Large Language Models (LLMs) have demonstrated remarkable human-like capabilities, yet their ability to replicate a specific individual remains underexplored. This paper presents a case study investigating LLM-based individual simulation using a volunteer-contributed archive of private messaging history spanning over ten years. Based on this dataset, we propose the ''Individual Turing Test'' to evaluate whether acquaintances of the volunteer can correctly identify which response in a multi-candidate pool most plausibly originates from the volunteer. We investigate prevalent approaches to LLM-based individual simulation, including fine-tuning, retrieval-augmented generation (RAG), memory-based methods, and hybrid approaches that integrate fine-tuning with RAG or memory. Empirical results show that current methods do not pass the Individual Turing Test, but perform substantially better when the same test is conducted on strangers to the target individual. Additionally, while fine-tuning improves performance in daily chats that reflect the individual's language style, retrieval-augmented and memory-based approaches demonstrate stronger performance on questions involving personal opinions and preferences. These findings reveal a fundamental trade-off between parametric and non-parametric approaches to individual simulation with LLMs under longitudinal context.

Large Language Models (LLMs) have achieved significant successes in various research domains by learning the relationship between words. However, while these models are capable of making predictions and inferences based on the learned patterns, they lack logical reasoning abilities, which are crucial for solving problems in both theoretical and practical domains. In addition, traditional logic inference methods are effective in solving problems that are based on logic, but not suitable for general tasks such as recommendations. In response to these challenges, this article introduces a Logical Large Language Model (L3M) that integrates the strengths of logical reasoning and LLMs. The data in L3M are represented in logical expressions, and the model uses logical constraints to learn the rules of basic logical operations such as And, Or, and Not. We conduct experiments on both theoretical tasks (solving logical equations) and practical tasks (recommender systems). The results of our theoretical experiments demonstrate that L3M is highly effective in solving logical expressions and variables. Additionally, L3M outperforms the state-of-the-art recommendation models in sequential recommendation tasks.

Multiple-choice question (MCQ) benchmarks such as MMLU and GPQA are widely used to assess the capabilities of large language models (LLMs). While accuracy remains the standard evaluation metric, recent work has introduced uncertainty quantification (UQ) methods, such as entropy, conformal prediction, and verbalized confidence, as complementary measures of model reliability and calibration. However, we find that these UQ methods, when applied to MCQ tasks, are unexpectedly fragile. Specifically, we show that fine-tuning a model on just 1,000 examples to adjust the probability of the first generated token, under the common prompting setup where the model is instructed to output only a single answer choice, can systematically distort a broad range of UQ methods across models, prompts, and domains, all while leaving answer accuracy unchanged. We validate this phenomenon through extensive experiments on five instruction-tuned LLMs, tested under standard prompting, zero-shot chain-of-thought reasoning, and a biomedical question answering setting. In all cases, models retain similar accuracy but exhibit significantly degraded calibration. These results suggest that current UQ practices for MCQs are ''one-token deep'', driven more by first-token decoding behavior than by any deeper representation of uncertainty, and are easily manipulated through minimal interventions. Our findings call for more robust and interpretable approaches to uncertainty estimation, particularly in structured formats like MCQs, where confidence signals are often reduced to token-level heuristics.

LLM-based Generative recommendation has attracted significant attention. However, in contrast to standard NLP tasks that inherently operate on human vocabulary, current generative recommendation approaches struggle to effectively encode items within the text-to-text framework. Due to this issue, the true potential of LLM-based generative recommendation remains largely unexplored. To better align LLMs with recommendation needs, we propose IDGenRec, representing each item as a unique, concise, semantically rich, platform-agnostic textual ID using human language tokens. This is achieved by training a textual ID generator alongside the LLM-based recommender, enabling seamless integration of personalized recommendations into natural language generation. Notably, as user history is expressed in natural language and decoupled from the original dataset, our approach suggests the potential for a foundational generative recommendation model.

In recent years, the integration of Large Language Models (LLMs) into recommender systems has garnered interest among both practitioners and researchers. Despite this interest, the field is still emerging, and the lack of open-source R&D platforms may impede the exploration of LLM-based recommendations. This paper introduces OpenP5, an open-source platform designed as a resource to facilitate the development, training, and evaluation of LLM-based generative recommender systems for research purposes. The platform is implemented using the encoder-decoder LLMs (e.g., T5) and the decoder-only LLMs (e.g., LLaMA-2) across 10 widely recognized public datasets, catering to two fundamental recommendation tasks: sequential and straightforward recommendations. Recognizing the crucial role of item IDs in LLM-based recommendations, we have also incorporated three item indexing methods within the OpenP5 platform: random indexing, sequential indexing and collaborative indexing. Built on the Transformers library, the platform facilitates easy customization of LLM-based recommendations for users. OpenP5 boasts a range of features including extensible data processing, task-centric optimization, comprehensive datasets and checkpoints, efficient acceleration, and standardized evaluations, making it a valuable tool for the implementation and evaluation of LLM-based recommender systems. The open-source code and pre-trained checkpoints for the OpenP5 library are publicly available at https://github.com/agiresearch/OpenP5.

Foundation Models such as Large Language Models (LLMs) have significantly advanced many research areas. In particular, LLMs offer significant advantages for recommender systems, making them valuable tools for personalized recommendations. For example, by formulating various recommendation tasks such as rating prediction, sequential recommendation, straightforward recommendation, and explanation generation into language instructions, LLMs make it possible to build universal recommendation engines that can handle different recommendation tasks. Additionally, LLMs have a remarkable capacity for understanding natural language, enabling them to comprehend user preferences, item descriptions, and contextual information to generate more accurate and relevant recommendations, leading to improved user satisfaction and engagement. This tutorial introduces Foundation Models such as LLMs for recommendation. We will introduce how recommender system advanced from shallow models to deep models and to large models, how LLMs enable generative recommendation in contrast to traditional discriminative recommendation, and how to build LLM-based recommender systems. We will cover multiple perspectives of LLM-based recommendation, including data preparation, model design, model pre-training, fine-tuning and prompting, multi-modality and multi-task learning, as well as trustworthy perspectives of LLM-based recommender systems such as fairness and transparency.