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 decentralized multi-agent reinforcement learning (MARL) for improving resilience in critical infrastructures.
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.
Research details the exact error conditions required for tractable, unbiased sampling with inexact score oracles, ruling out weaker error assumptions.
Research paper proposes a deep reinforcement learning approach to optimize service placement, computational delegation, and power in edge-cloud networks.
Research introduces a systematic framework and benchmark scenarios for continual anomaly detection (CAD) in tabular data, addressing limitations of current benchmarks.
ABot-World-0, an action-conditioned video world model, enables real-time, long-horizon interactive world rollout on a single desktop GPU.
Research introduces 1-Lipschitz neural networks on Hadamard manifolds, aiming to improve robustness and stability beyond Euclidean space methods.
Research introduces a method for exact recovery of Gaussian graphical models from a single trajectory of Gaussian Glauber dynamics.
Research explores using algebraic signatures derived from observed vanishing binomials in probability tensors to identify probabilistic model structure.
Research explores federated learning (FL) for continuous client onboarding without catastrophic forgetting, using hypernetworks and data-free replay.
Research introduces E2E-CDiff, a conditional diffusion model generating realistic, controllable visual traffic scenarios for autonomous driving evaluation.
Research details Quantum Reservoir Computing (QRC), using quantum system dynamics for temporal input transformation and classical readout training.
A new benchmark, HALLMARK, diagnoses three failure modes in LLM citation verifiers, addressing high rates of hallucinated references in academic papers.
Research proposes Decode-Time Grammars for constrained LLM generation, improving code accuracy for domain-specific languages and APIs.
Research benchmarks deep learning models for layout detection and information extraction specifically on Architecture, Engineering, and Construction (AEC) drawings.
Research presents a generative state-space model learning directly from sparse, noisy observations to improve ocean modeling beyond complete datasets.
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 explores shallow recurrent decoder networks for real-time optimal control in computationally demanding dynamical systems.
New research argues that 'circuit extraction' methods for identifying AI model mechanisms are under-determined, with findings depending on reporting and comparison choices.
Researchers propose SFGA, a statistics-first gating architecture for evaluating the quality of supervised fine-tuning (SFT) data based on diversity, utility, and redundancy.
Research explores network-agnostic feature initialization for Graph Neural Networks to improve spatial transferability in traffic assignment models.
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.
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.
Research explores using binary silver labels, like diagnosis codes or NLP mentions, to improve weakly supervised phenotyping algorithms in EHR data.
Research proposes a censoring-aware in-context learning method for generalized supplier lead time estimation, addressing right-censored data.
New research proposes an optimization stack for Reinforcement Learning with Verifiable Rewards (RLVR), focusing on reusing base model weight spectra.
Interactive Training 2 introduces an open-source control plane for auditable steering of live model training, abstracting trainer-specific code.
Research introduces the 'information shadow,' structural limits on what language models can learn from text alone, beyond data coverage gaps.
Research proposes a federated fine-tuning method that reduces communication bottleneck by averaging small trainable latents instead of full model weights.
Research explores Wireless Physical Neural Networks (WPNNs) that embed neural computation in analog hardware using MIMO relays and power amplifiers.
Research introduces Segmented Continuous Optimization (SCO) for piecewise fitting of non-stationary time-series with oscillatory behavior.
A research paper systematically compares two geospatial foundation models, TerraMind and THOR, developed under the European Space Agency's $\Phi$-lab.
Research introduces a unified framework for Riemannian deep learning with reusable modules and manifold-specific architectures.
Research indicates that current LLM alignment schemes and safety filters may not reduce harmful outputs to zero, posing challenges for robust safety.
Research explores compound sparsity, combining parameter pruning and token-level computation, to improve LLM compression limits and maintain performance.
Researchers explored diffusion generative models to address limited training data for Auditory Attention Decoding (AAD) using EEG in hearing aids.
Research proposes 'Spectral Evidence Bundling' to improve reliability estimation for time-series classification models beyond output score calibration.
FALCON-Discover proposes a model-agnostic framework to find concentrated regions of false confidence in ML predictions, improving local calibration.
Research introduces a decision-aware weak-to-strong (W2S) learning framework to improve predictive models using both labeled and unlabeled data.
BearingNAS introduces a hardware-aware neural architecture search framework to deploy intelligent fault diagnosis directly on sensor microcontrollers.
Research explores 'solver-in-the-loop' interactive optimization modeling using natural language, moving beyond one-shot LLM approaches.
AdaFlash introduces adaptive speculative decoding using diffusion drafters to accelerate large language model inference by generating drafts in a single pass.
Researchers introduced MedDDC-Eval, a diagnosis-decoupled evaluation framework for multi-turn medical consultation agents to isolate policy elicitation from diagnosis generation.
Research explores enhancing legal machine translation using reasoning-capable language models to improve precision and address linguistic complexity.
Research explores small language models (SLMs) for creative plot generation, focusing on global coherence, character, pacing, and emotional progression.
ChineseBERT incorporates glyph and pinyin information into pretraining models for enhanced Chinese language understanding.
© 2026 OneBench: AI Insights. All rights reserved.
Evidence before opinion