Binge Watch: Reproducible Multimodal Benchmarks Datasets for Large-Scale Movie Recommendation on MovieLens-10M and 20M

With the growing interest in Multimodal Recommender Systems (MRSs), collecting high-quality datasets provided with multimedia side information (text, images, audio, video) has become a fundamental step. However, most of the current literature in the field relies on small- or medium-scale datasets that are either not publicly released or built using undocumented processes. In this paper, we aim to fill this gap by releasing M3L-10M and M3L-20M, two large-scale, reproducible, multimodal datasets for the movie domain, obtained by enriching with multimodal features the popular MovieLens-10M and MovieLens-20M, respectively. By following a fully documented pipeline, we collect movie plots, posters, and trailers, from which textual, visual, acoustic, and video features are extracted using several state-of-the-art encoders. We publicly release mappings to download the original raw data, the extracted features, and the complete datasets in multiple formats, fostering reproducibility and advancing the field of MRSs. In addition, we conduct qualitative and quantitative analyses that showcase our datasets across several perspectives. This work represents a foundational step to ensure reproducibility and replicability in the large-scale, multimodal movie recommendation domain. Our resource can be fully accessed at the following link: https://zenodo.org/records/18499145, while the source code is accessible at https://github.com/giuspillo/M3L_10M_20M.

Extracting Consumer Insight from Text: A Large Language Model Approach to Emotion and Evaluation Measurement

Accurately measuring consumer emotions and evaluations from unstructured text remains a core challenge for marketing research and practice. This study introduces the Linguistic eXtractor (LX), a fine-tuned, large language model trained on consumer-authored text that also has been labeled with consumers' self-reported ratings of 16 consumption-related emotions and four evaluation constructs: trust, commitment, recommendation, and sentiment. LX consistently outperforms leading models, including GPT-4 Turbo, RoBERTa, and DeepSeek, achieving 81% macro-F1 accuracy on open-ended survey responses and greater than 95% accuracy on third-party-annotated Amazon and Yelp reviews. An application of LX to online retail data, using seemingly unrelated regression, affirms that review-expressed emotions predict product ratings, which in turn predict purchase behavior. Most emotional effects are mediated by product ratings, though some emotions, such as discontent and peacefulness, influence purchase directly, indicating that emotional tone provides meaningful signals beyond star ratings. To support its use, a no-code, cost-free, LX web application is available, enabling scalable analyses of consumer-authored text. In establishing a new methodological foundation for consumer perception measurement, this research demonstrates new methods for leveraging large language models to advance marketing research and practice, thereby achieving validated detection of marketing constructs from consumer data.

From User Preferences to Base Score Extraction Functions in Gradual Argumentation (with Appendix)

Gradual argumentation is a field of symbolic AI which is attracting attention for its ability to support transparent and contestable AI systems. It is considered a useful tool in domains such as decision-making, recommendation, debate analysis, and others. The outcomes in such domains are usually dependent on the arguments' base scores, which must be selected carefully. Often, this selection process requires user expertise and may not always be straightforward. On the other hand, organising the arguments by preference could simplify the task. In this work, we introduce \emph{Base Score Extraction Functions}, which provide a mapping from users' preferences over arguments to base scores. These functions can be applied to the arguments of a \emph{Bipolar Argumentation Framework} (BAF), supplemented with preferences, to obtain a \emph{Quantitative Bipolar Argumentation Framework} (QBAF), allowing the use of well-established computational tools in gradual argumentation. We outline the desirable properties of base score extraction functions, discuss some design choices, and provide an algorithm for base score extraction. Our method incorporates an approximation of non-linearities in human preferences to allow for better approximation of the real ones. Finally, we evaluate our approach both theoretically and experimentally in a robotics setting, and offer recommendations for selecting appropriate gradual semantics in practice.

FeDecider: An LLM-Based Framework for Federated Cross-Domain Recommendation

Federated cross-domain recommendation (Federated CDR) aims to collaboratively learn personalized recommendation models across heterogeneous domains while preserving data privacy. Recently, large language model (LLM)-based recommendation models have demonstrated impressive performance by leveraging LLMs' strong reasoning capabilities and broad knowledge. However, adopting LLM-based recommendation models in Federated CDR scenarios introduces new challenges. First, there exists a risk of overfitting with domain-specific local adapters. The magnitudes of locally optimized parameter updates often vary across domains, causing biased aggregation and overfitting toward domain-specific distributions. Second, unlike traditional recommendation models (e.g., collaborative filtering, bipartite graph-based methods) that learn explicit and comparable user/item representations, LLMs encode knowledge implicitly through autoregressive text generation training. This poses additional challenges for effectively measuring the cross-domain similarities under heterogeneity. To address these challenges, we propose an LLM-based framework for federated cross-domain recommendation, FeDecider. Specifically, FeDecider tackles the challenge of scale-specific noise by disentangling each client's low-rank updates and sharing only their directional components. To handle the need for flexible and effective integration, each client further learns personalized weights that achieve the data-aware integration of updates from other domains. Extensive experiments across diverse datasets validate the effectiveness of our proposed FeDecider.

Can Recommender Systems Teach Themselves? A Recursive Self-Improving Framework with Fidelity Control

The scarcity of high-quality training data presents a fundamental bottleneck to scaling machine learning models. This challenge is particularly acute in recommendation systems, where extreme sparsity in user interactions leads to rugged optimization landscapes and poor generalization. We propose the Recursive Self-Improving Recommendation (RSIR) framework, a paradigm in which a model bootstraps its own performance without reliance on external data or teacher models. RSIR operates in a closed loop: the current model generates plausible user interaction sequences, a fidelity-based quality control mechanism filters them for consistency with user's approximate preference manifold, and a successor model is augmented on the enriched dataset. Our theoretical analysis shows that RSIR acts as a data-driven implicit regularizer, smoothing the optimization landscape and guiding models toward more robust solutions. Empirically, RSIR yields consistent, cumulative gains across multiple benchmarks and architectures. Notably, even smaller models benefit, and weak models can generate effective training curricula for stronger ones. These results demonstrate that recursive self-improvement is a general, model-agnostic approach to overcoming data sparsity, suggesting a scalable path forward for recommender systems and beyond. Our anonymized code is available at https://anonymous.4open.science/r/RSIR-7C5B .

Semantics-Aware Denoising: A PLM-Guided Sample Reweighting Strategy for Robust Recommendation

Implicit feedback, such as user clicks, serves as the primary data source for modern recommender systems. However, click interactions inherently contain substantial noise, including accidental clicks, clickbait-induced interactions, and exploratory browsing behaviors that do not reflect genuine user preferences. Training recommendation models with such noisy positive samples leads to degraded prediction accuracy and unreliable recommendations. In this paper, we propose SAID (Semantics-Aware Implicit Denoising), a simple yet effective framework that leverages semantic consistency between user interests and item content to identify and downweight potentially noisy interactions. Our approach constructs textual user interest profiles from historical behaviors and computes semantic similarity with target item descriptions using pre-trained language model (PLM) based text encoders. The similarity scores are then transformed into sample weights that modulate the training loss, effectively reducing the impact of semantically inconsistent clicks. Unlike existing denoising methods that require complex auxiliary networks or multi-stage training procedures, SAID only modifies the loss function while keeping the backbone recommendation model unchanged. Extensive experiments on two real-world datasets demonstrate that SAID consistently improves recommendation performance, achieving up to 2.2% relative improvement in AUC over strong baselines, with particularly notable robustness under high noise conditions.

In Agents We Trust, but Who Do Agents Trust? Latent Source Preferences Steer LLM Generations

Agents based on Large Language Models (LLMs) are increasingly being deployed as interfaces to information on online platforms. These agents filter, prioritize, and synthesize information retrieved from the platforms' back-end databases or via web search. In these scenarios, LLM agents govern the information users receive, by drawing users' attention to particular instances of retrieved information at the expense of others. While much prior work has focused on biases in the information LLMs themselves generate, less attention has been paid to the factors that influence what information LLMs select and present to users. We hypothesize that when information is attributed to specific sources (e.g., particular publishers, journals, or platforms), current LLMs exhibit systematic latent source preferences- that is, they prioritize information from some sources over others. Through controlled experiments on twelve LLMs from six model providers, spanning both synthetic and real-world tasks, we find that several models consistently exhibit strong and predictable source preferences. These preferences are sensitive to contextual framing, can outweigh the influence of content itself, and persist despite explicit prompting to avoid them. They also help explain phenomena such as the observed left-leaning skew in news recommendations in prior work. Our findings advocate for deeper investigation into the origins of these preferences, as well as for mechanisms that provide users with transparency and control over the biases guiding LLM-powered agents.

FrameRef: A Framing Dataset and Simulation Testbed for Modeling Bounded Rational Information Health

Information ecosystems increasingly shape how people internalize exposure to adverse digital experiences, raising concerns about the long-term consequences for information health. In modern search and recommendation systems, ranking and personalization policies play a central role in shaping such exposure and its long-term effects on users. To study these effects in a controlled setting, we present FrameRef, a large-scale dataset of 1,073,740 systematically reframed claims across five framing dimensions: authoritative, consensus, emotional, prestige, and sensationalist, and propose a simulation-based framework for modeling sequential information exposure and reinforcement dynamics characteristic of ranking and recommendation systems. Within this framework, we construct framing-sensitive agent personas by fine-tuning language models with framing-conditioned loss attenuation, inducing targeted biases while preserving overall task competence. Using Monte Carlo trajectory sampling, we show that small, systematic shifts in acceptance and confidence can compound over time, producing substantial divergence in cumulative information health trajectories. Human evaluation further confirms that FrameRef's generated framings measurably affect human judgment. Together, our dataset and framework provide a foundation for systematic information health research through simulation, complementing and informing responsible human-centered research. We release FrameRef, code, documentation, human evaluation data, and persona adapter models at https://github.com/infosenselab/frameref.

Visual Persuasion: What Influences Decisions of Vision-Language Models?

The web is littered with images, once created for human consumption and now increasingly interpreted by agents using vision-language models (VLMs). These agents make visual decisions at scale, deciding what to click, recommend, or buy. Yet, we know little about the structure of their visual preferences. We introduce a framework for studying this by placing VLMs in controlled image-based choice tasks and systematically perturbing their inputs. Our key idea is to treat the agent's decision function as a latent visual utility that can be inferred through revealed preference: choices between systematically edited images. Starting from common images, such as product photos, we propose methods for visual prompt optimization, adapting text optimization methods to iteratively propose and apply visually plausible modifications using an image generation model (such as in composition, lighting, or background). We then evaluate which edits increase selection probability. Through large-scale experiments on frontier VLMs, we demonstrate that optimized edits significantly shift choice probabilities in head-to-head comparisons. We develop an automatic interpretability pipeline to explain these preferences, identifying consistent visual themes that drive selection. We argue that this approach offers a practical and efficient way to surface visual vulnerabilities, safety concerns that might otherwise be discovered implicitly in the wild, supporting more proactive auditing and governance of image-based AI agents.

Eco-Amazon: Enriching E-commerce Datasets with Product Carbon Footprint for Sustainable Recommendations

In the era of responsible and sustainable AI, information retrieval and recommender systems must expand their scope beyond traditional accuracy metrics to incorporate environmental sustainability. However, this research line is severely limited by the lack of item-level environmental impact data in standard benchmarks. This paper introduces Eco-Amazon, a novel resource designed to bridge this gap. Our resource consists of an enriched version of three widely used Amazon datasets (i.e., Home, Clothing, and Electronics) augmented with Product Carbon Footprint (PCF) metadata. CO2e emission scores were generated using a zero-shot framework that leverages Large Language Models (LLMs) to estimate item-level PCF based on product attributes. Our contribution is three-fold: (i) the release of the Eco-Amazon datasets, enriching item metadata with PCF signals; (ii) the LLM-based PCF estimation script, which allows researchers to enrich any product catalogue and reproduce our results; (iii) a use case demonstrating how PCF estimates can be exploited to promote more sustainable products. By providing these environmental signals, Eco-Amazon enables the community to develop, benchmark, and evaluate the next generation of sustainable retrieval and recommendation models. Our resource is available at https://doi.org/10.5281/zenodo.18549130, while our source code is available at: http://github.com/giuspillo/EcoAmazon/.