Search: “concepts”

arXiv:2602.05126v2 Announce Type: replace Abstract: Human papillomavirus (HPV) status is a critical determinant of prognosis and treatment response in head and neck and cervical cancers. Although atte

arXiv:2605.08482v1 Announce Type: cross Abstract: Automated ICD-10 coding from clinical discharge summaries requires models that are both accurate on long-tailed multi-label classification tasks and i

arXiv:2603.15525v2 Announce Type: replace Abstract: Deep learning models for chest X-ray diagnosis are constrained by limited coverage of clinically meaningful concept combinations in publicly availab

arXiv:2604.07965v1 Announce Type: new Abstract: Model editing aims to update knowledge to add new concepts and change relevant information without retraining. Lifelong editing is a challenging task, p

arXiv:2609.39605v1 Announce Type: new Abstract: The rapid advancement of generative video models has enabled the synthesis of increasingly realistic and temporally coherent videos, while also raising

arXiv:2606.14586v2 Announce Type: replace Abstract: Self-supervised learning can learn useful representations from unlabeled images. Yet these representations remain largely opaque, making it difficul

arXiv:2609.20267v1 Announce Type: cross Abstract: Concept erasure aims to selectively eliminate undesired visual semantics from pre-trained generative models without compromising their general utility

arXiv:2609.20303v1 Announce Type: new Abstract: Classical philosophical corpora pose three compounding challenges for language resources: they exist in several languages without parallel alignment, th

arXiv:2609.11253v1 Announce Type: cross Abstract: Erasing concept-specific information from representations has been proven useful for mitigating bias or interpreting model decisions. The joint object

arXiv:2609.10715v1 Announce Type: new Abstract: We introduce NCP-ArchPreview, a latent-space language model that pushes autoregressive pretraining beyond standard next-token prediction (NTP). Alongsid

arXiv:2609.01456v1 Announce Type: cross Abstract: Composed image retrieval (CIR) retrieves a target image from a reference image and a text modification. This paper studies metadata-available CIR rera

arXiv:2609.00476v1 Announce Type: new Abstract: Graphic designers often blend visual concepts to communicate multiple ideas within a single image, leveraging positive and negative space to create bala

arXiv:2608.23893v1 Announce Type: new Abstract: Learning systems deployed over long periods must adapt not only to statistical changes in incoming data, but also to revisions of the definitions that g

arXiv:2608.22956v1 Announce Type: new Abstract: Concept-bottleneck controllable generation routes multi-attribute control through a low-dimensional concept code that, at deployment, must be synthesise

arXiv:2608.15404v1 Announce Type: new Abstract: Concept Bottleneck Models (CBMs) are designed to make visual classification interpretable by expressing predictions through human-understandable concept

arXiv:2608.13478v1 Announce Type: new Abstract: We present our submission to Task 3 of the Genmu 2.0 Challenge on visual concept unlearning. Building on MapRoute, we introduce task-specific training o

arXiv:2608.10566v1 Announce Type: cross Abstract: How many directions does a neural representation use to encode a concept? A common answer repeatedly erases probe directions and reports the stopping

arXiv:2608.05162v1 Announce Type: new Abstract: Pooling is a consequential but under-examined design choice in decoder-only concept representation work: practitioners must collapse token-level hidden

arXiv:2608.03225v1 Announce Type: new Abstract: Human-interpretable computer-aided diagnosis is crucial for clinical decision making. Concept-based models excel by providing transparent reasoning and

arXiv:2512.12816v2 Announce Type: replace Abstract: We study how to allocate resources for training and deployment of machine learning (ML) models under concept drift and limited budgets. We consider

arXiv:2607.22544v1 Announce Type: new Abstract: Visual counterfactual explanations aim to answer 'what minimal change to this image would flip the model's prediction?', and are increasingly important

arXiv:2505.18672v2 Announce Type: replace Abstract: Representation intervention aims to localize and modify the representations that encode the underlying concepts in large language models (LLMs) to e

arXiv:2512.20666v2 Announce Type: replace-cross Abstract: Text-to-image diffusion models have attracted significant attention for their ability to generate diverse, high-fidelity images. However, in m

arXiv:2606.30313v1 Announce Type: new Abstract: Longitudinal glioblastoma response assessment requires comparing subtle tumor changes across MRI time points using structured clinical criteria such as

arXiv:2606.25434v1 Announce Type: new Abstract: Early and scalable detection of mild cognitive impairment (MCI) remains an unresolved clinical challenge. Existing EEG-based screening approaches are co

arXiv:2510.04180v2 Announce Type: replace Abstract: Deep neural networks can achieve high accuracy while relying on evidence that is hard to inspect or misaligned with the intended task. Concept Bottl

arXiv:2606.10669v1 Announce Type: cross Abstract: Concept-based models (CMs), deep neural networks that ground their predictions on representations aligned with human-understandable concepts (e.g., 'r

arXiv:2606.09653v1 Announce Type: new Abstract: Learned representations across models and modalities often exhibit striking structural similarities, suggesting shared underlying concept decompositions

arXiv:2606.00351v1 Announce Type: new Abstract: Personalized visual understanding has advanced significantly, yet existing approaches struggle to localize and extract specific concepts when input imag

arXiv:2602.02886v2 Announce Type: replace-cross Abstract: Concept Bottleneck Models (CBMs) promote interpretability by grounding predictions in human-understandable concepts. However, existing CBMs ty

arXiv:2605.24399v1 Announce Type: new Abstract: Healthcare models are transitioning from unimodal prediction toward multimodal reasoning over heterogeneous diagnostic inputs. In computational patholog

arXiv:2605.18150v1 Announce Type: new Abstract: Diffusion models (DMs) are widely used for text-to-image generation, but their strong generative capabilities also raise concerns about unsafe or undesi

arXiv:2603.25074v2 Announce Type: replace Abstract: Concept erasure serves as a vital safety mechanism for removing unwanted concepts from text-to-image (T2I) models. While extensively studied in U-Ne

arXiv:2604.21555v1 Announce Type: new Abstract: Sentence embedding techniques aim to encode key concepts of a sentence's meaning in a vector space. However, the majority of evaluation approaches for s

arXiv:2604.16483v1 Announce Type: new Abstract: Concept erasure in Text-To-Image (T2I) diffusion models is vital for safe content generation, but existing inference-time methods face significant limit

arXiv:2604.15829v1 Announce Type: new Abstract: Text-to-image generative models have achieved impressive fidelity and diversity, but can inadvertently produce unsafe or undesirable content due to impl

arXiv:2604.13331v1 Announce Type: new Abstract: In electronic health record (EHR) mining, learning high-quality representations of medical concepts (e.g., standardized diagnosis, medication, and proce

arXiv:2604.10032v1 Announce Type: cross Abstract: While modern generative models such as diffusion-based architectures have enabled impressive creative capabilities, they also raise important safety a

arXiv:2602.00377v3 Announce Type: replace Abstract: Existing knowledge probing methods rely on pre-defined queries, limiting extraction to known concepts. We introduce DecompressionLM, a stateless fra

arXiv:2609.04415v1 Announce Type: cross Abstract: Learning rich medical concept representations is essential for EHR prediction. Text-attributed knowledge graphs (TKGs) provide a natural foundation by

arXiv:2609.00469v1 Announce Type: new Abstract: Promptable segmentation foundation models such as SAM3 accept an open-vocabulary text concept and return every instance matching it, but adapting them t

arXiv:2608.27510v1 Announce Type: new Abstract: Transcoder attribution graphs are usually trained to explain why a model assigns high probability to a particular next token. We introduce Concept-Targe

arXiv:2608.24142v1 Announce Type: new Abstract: Prompt learning adapts vision-language models such as CLIP by optimizing continuous prompt vectors, but the learned prompts are difficult to interpret i

arXiv:2601.13798v3 Announce Type: replace-cross Abstract: Language-aligned vision foundation models perform strongly across diverse downstream tasks. Yet, their learned representations remain opaque,

arXiv:2608.15731v1 Announce Type: cross Abstract: Deep Neural Networks (DNNs) deployed in high-risk domains, such as healthcare and autonomous driving, must be not only accurate but also understandabl

arXiv:2608.07620v1 Announce Type: new Abstract: Recent advances in flow matching models have substantially improved the quality of text-to-image generation, but have also raised increasing safety conc