Cora
PulseAugur coverage of Cora — every cluster mentioning Cora across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
AI experiment finds limited autonomous capability in Cora
The Cora AI experiment, focused on testing autonomous capabilities rather than 'AI agents,' has reached its limits. Despite efforts to stimulate agency through extended context windows, community engagement, and tool ac…
-
New CoRA framework enables gradient-free on-device AI retrieval
Researchers have developed a new gradient-free framework called Conditional Retrieval Alignment (CoRA) for on-device in-context learning. CoRA converts a frozen encoder into a task-conditioned retriever by aligning cand…
-
New quantum graph learning architecture designed for NISQ era
Researchers have developed a novel quantum graph convolutional architecture specifically designed for unsupervised learning within the noisy intermediate-scale quantum (NISQ) era. This approach utilizes a variational qu…
-
Every launches All Access membership with $7,000+ AI tool bundle
Every has launched a new annual membership called All Access, aimed at individuals looking to build with AI. The membership includes a "Builder Pack" valued at over $7,000, offering credits and discounts for various AI …
-
AI automates coding, shifting engineer focus to final polish
Software development is shifting from manual coding to agent-assisted processes, where AI handles tasks like writing code, running tests, and fixing bugs. The human role is evolving to focus on the final polish and judg…
-
EdgeRefine framework improves privacy-utility balance in Graph Neural Networks
Researchers have developed EdgeRefine, a novel framework designed to enhance the privacy-utility balance in Graph Neural Networks (GNNs). This method addresses the challenge of sensitive link information leakage in grap…
-
New AI model CORA improves coronary artery disease assessment from CT scans
Researchers have developed CORA, a new self-supervised learning model designed to improve the assessment of coronary artery disease from CT angiography scans. Unlike previous methods that focus on global anatomy, CORA u…
-
New CORA method cuts LLM fine-tuning parameters by 4x
Researchers have introduced CORA (Coherent Orthogonal Rotation Adaptation), a novel parameter-efficient fine-tuning method for large language models. CORA leverages singular value decomposition (SVD) to preserve the geo…
-
Claude Fable 5 returns with subscription access, users share prompting tips
Anthropic's Claude Fable 5 model has returned after a brief unavailability, with access now included in Claude subscription plans until July 7th. Users are sharing prompts and strategies for effectively utilizing the mo…
-
PromptGNN-sim fuses GNNs and LLMs for advanced text-attributed graph learning
Researchers have developed PromptGNN-sim, a novel framework designed to enhance text-attributed graph learning by enabling deeper interaction between Graph Neural Networks (GNNs) and Large Language Models (LLMs). Unlike…
-
New PromptGNN-sim framework fuses GNNs and LLMs for enhanced graph learning
Researchers have introduced PromptGNN-sim, a novel framework designed to enhance the learning capabilities of Text-Attributed Graphs (TAGs) by deeply integrating Graph Neural Networks (GNNs) and Large Language Models (L…
-
New CGSD Algorithm Enhances Unsupervised Community Detection on Heterophilic Graphs
Researchers have developed a new unsupervised algorithm called Curvature-Guided Sheaf Diffusion (CGSD) for detecting communities in heterophilic graphs. This method utilizes the discrete Forman--Ricci curvature of edges…
-
New CGSD Algorithm Enhances Unsupervised Community Detection on Heterophilic Graphs
Researchers have introduced Curvature-Guided Sheaf Diffusion (CGSD), a novel unsupervised algorithm for community detection in heterophilic graphs. This method uniquely utilizes the discrete Forman--Ricci curvature of e…
-
AI enters allocation era: Token budgets to be managed like trading portfolios
Companies are shifting from aggressive AI adoption to a more strategic allocation of resources, as the cost of powerful models and long-running agents leads to significant enterprise bills without tangible results. This…
-
LLM Features Can Harm GNN Performance on Homophilous Graphs
A new research paper reveals that incorporating features generated by large language models (LLMs) into graph neural networks (GNNs) can sometimes decrease performance on specific benchmarks. This effect, termed 'concat…
-
New TNODEV Toolbox Enhances Neural ODE Verification
Researchers have developed TNODEV, a new toolbox designed for the formal verification of neural ordinary differential equations (neural ODEs). This tool addresses limitations in existing methods by integrating a falsifi…
-
New CORA method bridges thinking-answer gap in multimodal AI
Researchers have introduced CORA, a new method to address the thinking-answer inconsistency in multimodal large vision-language models (LVLMs). This inconsistency, where the reasoning process does not align semantically…
-
LLM-GNN Co-Teaching boosts few-shot graph learning accuracy
Researchers have developed a new method called LLM-GNN Co-Teaching to improve few-shot graph learning. This approach avoids designating one model as a "golden teacher," instead allowing a Graph Neural Network (GNN) and …
-
Compound engineering expands to include human ideation and polishing
The concept of compound engineering, which involves an AI agent executing tasks based on human goals, is evolving. Initially focused on the AI's execution loop of planning, working, and reviewing, the process now emphas…
-
Every's Guide to Codex: AI Workspace for Knowledge Work
The newsletter "Every" has published a comprehensive guide on leveraging Codex, an AI-powered workspace, for knowledge work. The guide details a five-step process for using Codex, from connecting data sources to compoun…