Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2610.07953

Benchmarking System One Models in Online Moderation

Federico Mazzoni·Andrea Failla
View original source ↗Read source document ↗Export metadata ↓

Abstract

Online moderation systems must apply changing platform policies, community rules, and prior decisions while producing decisions that can be audited and routed to human review. We evaluate whether System One Models, which accept natural-language context but return typed choices, probabilities, or scores, can support this setting. Across five moderation benchmarks, Jev is competitive with specialized reference systems, matching or exceeding them in several policy-grounded and harmful-content settings. We then use controlled information conditions to separate written rules, retrieved precedents, and restrictions on the candidate answer space. Jev generally benefits from retrieved precedents, improving exact policy selection and harmful-content discrimination when the answer space is held fixed. Laya is less consistent: retrieval often shifts its positive prediction rate or no-violation rate without improving discrimination. Jev confidence can support selective review in several settings, although it is not consistently calibrated; Laya confidence is less useful for ranking errors. These results show a promising future for SOM-powered content moderation.

Explore related subjects

cs.SIcs.CY

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Federico Mazzoni, Andrea Failla. 2026-10-06. Benchmarking System One Models in Online Moderation. https://arxiv.org/abs/2610.07953

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

A local eigenvector centrality

Eigenvector centrality is an established measure of global connectivity, from which the importance and influence of nodes can be inferred. We introduce local eigenvector centrality (LEC), a scale-adaptive extension that integrates local and global connectivity into a single centrality measure. LEC identifies a topologically relevant mesoscopic scale for community-level centrality assessment, informed by prominent eigengaps, and combines information from the associated eigenvectors, via the Euclidean norm, to yield a centrality that reflects community-driven node influence. Networks with known community structures are employed to demonstrate that LEC closely reproduces eigenvector centrality for each community in isolation. Discrepancies between these measures are shown to be attributable to the global connectivity of affected nodes. LEC therefore provides a principled integration of community-level and global notions of node importance. Comparisons with PageRank and Katz centrality demonstrate that damping-based approaches do not recover community-scale centrality as effectively as LEC.

cs.SI↗
arXiv2026

Recommender system in X inadvertently profiles ideological positions of users

Several data protection laws restrict processing that reveals political opinions, irrespective of the controller's intent. Whether recommender systems do so as a by-product of optimizing relevance has not been measured. From 2.5 million ``Who to Follow'' recommendations shown to 682 volunteers in France, we reconstructed an approximation of the embedding used by X's recommender for 26,509 accounts, computing survey-calibrated ideology scores. One direction in this embedding orders users by Left-Right position (Pearson rho = 0.887), distinct from directions tracking age, gender or popularity. We show this scale exists and affects the recommendations computed from the embedding. Removing it diversified recommendations at a limited cost in accuracy. We document a consequential form of emergent political representation that current definitions of profiling do not clearly address.

cs.SI↗
arXiv2026

Scalable detection of higher-order interactions in network data

Complex systems are routinely measured and represented through pairwise networks, even when the underlying interactions involve more than two units at once. Recovering this latent hypergraph structure from pairwise measurements is a fundamental inverse problem, but as the space of candidate hyperedges grows exponentially with system size, scalable hypergraph reconstruction at arbitrary interaction orders is out of reach for existing methods. Here we cast hypergraph reconstruction as a local signal-to-noise discrimination problem and use this locality to build a fast algorithm that reconstructs hypergraphs up to any interaction order. Across diverse synthetic and real-world systems our method achieves a high recovery accuracy of latent hypergraph structure while reconstructing hypergraphs up to orders of magnitude more quickly than current approaches. Our approach also yields an information-theoretic detectability boundary that sharply predicts which latent hyperedges are recoverable from pairwise measurements.

cs.SI↗