CloSeR is a plug‑and‑play framework that enhances Generalized Category Discovery (GCD) by injecting closed‑set relational knowledge from a lightweight teacher model. The teacher is built by fine‑tuning adapters on labeled data while keeping the backbone frozen, preserving pretrained priors. Unified Relational Distillation then transfers both global sample‑to‑prototype and local sample‑to‑sample relations to the GCD task, reducing optimization interference and improving performance across six benchmarks with DINO and DINOv2 backbones.
Generalized Category Discovery (GCD) is an intriguing open-world problem that has garnered increasing attention: given partially labelled data, the goal is to correctly recognize known classes while d...
arXiv:2607. 00620v1 Announce Type: cross Abstract: Generalized Category Discovery (GCD) aims to recognize known classes while autonomously discovering novel ones in open-world settings.
arXiv:2607. 15919v1 Announce Type: cross Abstract: Data-free knowledge distillation transfers the knowledge encoded in a teacher model to a student model without access to the original training data.
arXiv:2602. 17395v2 Announce Type: replace-cross Abstract: Generalized Category Discovery (GCD) aims to identify novel categories in unlabeled data while leveraging a small labeled subset of known classes.
arXiv:2605. 09420v2 Announce Type: replace-cross Abstract: In this study, we tackle Generalized Category Discovery (GCD) via a Relational Retrieval perspective, explicitly coupling labeled and unlabeled data through bidirectional knowledge transfer.
The paper introduces Semantic Localization-Enhanced Teacher (SLE‑T), a knowledge‑distillation framework that aligns spatial‑scale and semantic features between a Vision Foundation Model (VFM) teacher and a student detector for cross‑domain object detection. SLE‑T employs a lightweight SLE Adapter that injects pretrained local‑texture priors into DINOv2 and reformulates its features into dense, spatially and semantically compatible representations, enabling effective pseudo‑label learning or feature alignment. Experiments on three domain‑adaptive object detection benchmarks show that SLE‑T with DINOv2‑B achieves state‑of‑the‑art performance while using only a quarter of the training time and less GPU memory compared to the larger DINOv2‑G teacher.
arXiv:2607. 04548v1 Announce Type: cross Abstract: Novel category discovery aims to identify unseen classes from unlabeled data by transferring knowledge from labeled categories, but most existing methods perform discovery in opaque latent feature spaces.
arXiv:2606. 04922v1 Announce Type: cross Abstract: Current prompt-based and adapter-based tuning of vision-language models (VLMs) is attractive for medical imaging, where clinical data sensitivity favors frozen backbones and annotations are limited.
The paper introduces a method that transforms knowledge graph facts into a fixed vocabulary representation, where each fact becomes a node linked to its subject, object, and relation type via six meta-relations. Using this representation, standard GNNs (e.g., GAT, GINE, GraphSAGE, R-GCN) trained on a single small graph can achieve zero‑shot link prediction on 40 inductive benchmarks, matching the performance of specialized foundation models like ULTRA. The approach also generalizes to relational databases, enabling foreign‑key prediction without cell values or schema text, and the authors provide code, checkpoints, and evaluation tools for all benchmarks.
arXiv:2608.22584v1 Announce Type: new Abstract: Two-stage neuro-symbolic architectures provide an elegant paradigm for visual problem solving by cleanly separating connectionist perception of predefi...
arXiv:2606. 18209v1 Announce Type: new Abstract: Dataset distillation (DD) has emerged as a prominent approach in data centric machine learning, aiming to synthesize compact training sets for efficient training by compressing the information in large datasets into a small number of synthetic samples.
The paper introduces G2D, a training‑free framework that combines a discriminative model (CLIP) for broad candidate retrieval with a generative vision‑language model for fine‑grained, image‑grounded verification. By using CLIP’s top‑K shortlist and a structured prior from candidate names and probabilities, G2D focuses generative reasoning on uncertain samples, achieving an average accuracy of 68.85% across eight benchmarks—higher than both CLIP alone (59.35%) and the standalone generative model (63.11%). The approach also adapts to various generator configurations and extends to other models such as DCLIP, WaffleCLIP, and CuPL.