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 introduces a Von Mises-Fisher Mixture Model with dynamic shrinkage to enhance vision-language model performance under imbalanced test-time data.
Research demonstrates a large-scale remote sensing VLM achieves strong performance using a simplified architecture, challenging specialized designs.
Research identifies 'code-poisoning property inference attacks' as a new threat, where malicious code on hosting platforms could leak private training data properties.
Researchers developed CanonicalPhys, an rPPG model achieving robust heart rate monitoring across varying head poses by using canonical-space priors.
Research introduces Pick-to-Learn methodology for calibrating Model Predictive Control (MPC) policies, demonstrated with an aircraft flight problem.
CRAFT is a research method that converts rubric-based evaluations into tools for diagnosing specific LLM capability failures and generating targeted fine-tuning data.
Researchers propose Visualized Learning for Machine Learning (VL4ML), a human-centered framework to communicate AI predictions and uncertainty visually.
AutoSpec automates the generation of neural network specifications, addressing the manual, error-prone process in model verification for safety-critical systems.
New research proposes the Composite Task Challenge (CTC) to benchmark cooperative multi-agent reinforcement learning (MARL) for division of labor.
Research improves CLIP for training-free open-vocabulary semantic segmentation by refining patch-wise representations for dense prediction tasks.
Research proposes MAnchors, a memorization-based framework to accelerate Anchors, a local model-agnostic explanation technique, improving efficiency.
AuditVotes introduces a framework to improve the certified robustness and accuracy of Graph Neural Networks (GNNs) against adaptive attacks.
Research introduces a multi-marginal temporal Schrödinger Bridge method to reconstruct dynamic processes from unpaired static snapshots, improving scalability.
Research questions the intrinsic effectiveness of Heterogeneous Graph Neural Networks (HGNNs) for node classification despite their success.
New research introduces Configuration-Mixed Prediction (CMP), a method for adaptively weighting clustering configurations per sample, moving beyond fixed resolutions.
DiffuMamba, a new diffusion language model, uses a bidirectional Mamba backbone to improve inference efficiency over Transformer-based DLMs.
SloMo-Fast proposes a source-free continual test-time adaptation method to prevent catastrophic forgetting in evolving AI models, without using source data.
Research on time-varying mixing matrices in decentralized federated learning aims to minimize per-node energy consumption in wireless networks.
Research paper proposes PASs-MoE, a method to mitigate router and expert co-drift in LoRA-based Mixture-of-Experts for continual learning in MLLMs.
Researchers propose SC-JEPA, a method to stabilize latent predictive learning for time-series anomaly prediction, addressing instability in JEPA.
New research proposes an improved method for optimizing orthogonal matrices, potentially scaling to thousands of constraints, building on the Landing algorithm.
Research explores pseudo-calibration with conformal prediction to maintain marginal coverage guarantees under bounded label-conditional covariate shift.
Research explores dynamic budget allocation for multi-turn LLM evaluation to efficiently identify jailbreaks or task completion in conversational settings.
Research proposes a new variational inference method for evidential deep learning (EDL) to address limitations in quantifying epistemic uncertainty.
Research proposes a classifier-based adaptive stopping framework for sampling kernels in Bayesian inference to improve MCMC efficiency.
Research introduces Factorized Neural Operators to better capture multiscale physical behavior by decomposing dynamic and persistent responses.
GeoRouteNet is a new non-autoregressive neural solver for the Traveling Salesman Problem (TSP) that uses geometric features to improve transferability.
Research proposes decoupling magnitude and direction in neural network weight updates to improve training efficiency and stability beyond current optimizers.
Research proposes an LLM agent framework to integrate natural language business context into hub capacity planning, using a chain-of-thought protocol.
Research proposes a method for generating differentially private synthetic data designed to preserve causal estimands for accurate causal inference.
© 2026 OneBench: AI Insights. All rights reserved.
Evidence before opinion