieval-augmented generation (RAG) solutions that construct personal context by retrieving relevant items from the user's profile. Existing methods use the user's query directly to retrieve personal documents and such strategies often lead to surface-level personalization. We propose PR2 (Personalized Retrieval-Augmented Reasoning), a reinforcement learning framework that integrates reasoning and retrieval from personal context for personalization. PR2 learns adaptive retrieval-reasoning policies, determining when to retrieve, what evidence to retrieve from user profiles, and how to incorporate it into intermediate reasoning steps. By optimizing multi-turn reasoning trajectories under a personalized reward function, the framework reinforces reasoning paths that better align with user-specific preferences and contextual signals reflected by the reward model. Extensive experiments on the LaMP-QA benchmark using three LLMs show that PR2 consistently outperforms strong baselines, achieving an average relative improvement of 8.26%-12.0% in personalized QA.

In the domain of artificial intelligence, the role of agents is rapidly expanding. An agent is an autonomous entity that perceives its environment, makes informed decisions, and takes actions to complete tasks. However, enabling agents to accurately interpret and respond to human requests-particularly through understanding human Theory of Mind (ToM)-remains a complex and pivotal challenge. ToM encompasses the ability to recognize that individuals possess distinct mental states, such as intentions, beliefs, and desires, which may differ from one's own. This cognitive faculty is fundamental to human interaction, enabling empathy and understanding across diverse contexts. As AI agents are increasingly deployed across new domains, approximating ToM becomes vital for producing appropriate and context-sensitive responses. Recent studies have explored methodologies to enhance ToM capabilities in large language models (LLMs), aiming to refine their ability to infer and track human mental states. By leveraging ToM, LLMs have been shown to approximate key aspects of human-like social cognition. However, most existing research has focused primarily on evaluating ToM within LLMs, overlooking the broader challenge of enabling AI agents to actively understand and respond to human cognitive and mental states.

As Large Language Models (LLMs) become increasingly integrated into applications demanding human-like understanding-such as mental health support, education, and social robotics-their capacity to exhibit Theory of Mind (ToM) reasoning is essential. Although previous research has evaluated LLMs' capability in ToM tasks, a critical gap remains as few studies have systematically investigated how LLMs' ToM reasoning diverges from human reasoning and the extent of these differences. This study introduces a reinforcement learning-based framework designed to bridge this gap. This approach seeks to enhance LLM alignment with human ToM reasoning, effectively narrowing the differences in their reasoning processes. Finally, future research directions to advance this field will be discussed, including strategies for developing LLMs that can better approximate human social cognition. This work lays a foundation for responsible LLM deployment, offering guidelines for applications in sensitive contexts where accurate ToM understanding is crucial.

As Large Language Models (LLMs) have rapidly advanced in social reasoning tasks, their applications have expanded to domains such as healthcare and psychology. Given the direct interaction of users with these applications, it is essential to evaluate the performance of LLMs, particularly in human-like social reasoning capabilities. While previous studies have explored human-aligned social reasoning in LLMs, they have not adequately assessed whether the generated reasoning answers stem from the LLMs' memorization of training data or their natural language understanding. In this study, we aim to address this gap by assessing the impact of training data memorization on the human-aligned social reasoning capabilities of LLMs. We introduce IR+CoT (Information Retrieval (IR) + Chain of Thought (CoT)), a framework that leverages retrieved information from input questions to fine-tune prompt templates and employs CoT methods. IR+CoT mitigates the effects of memorization and enhances the LLMs' social reasoning performance. Experiments on three LLMs, using seen (present during the training of the LLMs) and unseen (introduced post-training) questions from Reddit and Lemmy, show that IR+CoT enhances social reasoning and reduces memorization effects. This research's novelty lies in using old and new questions to assess memorization's impact on social reasoning.

A recommender system is tasked with effectively analyzing a user’s preferences and interactions to provide personalized recommendations. This calls for extracting and connecting various heterogeneous data while preserving their temporal relations. Graph neural networks (GNNs) have proven to be highly suitable in recommendation systems for connecting different types of user behavioral signals. However, they inherently lack ability to capture temporal aspects of underlying data. This shortcoming prevents them from explicating and utilizing task information, which is shown to be instrumental in many information retrieval applications. To overcome this limitation, we propose a new Task-based Graph Neural Network model (TGNN) focusing on identifying users’ underlying tasks within their temporal multi-behavior, specifically in each session. The model consists of three modules: (1) a sequential meta-path module that captures a temporal sequence of users’ behaviors; (2) a graph neural network layer that models the relationships between different information items and users into task representations; and (3) a recommendation layer that utilizes a collaborative filtering method to generate top-N recommendations based on the model’s comprehension of users’ tasks. The novelty of our approach lies in understanding users’ tasks through their temporal behavior, enabling more accurate personalization. The results of evaluative experiments on three publicly available datasets demonstrate the effectiveness of our task-based recommendation model compared to 10 baselines and indicate a promising research direction for task-oriented recommender systems.

Theory of Mind (ToM) reasoning involves understanding that others have unique mental states-like beliefs, thoughts, intentions, viewpoints, and emotions-different from one's own, and incorporating this into one's reasoning. While some research suggests that LLMs possess reasoning abilities, other studies challenge this assertion, often focusing on structured responses and overlooking the complexities of open-ended interactions. As LLMs are increasingly employed in different sectors, their ability to accurately interpret human mental states in reasoning becomes critical. For example, in psychological services, if LLMs generate reasoning responses without understanding human mental states, their answers may lack logical soundness and potentially exacerbate client distress. Therefore, understanding LLMs' ToM capabilities is crucial to ensure they deliver effective and appropriate responses in real-world scenarios. In this research, I investigate the effectiveness of incorporating questioners' viewpoints in the questions-whether posed in a rational or intuitive manner-on the generation of reasoning answers by LLMs and how these generated answers align with human-written responses. The results demonstrate that incorporating these viewpoints into the prompt instructions enhances the reasoning performance of LLMs, although the responses still fall short of being truly human-like. This research contributes to the information retrieval and generative AI community by raising awareness about the limitations of LLMs in reasoning and their alignment with human responses in this domain.

As Large Language Models (LLMs) are integrated into various sectors, ensuring their reliability and safety is crucial. This necessitates rigorous probing and auditing to maintain their effectiveness and trustworthiness in practical applications. Subjecting LLMs to varied iterations of a single query can unveil potential inconsistencies in their knowledge base or functional capacity. However, a tool for performing such audits with a easy to execute workflow, and low technical threshold is lacking. In this demo, we introduce "AuditLLM," a novel tool designed to audit the performance of various LLMs in a methodical way. AuditLLM's primary function is to audit a given LLM by deploying multiple probes derived from a single question, thus detecting any inconsistencies in the model's comprehension or performance. A robust, reliable, and consistent LLM is expected to generate semantically similar responses to variably phrased versions of the same question. Building on this premise, AuditLLM generates easily interpretable results that reflect the LLM's consistency based on a single input question provided by the user. A certain level of inconsistency has been shown to be an indicator of potential bias, hallucinations, and other issues. One could then use the output of AuditLLM to further investigate issues with the aforementioned LLM. To facilitate demonstration and practical uses, AuditLLM offers two key modes: (1) Live mode which allows instant auditing of LLMs by analyzing responses to real-time queries; and (2) Batch mode which facilitates comprehensive LLM auditing by processing multiple queries at once for in-depth analysis. This tool is beneficial for both researchers and general users, as it enhances our understanding of LLMs' capabilities in generating responses, using a standardized auditing platform.

Theory of mind (ToM) reasoning involves understanding that others have intentions, emotions, and thoughts, which is crucial for regulating one's reasoning. Although large language models (LLMs) excel in tasks such as summarization, question answering, and translation, they still face challenges with ToM reasoning, especially in open-ended questions. Despite advancements, the extent to which LLMs truly understand ToM reasoning and how closely it aligns with human ToM reasoning remains inadequately explored in open-ended scenarios. Motivated by this gap, we assess the abilities of LLMs to perceive and integrate human intentions and emotions into their ToM reasoning processes within open-ended questions. Our study utilizes posts from Reddit's ChangeMyView platform, which demands nuanced social reasoning to craft persuasive responses. Our analysis, comparing semantic similarity and lexical overlap metrics between responses generated by humans and LLMs, reveals clear disparities in ToM reasoning capabilities in open-ended questions, with even the most advanced models showing notable limitations. To enhance LLM capabilities, we implement a prompt tuning method that incorporates human intentions and emotions, resulting in improvements in ToM reasoning performance. However, despite these improvements, the enhancement still falls short of fully achieving human-like reasoning. This research highlights the deficiencies in LLMs' social reasoning and demonstrates how integrating human intentions and emotions can boost their effectiveness.

Despite the considerable advancements in modern search systems for assisting users in search tasks of varying types, support for complex tasks that call for multi-round interactions remains challenging. Identifying users’ tasks is essential to understanding their evolving information needs and search goals during search sessions to simulate and achieve real-time adaptive search retrievals; thus, it is a crucial research thrust in interactive information retrieval (IIR). While a series of descriptive and formal models have been proposed to characterize complex information search sessions, only a few focus on leveraging dynamic task features in search personalizations to support users in different task stages in an adaptive fashion. This preliminary study presents a heterogeneous graph neural network model for extracting and representing tasks to better understand users’ interactive search processes by connecting tasks with search interactions. Our approach’s novelty lies in our application of task representation learning, which enables systems to extract hidden task information from users’ search behaviors. The results of our evaluative experiments on TREC Session track data highlight the value of our proposed task representation model and illustrate a promising research direction on task-oriented intelligent systems.