Newest papers

ReCoG: Relational Concept Graph for Training-Free Personalization

Changseon Yu, Hyeokjun Kweon · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.31282 · cited by 0 works (OpenAlex) · open-access PDF

French

Bien que les méthodes récentes basées sur le retriever enrichissent chaque concept avec des preuves descriptives, elles traitent largement les concepts en isolement, de sorte que les informations récupérées peuvent être plausibles mais non discriminatoires, en particulier lorsque le concept personnalisé contient des catégories très confuses et finement graines. Nous affirmons que la personnalisation est intrinsèquement relationnelle: la reconnaissance correcte dépend des distinctions spécifiques entre un concept et ses alternatives les plus proches. À cette fin, nous proposons ReCoG, une preuve conceptuelle relative pour la personnalisation augmentée par la récupération. ReCoG représente des concepts comme des nœuds et des augmentations avec des bords directs qui capturent explicitement des concepts confidentiels entre des paires,

Machine translation (Meta M2M100); the English original follows.

Abstract

Training-free personalization adapts a frozen vision-language model (VLM) to recognize user-defined concepts from only a few exemplars by retrieving concept entries at inference time. While recent retriever-based methods enrich each concept with descriptive evidence, they largely treat concepts in isolation, so the retrieved information can be plausible yet non-discriminative, especially when the personalized concept set contains highly confusable, fine-grained categories. We argue that personalization is inherently relational: correct recognition depends on the specific distinctions between a concept and its nearest alternatives. To this end, we propose ReCoG, a Relational Concept Graph for retrieval-augmented personalization. ReCoG represents concepts as nodes and augments them with directed edges that explicitly capture discriminative cues between concept pairs, elicited by a frozen VLM during database construction. At inference, ReCoG retrieves a shortlist and resolves ambiguity via explicit pairwise comparisons conditioned on the corresponding relational edges, selecting the concept that consistently wins against competing candidates. We also introduce FINGER benchmark to test personalized recognition on confusable concept sets. Across FINGER, ReCoG substantially outperforms prior methods, with the largest gains in the most confusable regimes. ReCoG also achieves consistently best performance on established personalization benchmarks including captioning and VQA, confirming that relational evidence is broadly beneficial beyond fine-grained settings. Project Page: https://csy-98.github.io/ReCoG/

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Personalization, Ambiguity, Pairwise comparison, Discriminative model, Artificial intelligence

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange