M-SYNTH is organized into a directory structure that indicates the parameters. Code and dataset is released with the Creative Commons 1.0 Universal License We now review the timing required to perform mass insertion and imaging. In Table 2, we review the imaging time required for each breast density. The time varies from 2.84 GPU), we were able to generate the complete dataset in about two weeks.Breast Density Time (min) Fatty 13.463809 Scattered 11.002291 Hetero 3.655613 Dense 2.842028 Table 2: Timing analysis for imaging by breast density. Additional renderings of the breast phantoms generated for the study are shown in Figure 1, demonstrating a high level of detail and anatomical variability within and among models.

IC3K 2025 (17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management) received 163 paper submissions from 40 countries. To evaluate each submission, a double-blind paper review was performed by the Program Committee. After a stringent selection process, 31 papers were published and presented as full papers, i.e. completed work (12 pages/25' oral presentation), 81 papers were accepted as short papers (54 as oral presentation). The organizing committee included the IC3K Conference Chairs: Ricardo da Silva Torres, Artificial Intelligence Group, Wageningen University & Research, Netherlands and Jorge Bernardino, Polytechnic University of Coimbra, Portugal, and the IC3K 2025 Program Chairs: Le Gruenwald, University of Oklahoma, School of Computer Science, United States, Frans Coenen, University of Liverpool, United Kingdom, Jesualdo Tomás Fernández-Breis, University of Murcia, Spain, Lars Nolle, Jade University of Applied Sciences, Germany, Elio Masciari, University of Napoli Federico II, Italy and David Aveiro, University of Madeira, NOVA-LINCS and ARDITI, Portugal. At the closing session, the conference acknowledged a few papers that were considered excellent in their class, presenting a "Best Paper Award", "Best Student Paper Award", and "Best Poster Award" for each of the co-located conferences.

The ability of Language Models (LMs) to understand natural language makes them a powerful tool for parsing human instructions into task plans for autonomous robots. Unlike traditional planning methods that rely on domain-specific knowledge and handcrafted rules, LMs generalize from diverse data and adapt to various tasks with minimal tuning, acting as a compressed knowledge base.

Knowledge-based Visual Question Answering (KB-VQA) requires VQA systems to utilize knowledge from external knowledge bases to answer visually-grounded questions. Retrieval-Augmented Visual Question Answering (RA-VQA), a strong framework to tackle KB-VQA, first retrieves related documents with Dense Passage Retrieval (DPR) and then uses them to answer questions. This paper proposes Fine-grained Late-interaction Multi-modal Retrieval (FLMR) which significantly improves knowledge retrieval in RA-VQA. FLMR addresses two major limitations in RA-VQA's retriever: (1) the image representations obtained via image-to-text transforms can be incomplete and inaccurate and (2) similarity scores between queries and documents are computed with one-dimensional embeddings, which can be insensitive to finer-grained similarities.FLMR overcomes these limitations by obtaining image representations that complement those from the image-to-text transform using a vision model aligned with an existing text-based retriever through a simple alignment network. FLMR also encodes images and questions using multi-dimensional embeddings to capture finer-grained similarities between queries and documents.

Continual learning of partially similar tasks poses a challenge for artificial neural networks, as task similarity presents both an opportunity for knowledge transfer and a risk of interference and catastrophic forgetting.However, it remains unclear how task similarity in input features and readout patterns influences knowledge transfer and forgetting, as well as how they interact with common algorithms for continual learning.Here, we develop a linear teacher-student model with latent structure and show analytically that high input feature similarity coupled with low readout similarity is catastrophic for both knowledge transfer and retention. Conversely, the opposite scenario is relatively benign. Our analysis further reveals that task-dependent activity gating improves knowledge retention at the expense of transfer, while task-dependent plasticity gating does not affect either retention or transfer performance at the over-parameterized limit. In contrast, weight regularization based on the Fisher information metric significantly improves retention, regardless of task similarity, without compromising transfer performance. Nevertheless, its diagonal approximation and regularization in the Euclidean space are much less robust against task similarity. We demonstrate consistent results in a permuted MNIST task with latent variables. Overall, this work provides insights into when continual learning is difficult and how to mitigate it.

Large Language Models (LLMs) are typically trained on data mixtures: most data come from web scrapes, while a small portion is curated from high-quality sources with dense domain-specific knowledge. In this paper, we show that when training LLMs on such data mixtures, knowledge acquisition from knowledge-dense datasets, unlike training exclusively on knowledge-dense data (arXiv:2404.05405), does not always follow a smooth scaling law but can exhibit phase transitions with respect to the mixing ratio and model size. Through controlled experiments on a synthetic biography dataset mixed with web-scraped data, we demonstrate that: (1) as we increase the model size to a critical value, the model suddenly transitions from memorizing very few to most of the biographies; (2) below a critical mixing ratio, the model memorizes almost nothing even with extensive training, but beyond this threshold, it rapidly memorizes more biographies. We attribute these phase transitions to a capacity allocation phenomenon: a model with bounded capacity must act like a knapsack problem solver to minimize the overall test loss, and the optimal allocation across datasets can change discontinuously as the model size or mixing ratio varies. We formalize this intuition in an information-theoretic framework and reveal that these phase transitions are predictable, with the critical mixing ratio following a power-law relationship with the model size. Our findings highlight a concrete case where a good mixing recipe for large models may not be optimal for small models, and vice versa.

We present a graphical, knowledge-based method for reviewing treatment-emergent adverse events (AEs) in clinical trials. The approach enhances MedDRA by adding a hidden medical knowledge layer (Safeterm) that captures semantic relationships between terms in a 2-D map. Using this layer, AE Preferred Terms can be regrouped automatically into similarity clusters, and their association to the trial disease may be quantified. The Safeterm map is available online and connected to aggregated AE incidence tables from ClinicalTrials.gov. For signal detection, we compute treatment-specific disproportionality metrics using shrinkage incidence ratios. Cluster-level EBGM values are then derived through precision-weighted aggregation. Two visual outputs support interpretation: a semantic map showing AE incidence and an expectedness-versus-disproportionality plot for rapid signal detection. Applied to three legacy trials, the automated method clearly recovers all expected safety signals. Overall, augmenting MedDRA with a medical knowledge layer improves clarity, efficiency, and accuracy in AE interpretation for clinical trials.

However, practical model deployment often operates in the gap between these two limit cases, as real-world applications demand adaptation to specific subdomains, tasks or concepts -- spread over the entire, varying life cycle of a model.

Imitating humans' mental world knowledge model which provides global prior knowledge before the task and maintains local dynamic knowledge during the task, in this paper, we introduce parametric W orld K nowledge M odel ( WKM) to facilitate agent