Aug 2026· Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2· pp. 13455-13456· 0 citations· 10 references
Abstract
Large language models (LLMs) and agentic AI systems are rapidly moving into user-facing applications, yet most remain fundamentally generic, optimized for population-level objectives under the assumption that one model can serve all users. This assumption is increasingly misaligned with real-world deployment, where AI systems interact continuously with individuals whose preferences, knowledge, goals, and values evolve over time. PILA'26 is motivated by the need to move beyond static general models toward personal intelligence ---AI systems that explicitly model users and dynamically adapt their reasoning, behavior, and decisions through memory, interaction, and lifelong learning. The workshop brings together researchers and practitioners from data mining, LLMs, NLP, IR, human-centered AI, and AI safety to position personalization as a central research direction for next-generation AI systems at KDD. Topics include user memory and personalized alignment, self-evolving and lifelong learning, datasets and evaluation, real-world applications, and trustworthiness in user-adaptive AI. Workshop website: https://pila26-workshop.github.io.
Personal intelligence is becoming a central frontier for user-facing AI agents. To be helpful in everyday life, agents must understand users across the digital contexts where their preferences, intents, habits, social relationships, and needs unfold over time. Today's systems can personalize within individual apps or tasks, but personal intelligence as a whole remains under-measured: how agents build cross-context user understanding, support steerable recommendation systems, act proactively across platforms, and avoid over-personalization. We introduce PersonaMem-v3, a real-world-grounded benchmark and evaluation harness for omni-platform personal intelligence. PersonaMem-v3 is seeded from more than one million anonymized real-world engagement histories, most of which are implicit signals, and uses them to construct time-indexed user digital worlds across social media, chatbot, calendar, and AI-companion with preference evolvement over time. The benchmark brings personalization, LLM-powered recommendation, proactiveness, agentic tool use, and geo-temporal reasoning into one framework, anchored in psychology, social-linguistics, and user-behavior theories. It evaluates whether AI agents can infer holistic user understanding from cross-platform evidence, personalize responses, rerank recommendations on social media, follow user steering through natural language, and hold back when personalization would be inappropriate, repetitive, outdated, or unnecessary. PersonaMem-v3 points toward LLM-powered personal intelligent agents that work with existing scalable recommendation infrastructure while making personalization more interactive, agentic, and aligned with how real users experience their digital lives.
Bowen Jiang, Yuan Yuan, Zhuoqun Hao et al.· 0 citations
AI agents driven by large language models (LLMs) are radically changing industries through methods like automation, decision support, and intelligent interactions. As a result, the efficiency of these systems is as crucial as their capabilities. In fact, one of the most critical factors influencing AI's cost-effectiveness, speed, scalability, and user experience is token optimization, a factor often ignored in performance considerations. Some of the characteristics of the very modern AI workflows that can lead to a token explosion include deep prompting, several agents' interactions, retrieval of memories, and persistent context sharing. Such overuse of tokens has a double effect of continually increasing the expenditure and leading to unpleasant situations like lag, context overflow, deterioration of expected response, and wasting of resources. Alongside the contribution of AI agents to real-time applications, their critical nature is reminding us of the necessity to find ways of managing tokens intelligently to ensure a balance between performance and efficiency. This article presents a series of feasible and potent methods for optimizing token consumption in AI-driven systems at a minimum level without sacrificing the quality of outputs or the extent of contextual understanding. The methods proposed are prompt engineering antediluvian, context compression, memory selection, response generation, retrieval and adaptive token allocation that are area-specific and task-oriented and adapted to various workflows. Besides that, we analyze how intelligent token control can facilitate multi-agent collaboration operations while preventing unnecessary data exchange and reductions in processing redundancies. We maintain that enhancing token control has implications for a cleaner environment, greater scalability, and a more dependable and responsive infrastructure. The purpose of this paper is to present a practical, human-centric approach to the creation of 'smarter' AI agents, which are not only robust and precise but also resource-efficient and financially sustainable for large-scale deployment in the future.
Madhurima Kommuru· International Journal of Mac...· 0 citations
ODYSSE is presented, a Reinforced Fine-Tuning (RFT) framework for personalized agentic reasoning designed to address long action horizons and strong cross-step dependencies in personalized agentic reasoning, and an episodic batch sampler that groups actions from the same episode into unified training batches, facilitating coherent optimization under ESPO.
AI agents are trained on population-scale data to encode broad capabilities spanning those of many practitioners. Yet the artifacts they produce rarely meet the personal bar professionals need to stake their reputation on. On realistic, open-ended tasks where success criteria are heterogeneous and insufficiently documented, individual expertise lives precisely in the elevation and departure from the average. In practice, iterative human-agent interaction surfaces criteria that users cannot fully specify up front, yet apply repeatedly across tasks. We argue this cross-session interaction data is a rich, underused signal for closing the gap to individual expertise. In this work, we propose test-time adaptation through human-agent interaction (TAHI), which integrates these signals into agent context and weights, and crystallizes each user's training and evaluation criteria via an evolving rubric module. We adapt agents to 30 individuals in two high-utility domains, writing and visual creation, on a total of 600 tasks. Our agents improve solo task success by 4.5-20.9% within only tens of tasks. Meanwhile, our evolving rubric module serves as a scalable annotation tool, creating evaluation rubrics that catch 16.0-22.3% more failures than those from LMs or humans alone. While agents are adapted towards individuals, we show these personalized agents also produce improvements in success of up to 8.8% that generalize across users.
Z. Wang, Apurva Gandhi, Rulin Shao et al.· 0 citations
Large Language Models (LLMs) are increasingly used in simulations, either in academic settings for research studies or in industry for prototyping. Previous research has investigated the extent to which agents can mimic human behavior in socioeconomic settings; however, there is limited research on greedy decision-making by agents in simulated resource allocation environments. Furthermore, there is limited work on cross-model evaluation. Our research investigates the decision-making of ten agents across two different experimental conditions: one in which agents are able to communicate with other agents, and one in which emotional contexts are directly injected into the prompt. Based on a conceptual framework of greed, we find that agents predominantly exhibited greed-like behavior across all conditions. Interaction and self-reported social connection did not meaningfully influence the agents’ decision-making. We evaluated four different models: gpt-5-mini, gemini-3-flash-preview, claude-haiku-4-5-20251001, and grok-3-mini-fast-beta, and observed that while models differed in their self-reported connection scores, they did not differ significantly in greediness scores. The code and experimental artifacts are available at https://github.com/tiaL-ops/simCo.
Landy Rakotoarison, Fanamby T. Randriamahenintsoa· International Conference on...· 0 citations
A personal AI system needs a model of the user's goals, constraints, and ongoing commitments to plan and act on their behalf, and the quality of that model is bounded by what the system can observe. Broader observation does not by itself improve assistance because a bounded system must select and compress information for the task at hand. We argue that this observation bottleneck has a cooperative structure: the system builds a partial model of the user's changing life, the user evaluates its actions, and the user's consent and control shape what it can observe next. Useful and inspectable behavior can give users a reason to maintain or expand the observation channel, while failures can lead them to correct, narrow, revoke, or abandon it. We use the term cooperative observation for this feedback loop among usefulness, trust, and future access, and propose it as a framework for personal intelligence. We report a preliminary single-subject account from Organizm, a prototype used over six months, and outline evaluation directions for measuring how observation quality shapes personal AI.
Yashar Talebirad, Osman Jime, Ali Parsaee et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
