Generative AI assistants offer significant potential to enhance productivity, streamline information access, and improve user experience in enterprise contexts. In this work, we present Summit Concierge, a domain-specific AI assistant developed for Adobe Summit. The assistant handles a wide range of event-related queries and operates under real-world constraints such as data sparsity, quality assurance, and rapid deployment. To address these challenges, we adopt a human-in-the-loop development workflow that combines prompt engineering, retrieval grounding, and lightweight human validation. We describe the system architecture, development process, and real-world deployment outcomes. Our experience shows that agile, feedback-driven development enables scalable and reliable AI assistants, even in cold-start scenarios.

We study the problem of aligning a generative model's response with a user's preferences. Recent works have proposed several different formulations for personalized alignment; however, they either require a large amount of user preference queries or require that the preference be explicitly specified as a text input. In this paper, we propose a novel inference-time personalized alignment method, UserAlign, that elicits the user's preferences with a few queries as pairwise response comparisons. In particular, UserAlign builds on the theoretical framework of best-arm identification in logistic bandits and selects a personalized response from a fixed pool of the model's generated responses. The key idea is to consider the user's feedback consistent and noise-free, and incorporate it into the theoretical framework to identify the best response quickly. Experimental results across several tasks, involving personalized text and image generation, showcase the effectiveness of UserAlign in achieving personalized alignment.

While prior work on group recommender systems (GRSs) has primarily focused on improving recommendation accuracy, most approaches assume static or predefined groups, making them unsuitable for dynamic, real-world scenarios. We reframe group formation as a core challenge in GRSs and propose DeepForm (Stochastic Deep Graph Clustering for Practical Group Formation), a framework designed to meet three key operational requirements: (1) the incorporation of high-order user information, (2) real-time group formation, and (3) dynamic adjustment of the number of groups. DeepForm employs a lightweight GCN architecture that effectively captures high-order structural signals. Stochastic cluster learning enables adaptive group reconfiguration without retraining, while contrastive learning refines groups under dynamic conditions. Experiments on multiple datasets demonstrate that DeepForm achieves superior group formation quality, efficiency, and recommendation accuracy compared with various baselines.

New York's new mayor Zohran Mamdani tells Trump'I have four words for you' in blistering victory speech quoting his socialist hero, bragging about'toppling a dynasty' and promising a'new dawn' Driver screaming'Allahu Akbar' ploughs in to pedestrians'trying to hit everyone he encountered' on French holiday island leaving ten injured This Leftist election landslide was caused by the same vile disease that's triggered a GOP civil war. Why Mamdani's socialist revolution in New York has sparked a civil war for Democrats... and Trump is secretly loving it Simone Biles details all the plastic surgery she's had after her boob job this summer Kim Kardashian's new TV show All's Fair is SAVAGED by critics as it's branded'the worst drama ever', 'existentially terrible' and'a crime against television' while debuting at 0% on Rotten Tomatoes Putin orders increased drone'incursions' in Europe as Russia ramps up'hybrid war' - with Belgium's biggest airport forced to close overnight Inside Kate and William's forever home: Princess is kitting out Forest Lodge in her preferred'classic contemporary style' to create a'lovely but absolutely inoffensive' look REVEALED: Fattest states in America ranked... including region where three-quarters of residents are obese I was so desperate for a baby I stole sperm from my husband's condom: It's the most shocking confession. Now for the first time LIZ JONES tells what happened next... and the consequence no one saw New footage reveals the moments before football manager collapsed and died mid-match, leaving his players in disbelief, as it emerges he'complained about fish he had eaten' hours before Hollywood A-listers may be blacklisted for'antisemitism' under Paramount's new anti-woke leadership Texas teen'tears masterpiece from wall at the Met in unhinged meltdown' before being handed in by his MOTHER Is this how the world will end? Scientists have revealed a grim prospect for humanity's future, as they warn Earth will eventually be consumed by the sun. In roughly five billion years, our star will burn the last of its hydrogen fuel and begin expanding into a monstrous red giant.

New York's new mayor Zohran Mamdani tells Trump'I have four words for you' in blistering victory speech quoting his socialist hero, bragging about'toppling a dynasty' and promising a'new dawn' Driver screaming'Allahu Akbar' ploughs in to pedestrians'trying to hit everyone he encountered' on French holiday island leaving ten injured This Leftist election landslide was caused by the same vile disease that's triggered a GOP civil war. Why Mamdani's socialist revolution in New York has sparked a civil war for Democrats... and Trump is secretly loving it Simone Biles details all the plastic surgery she's had after her boob job this summer Kim Kardashian's new TV show All's Fair is SAVAGED by critics as it's branded'the worst drama ever', 'existentially terrible' and'a crime against television' while debuting at 0% on Rotten Tomatoes Putin orders increased drone'incursions' in Europe as Russia ramps up'hybrid war' - with Belgium's biggest airport forced to close overnight Inside Kate and William's forever home: Princess is kitting out Forest Lodge in her preferred'classic contemporary style' to create a'lovely but absolutely inoffensive' look REVEALED: Fattest states in America ranked... including region where three-quarters of residents are obese I was so desperate for a baby I stole sperm from my husband's condom: It's the most shocking confession. Now for the first time LIZ JONES tells what happened next... and the consequence no one saw New footage reveals the moments before football manager collapsed and died mid-match, leaving his players in disbelief, as it emerges he'complained about fish he had eaten' hours before Hollywood A-listers may be blacklisted for'antisemitism' under Paramount's new anti-woke leadership Texas teen'tears masterpiece from wall at the Met in unhinged meltdown' before being handed in by his MOTHER Bizarre TikTok trend sees Gen Z'rawdogging boredom' to fix their attention spans - so, does it really work? READ MORE: Mark Zuckerberg's'rawdog' routine that made him a billionaire A bizarre new trend has emerged on TikTok, in which Gen Z put themselves in timeout to try to fix their attention spans. Dubbed'rawdogging boredom', users set a timer and simply sit there without any distractions.

While existing work focuses primarily on task success, we argue that effective real-world agents require optimizing three dimensions: productivity (task completion), proactivity (asking essential questions), and personalization (adapting to diverse user preferences). We introduce UserVille, an interactive environment with LLM-based user simulators enabling diverse, configurable user preferences. Leveraging UserVille, we introduce PPP, a multi-objective reinforcement learning approach that jointly optimizes all three dimensions: Productivity, Proactivity, and Personalization. Experiments on software engineering and deep research tasks show that agents trained with PPP achieve substantial improvements over strong baselines such as GPT-5 (+21.6 on average), demonstrating the ability to ask strategic clarifying questions, adapt to unseen user preferences, and improve task success through better interaction. This work demonstrates that explicitly optimizing for user-centered interaction is critical for building practical and effective AI agents.

We present a scalable recommender system implementation based on RippleNet, tailored for the media domain with a production deployment in Onet.pl, one of Poland's largest online media platforms. Our solution addresses the cold-start problem for newly published content by integrating content-based item embeddings into the knowledge propagation mechanism of RippleNet, enabling effective scoring of previously unseen items. The system architecture leverages Amazon SageMaker for distributed training and inference, and Apache Airflow for orchestrating data pipelines and model retraining workflows. To ensure high-quality training data, we constructed a comprehensive golden dataset consisting of user and item features and a separate interaction table, all enabling flexible extensions and integration of new signals.

Lexical alignment, where speakers start to use similar words across conversation, is known to contribute to successful communication. However, its implementation in conversational agents remains underexplored, particularly considering the recent advancements in large language models (LLMs). As a first step towards enabling lexical alignment in human-agent dialogue, this study draws on strategies for personalising conversational agents and investigates the construction of stable, personalised lexical profiles as a basis for lexical alignment. Specifically, we varied the amounts of transcribed spoken data used for construction as well as the number of items included in the profiles per part-of-speech (POS) category and evaluated profile performance across time using recall, coverage, and cosine similarity metrics. It was shown that smaller and more compact profiles, created after 10 min of transcribed speech containing 5 items for adjectives, 5 items for conjunctions, and 10 items for adverbs, nouns, pronouns, and verbs each, offered the best balance in both performance and data efficiency. In conclusion, this study offers practical insights into constructing stable, personalised lexical profiles, taking into account minimal data requirements, serving as a foundational step toward lexical alignment strategies in conversational agents.

The rapid development of Vision-language models (VLMs) enables open-ended perception and reasoning. Recent works have started to investigate how to adapt general-purpose VLMs into personalized assistants. Even commercial models such as ChatGPT now support model personalization by incorporating user-specific information. However, existing methods either learn a set of concept tokens or train a VLM to utilize user-specific information. However, both pipelines struggle to generate accurate answers as personalized assistants. We introduce Jarvis, an innovative framework for a personalized AI assistant through personal KV-Cache retrieval, which stores user-specific information in the KV-Caches of both textual and visual tokens. The textual tokens are created by summarizing user information into metadata, while the visual tokens are produced by extracting distinct image patches from the user's images. When answering a question, Jarvis first retrieves related KV-Caches from personal storage and uses them to ensure accuracy in responses. We also introduce a fine-grained benchmark built with the same distinct image patch mining pipeline, emphasizing accurate question answering based on fine-grained user-specific information. Jarvis is capable of providing more accurate responses, particularly when they depend on specific local details. Jarvis achieves state-of-the-art results in both visual question answering and text-only tasks across multiple datasets, indicating a practical path toward personalized AI assistants. The code and dataset will be released.

Large language models (LLMs) have demonstrated significant potential in solving recommendation tasks. With proven capabilities in understanding user preferences, LLM personalization has emerged as a critical area for providing tailored responses to individuals. Current studies explore personalization through prompt design and fine-tuning, paving the way for further research in personalized LLMs. However, existing approaches are either costly and inefficient in capturing diverse user preferences or fail to account for timely updates to user history. To address these gaps, we propose the Memory-Assisted Personalized LLM (MAP). Through user interactions, we first create a history profile for each user, capturing their preferences, such as ratings for historical items. During recommendation, we extract relevant memory based on similarity, which is then incorporated into the prompts to enhance personalized recommendations. In our experiments, we define a new task that enables testing with varying memory size under two scenarios: single domain where memory and tasks are from the same category and cross-domain (e.g. memory from movies and recommendation tasks in books). The results show that MAP outperforms regular LLM-based recommenders that integrate user history directly through prompt design. Moreover, as user history grows, MAP's advantage increases in both scenarios, making it more suitable for addressing successive personalized user requests.