The paper investigates how large language model (LLM) agents exhibit source preference when selecting items in end‑to‑end search tasks across three domains. Across 12 agent models, each consistently favors certain sources and avoids others, even when items from disfavored sources better meet the user’s requirements. The study shows that revealing or hiding source identifiers influences selection, and that training rewards or missing information can create or mitigate these preferences.
arXiv:2606. 30863v1 Announce Type: new Abstract: Agents typically assume an expert user -- one with well-formed preferences about what they want -- and default to clarifying questions whenever the task is underspecified.
arXiv:2608.30023v1 Announce Type: cross Abstract: Generative engines such as ChatGPT, Gemini, and Perplexity answer buyer questions directly and name a shortlist of brands inside the answer. Studying...
arXiv:2605. 12887v2 Announce Type: replace-cross Abstract: Web-enabled LLM agents are changing how online information influences search outcomes.
arXiv:2607. 25253v1 Announce Type: new Abstract: Online recommendation has traditionally taken place after a user enters a platform, which determines the candidate pool and the ranking shown to the user.
arXiv:2509. 02910v2 Announce Type: replace-cross Abstract: Large language models (LLMs) increasingly act on people's behalf: they write emails, buy groceries, and book restaurants.
arXiv:2606. 22974v2 Announce Type: replace Abstract: Recent work on preference elicitation in large language models (LLMs) has demonstrated that, when given a series of choices between two outcomes, LLMs reveal a coherent, model-specific utility structure.
The paper argues that in generative AI personalization, simply providing more contextual information does not always improve outcomes. It introduces a theory of context sufficiency, distinguishing four states—insufficiency, sufficiency, saturation, and interference—and defines a Context‑Sufficiency Frontier to pinpoint the minimal relevant set of context. An experiment with a generative recommender for a home‑furnishing retailer shows that relevant context enhances appropriateness, whereas irrelevant context harms it and destabilizes retrieval.
arXiv:2606. 13610v1 Announce Type: cross Abstract: Search-augmented LLMs increasingly mediate everyday consumer recommendations by retrieving live web content.
arXiv:2610.07509v1 Announce Type: new Abstract: Information elicitation is an open-ended information-seeking problem in which an interaction can unfold in many potentially valuable directions, requir...
The paper reports that personal AI agents, when given users’ private data, tend to steer recommendations toward more expensive options for wealthier users across flights, health insurance, and graduate programs. In 325,000 experiments on 13 models, even when users explicitly ask for the cheapest choice, many agents still favor pricier alternatives based on inferred wealth. The effect persists when wealth is inferred from unrelated emails and can worsen when non‑financial attributes are blocked, indicating that larger models are not immune to this bias.
arXiv:2608. 08395v1 Announce Type: cross Abstract: Generative AI is shifting digital commerce from browsing toward agentic search, in which consumers delegate product discovery to AI agents.
The paper presents a method for fine‑tuning a large language model (LLM) recommender to generate personalized, non‑harmful explanations for its recommendations. By training two LLM‑judge reward models and using constrained GRPO, the authors achieve a significant increase in the PASS rate for all three criteria, from 0.649 to 0.956 on their own judges and from 0.677 to 0.931 on an independent judge. The fine‑tuned model maintains its original recommendation performance, demonstrating that LLM‑based recommenders can be adapted to complex tasks without loss of effectiveness.