A study from an unnamed research team challenges the dominance of neural networks in evolutionary reinforcement learning, showing that programmatic policies implemented as soft, differentiable decisio…
A new method called DoLQ uses a multi-agent AI architecture, including a Scientist Agent powered by a large language model, to discover differential equations by combining qualitative and quantitative…
Inverse-LLaVA, a new multimodal architecture from researchers, challenges traditional multimodal learning by eschewing alignment pretraining and instead projecting text embeddings into visual represen…
Lighthouse RL, a new reinforcement learning method for analog circuit sizing, achieves up to 1.72x improvement in sample efficiency and 100% optimization success rate by using strategic resets from hi…
A new multi-agent reinforcement learning framework, Sensed Information Value Realization Multi-Agent Reinforcement Learning (SVR-MARL), aims to improve autonomous underwater vehicle (AUV) communicatio…
A new modular Context-Augmented Prompting framework boosts small language model accuracy in molecular property predictions by up to 74% on the Tox21 dataset, using graph-derived context from a trained…
MxGPS, a novel multiplex graph transformer, achieves a 0% boundary violation rate across four unseen power grid topologies, demonstrating that topology overfitting—not model capacity—is the primary ba…
Skip connections and normalization layers in Transformer architectures serve a critical function beyond gradient stabilization: they preserve the rank of representations across layers, preventing rank…
Neural networks trained on modular arithmetic tasks either succeed immediately or fail completely, according to research showing that algebraic constraints limit their expressible function class. Acro…
Researchers introduced a topology-aligned inductive bias that aligns model architecture with molecular bond graphs, embodied in two architectures: the Iso-QGNN variational quantum circuit and the Iso-…
Researchers have unveiled a new AI framework that acts as a digital twin for brain tumors, predicting tumor evolution and optimizing treatment. In tests with 387 synthetic tumor trajectories, the hybr…
A new machine learning optimization framework focuses on five real-world constraints — data availability, latency budget, memory budget, accuracy tolerance, and retraining budget — to guide deployment…
Quantum Circuit Vision, a new benchmark of 132 quantum circuits across 13 categories, evaluates multimodal AI agents' ability to understand circuit diagrams and generate executable code for Amazon Bra…
An influential game designer argues that video games, which mirror human cognition through complex problem-solving and strategic planning, could provide a blueprint for understanding and improving lar…
AI revenue growth expectations may be overhyped, with global AI revenue estimated at $150 billion in 2022 and projected to reach $300 billion by 2025, but scaling profitably remains a hurdle for hyper…
TRACE, a novel dense credit-assignment framework for reinforcement learning, dramatically improved agent performance on the BrowseComp-Plus benchmark, boosting Qwen3-4B from 7.2 to 35.6 and Qwen3-30B-…
DAGR, a new approach to goal-conditioned reinforcement learning, refines state-independent goal encoders into state-conditioned ones using multi-scale gated cross-attention, improving navigation tasks…
FastCentNN, an upgrade to Centroid Neural Networks, reduces runtime by up to 16% on synthetic 2D datasets and about 5% on high-dimensional ones while maintaining clustering quality, according to its d…
A survey of 49 operational sources found that fraud detection studies using large language models (LLMs) lack transparency in key metrics like per-decision latency and cost, while content moderation s…
FaStR, a new AI model using tensor decompositions for reinforcement learning, reduces sample size requirements by a factor scaling with the smaller of state and action dimensions. Developed by researc…