AI News — Latest Developments & Breakthroughs

ArXiv cs.AI 📄 Paper ⚡ AI Lesson 12h ago

On the Complexity of Optimal Graph Rewiring for Oversmoothing and Oversquashing in Graph Neural Networks

arXiv:2603.26140v1 Announce Type: cross Abstract: Graph Neural Networks (GNNs) face two fundamental challenges when scaled to deep architectures: oversmoothing,

ArXiv cs.AI 📄 Paper 12h ago

Progressive Learning with Anatomical Priors for Reliable Left Atrial Scar Segmentation from Late Gadolinium Enhancement MRI

arXiv:2603.26186v1 Announce Type: cross Abstract: Cardiac MRI late gadolinium enhancement (LGE) enables non-invasive identification of left atrial (LA) scar, wh

ArXiv cs.AI 📄 Paper 12h ago

MemCam: Memory-Augmented Camera Control for Consistent Video Generation

arXiv:2603.26193v1 Announce Type: cross Abstract: Interactive video generation has significant potential for scene simulation and video creation. However, exist

ArXiv cs.AI 👁️ Computer Vision 📄 Paper ⚡ AI Lesson 12h ago

An Object Web Seminar: A Retrospective on a Technical Dialogue Still Reverbarating

arXiv:2603.26203v1 Announce Type: cross Abstract: Technology change happens quickly such that new trends tend to crowd out the focus on what was new just yester

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

Sparse Auto-Encoders and Holism about Large Language Models

arXiv:2603.26207v1 Announce Type: cross Abstract: Does Large Language Model (LLM) technology suggest a meta-semantic picture i.e. a picture of how words and com

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

Towards GUI Agents: Vision-Language Diffusion Models for GUI Grounding

arXiv:2603.26211v1 Announce Type: cross Abstract: Autoregressive (AR) vision-language models (VLMs) have long dominated multimodal understanding, reasoning, and

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

Clawed and Dangerous: Can We Trust Open Agentic Systems?

arXiv:2603.26221v1 Announce Type: cross Abstract: Open agentic systems combine LLM-based planning with external capabilities, persistent memory, and privileged

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

Automating Domain-Driven Design: Experience with a Prompting Framework

arXiv:2603.26244v1 Announce Type: cross Abstract: Domain-driven design (DDD) is a powerful design technique for architecting complex software systems. This pape

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

Physics-Informed Neural Networks and Sequence Encoder: Application to heating and early cooling of thermo-stamping process

arXiv:2603.26245v1 Announce Type: cross Abstract: In a previous work (Elaarabi et al., 2025b), the Sequence Encoder for online dynamical system identification (

ArXiv cs.AI 📄 Paper 12h ago

Distilling Conversations: Abstract Compression of Conversational Audio Context for LLM-based ASR

arXiv:2603.26246v1 Announce Type: cross Abstract: Standard LLM-based speech recognition systems typically process utterances in isolation, limiting their abilit

ArXiv cs.AI 📄 Paper ⚡ AI Lesson 12h ago

Automatic Speech Recognition for Documenting Endangered Languages: Case Study of Ikema Miyakoan

arXiv:2603.26248v1 Announce Type: cross Abstract: Language endangerment poses a major challenge to linguistic diversity worldwide, and technological advances ha

ArXiv cs.AI 📄 Paper ⚡ AI Lesson 12h ago

Channelling, Coordinating, Collaborating: A Three-Layer Framework for Disability-Centered Human-Agent Collaboration

arXiv:2603.26252v1 Announce Type: cross Abstract: AI accessibility tools have mostly been designed for individual use, helping one person overcome a specific fu

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

ARTA: Adaptive Mixed-Resolution Token Allocation for Efficient Dense Feature Extraction

arXiv:2603.26258v1 Announce Type: cross Abstract: We present ARTA, a mixed-resolution coarse-to-fine vision transformer for efficient dense feature extraction.

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

Working Notes on Late Interaction Dynamics: Analyzing Targeted Behaviors of Late Interaction Models

arXiv:2603.26259v1 Announce Type: cross Abstract: While Late Interaction models exhibit strong retrieval performance, many of their underlying dynamics remain u

ArXiv cs.AI 👁️ Computer Vision 📄 Paper ⚡ AI Lesson 12h ago

GeoGuide: Hierarchical Geometric Guidance for Open-Vocabulary 3D Semantic Segmentation

arXiv:2603.26260v1 Announce Type: cross Abstract: Open-vocabulary 3D semantic segmentation aims to segment arbitrary categories beyond the training set. Existin

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

Knowdit: Agentic Smart Contract Vulnerability Detection with Auditing Knowledge Summarization

arXiv:2603.26270v1 Announce Type: cross Abstract: Smart contracts govern billions of dollars in decentralized finance (DeFi), yet automated vulnerability detect

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

PhysVid: Physics Aware Local Conditioning for Generative Video Models

arXiv:2603.26285v1 Announce Type: cross Abstract: Generative video models achieve high visual fidelity but often violate basic physical principles, limiting rel

ArXiv cs.AI 📄 Paper ⚡ AI Lesson 12h ago

findsylls: A Language-Agnostic Toolkit for Syllable-Level Speech Tokenization and Embedding

arXiv:2603.26292v1 Announce Type: cross Abstract: Syllable-level units offer compact and linguistically meaningful representations for spoken language modeling

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

Preference-Aligned LoRA Merging: Preserving Subspace Coverage and Addressing Directional Anisotropy

arXiv:2603.26299v1 Announce Type: cross Abstract: Merging multiple Low-Rank Adaptation (LoRA) modules is promising for constructing general-purpose systems, yet

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

Label-Free Cross-Task LoRA Merging with Null-Space Compression

arXiv:2603.26317v1 Announce Type: cross Abstract: Model merging combines independently fine-tuned checkpoints without joint multi-task training. In the era of f

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

From Human Cognition to Neural Activations: Probing the Computational Primitives of Spatial Reasoning in LLMs

arXiv:2603.26323v1 Announce Type: cross Abstract: As spatial intelligence becomes an increasingly important capability for foundation models, it remains unclear

ArXiv cs.AI 📄 Paper 12h ago

PRISMA: Toward a Normative Information Infrastructure for Responsible Pharmaceutical Knowledge Management

arXiv:2603.26324v1 Announce Type: cross Abstract: Most existing approaches to AI in pharmacy collapse three epistemologically distinct operations into a single

ArXiv cs.AI 📄 Paper 12h ago

Mitigating the Reasoning Tax in Vision-Language Fine-Tuning with Input-Adaptive Depth Aggregation

arXiv:2603.26330v1 Announce Type: cross Abstract: Supervised fine-tuning (SFT) on visual instruction data often improves perceptual capabilities in vision-langu

ArXiv cs.AI 🧠 Large Language Models 📄 Paper ⚡ AI Lesson 12h ago

CALRK-Bench: Evaluating Context-Aware Legal Reasoning in Korean Law

arXiv:2603.26332v1 Announce Type: cross Abstract: Legal reasoning requires not only the application of legal rules but also an understanding of the context in w

📰 AI News