Search signals, briefings, company results, benchmarks and glossary terms.
Search signals, briefings, company results, benchmarks and glossary terms.
Use this view to inspect the underlying evidence corpus. For ranked developments, decision posture and interpretation, use Signals.
Raw feed or Signals?
Raw feed is chronological evidence. Signals ranks and interprets material change.
Research explores specialized multi-generator adversarial learning to detect rare, non-homogeneous failures in predictive maintenance using imbalanced datasets.
Researchers explored diffusion generative models to address limited training data for Auditory Attention Decoding (AAD) using EEG in hearing aids.
Research paper proposes Utility-Augmented Transformer (UAT) to address 'feedback-blind' retrieval in sequential decision-making models, enhancing adaptation.
New research proposes ROMS-IMLE, a minimalist, single-step generative model challenging multi-step diffusion/flow matching complexities.
Research explores decoding visual perception from brain activity using ECoG and deep learning to predict visual categories from videos.
Research explores using LLMs for agentic calibration of grey-box simulation models, addressing high-dimensional parameter spaces and expensive evaluations.
Research introduces JD-BP, a joint-decision generative framework to optimize auto-bidding and pricing strategies for advertisers, addressing uncertainties.
Researchers introduced PertReason, a new benchmark and framework for evaluating mechanistic reasoning in ML models within scientific domains.
Research proposes a ReRAM-aware finetuning method for large-scale models to mitigate hardware non-idealities in In-Memory Computing, avoiding full retraining.
Research introduces 'partial fusion' of neural networks, offering a tradeoff between computational cost and performance, interpolating between ensembles and weight aggregation.
Research proposes Latent Distribution Matching as a unifying theoretical framework for self-supervised learning, aiming to explain diverse methods.
Research introduces RF-Deep, a random forest method to improve out-of-distribution detection in lung cancer segmentation from CT scans.
Research proposes a drift-adaptive two-stream AI architecture for ICU prediction, decoupling physiological and treatment data to prevent silent degradation.
New research details cost accounting for exhaustive site-by-site interventions on neural network computational graphs, focusing on recomputation costs.
Research explores federated learning (FL) for continuous client onboarding without catastrophic forgetting, using hypernetworks and data-free replay.
Research details BETA, an indirect evasion attack targeting Graph Neural Network-based anomaly detection in sensor networks by perturbing non-target sensors.
Research explores Topology-Driven Clustering using Betti Number Filtration to enhance performance for datasets with complex geometric structures.
A research survey explores Parameter-Efficient Continual Fine-Tuning (PECFT) methods for large pre-trained models to adapt to dynamic data.
MAAT framework reconstructs latent dynamical states from partial, noisy, and heterogeneous observations using a knowledge-informed kernel approach.
Research compares Monte Carlo Dropout and Deep Ensemble for uncertainty quantification in AI-driven crash simulation surrogates, using an open-source bumper beam.
Research explores using LLM agents to automatically construct clinical scoring systems, aiming to improve interpretability and workflow alignment.
Research introduces geometric analysis of transformer residual streams, measuring representation movement and structural changes across model layers.
Research benchmarks deep learning models for layout detection and information extraction specifically on Architecture, Engineering, and Construction (AEC) drawings.
Research details the exact error conditions required for tractable, unbiased sampling with inexact score oracles, ruling out weaker error assumptions.
Research explores using algebraic signatures derived from observed vanishing binomials in probability tensors to identify probabilistic model structure.
ABot-World-0, an action-conditioned video world model, enables real-time, long-horizon interactive world rollout on a single desktop GPU.
Research introduces a method for exact recovery of Gaussian graphical models from a single trajectory of Gaussian Glauber dynamics.
New research introduces S3, a Hierarchical Reinforcement Learning approach using uncertainty-constrained subgoal selection to improve long-horizon task solving.
Research details Quantum Reservoir Computing (QRC), using quantum system dynamics for temporal input transformation and classical readout training.
Research introduces E2E-CDiff, a conditional diffusion model generating realistic, controllable visual traffic scenarios for autonomous driving evaluation.
Research proposes Decode-Time Grammars for constrained LLM generation, improving code accuracy for domain-specific languages and APIs.
Research introduces 1-Lipschitz neural networks on Hadamard manifolds, aiming to improve robustness and stability beyond Euclidean space methods.
A new benchmark, HALLMARK, diagnoses three failure modes in LLM citation verifiers, addressing high rates of hallucinated references in academic papers.
Research applies machine learning super-resolution to downscale atmospheric data for better detection and forecasting of extreme weather events.
New research addresses quantifying the risk of rare failures in language models, proposing improved methods for low probability estimation to avoid bias.
Research explores using reinforcement learning with verifiable rewards for molecular generation, improving LLM-based chemical and drug design.
Research explores Parallel Noising in Neural Markov Logic Networks (NMLNs) to improve generative modeling performance on larger relational structures.
Neural Kolmogorov Equations (NKEs) are proposed for learning stochastic dynamics with parallelizable training and support for general noise types.
Research presents a generative state-space model learning directly from sparse, noisy observations to improve ocean modeling beyond complete datasets.
Research explores network-agnostic feature initialization for Graph Neural Networks to improve spatial transferability in traffic assignment models.
Research introduces Decafs, a flow-based generative model using Lie groups to disentangle latent spaces, improving interpretability and controlled generation.
New research introduces Posterior Prefix Tuning (PPT), a method to steer transformer behavior by optimizing the latent posterior without backpropagation.
New research argues that 'circuit extraction' methods for identifying AI model mechanisms are under-determined, with findings depending on reporting and comparison choices.
New research proposes an optimization stack for Reinforcement Learning with Verifiable Rewards (RLVR), focusing on reusing base model weight spectra.
Research proposes variational meta-learning for neural system identification to improve data efficiency and uncertainty quantification in low-data regimes.
Research proposes a censoring-aware in-context learning method for generalized supplier lead time estimation, addressing right-censored data.
Research introduces GNNAS-TSP, a Graph Neural Network (GNN)-based framework for automated algorithm selection (AS) in the Traveling Salesman Problem.
ConceptCF introduces a method for generating counterfactual explanations for time series models based on human-interpretable concepts.
Researchers propose SFGA, a statistics-first gating architecture for evaluating the quality of supervised fine-tuning (SFT) data based on diversity, utility, and redundancy.
A research paper systematically compares two geospatial foundation models, TerraMind and THOR, developed under the European Space Agency's $\Phi$-lab.
© 2026 OneBench: AI Insights. All rights reserved.
Evidence before opinion