Ursa Major
PulseAugur coverage of Ursa Major — every cluster mentioning Ursa Major across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
New GUARD framework autoformalizes argumentative inferences with LLMs and theorem provers
Researchers have developed a neuro-symbolic framework called GUARD to address the challenge of autoformalizing argumentative material inferences. This system uses large language models to construct and formalize candida…
-
New neuro-symbolic method enhances enthymeme completion with logical resistance scores
Researchers have developed a new neuro-symbolic method called Possible-World Atom-Link Formalization (PWAL) for completing enthymemes, which are arguments with missing premises or claims. PWAL extends prior work by repl…
-
New agent system automates complex atomistic simulations
Researchers have developed an agent-based system, integrated into the URSA framework, designed to automate complex, multi-step atomistic simulations. This system can autonomously select interatomic potentials, construct…
-
New E-SpecFormer model offers efficient RF spectrum monitoring for IoT
Researchers have developed E-SpecFormer, a new Transformer model designed for efficient RF spectrum monitoring on edge devices. This model incorporates a novel attention mechanism called LiTAN, which reduces computation…
-
New URSA benchmark evaluates AI for drug discovery synthesis planning
Researchers have introduced URSA, a new benchmark framework designed to evaluate retrosynthesis systems in drug discovery. URSA assesses synthetic routes not only for convergence to starting materials but also for chemi…
-
NVIDIA Vera CPUs to accelerate agentic AI for science at Los Alamos National Laboratory
NVIDIA's Vera CPU is set to power new supercomputers at Los Alamos National Laboratory (LANL), enhancing scientific discovery and agentic AI capabilities. The new systems, named Mission, Vision, and Veritas, will integr…
-
KaLM-Reranker-V1: Efficient Document Reranking Model Unveiled
Researchers have introduced KaLM-Reranker-V1, a novel reranking model designed for efficiency in large-scale retrieval systems. This model decouples query and passage computation using an encoder-decoder architecture wi…
-
Tessera offers secure, near-line-rate weight streaming for edge AI accelerators
Researchers have developed Tessera, a new architecture designed to securely stream model weights to edge accelerators in Unified Memory Architecture (UMA) systems. This approach addresses the challenge of protecting pro…