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 highlights a 'render confound' in evaluating LLM memory with revising records, where prompt presentation, not underlying mechanism, skews results.
Research explores constrained Hebbian learning for efficient neural representations under biological constraints, testing synaptic resource allocation.
Research explores using a tabular foundation model for dynamic security assessment (DSA) in power systems, aiming to reduce data labeling needs.
Researchers developed DELUGE, a multimodal deep learning framework for daily, continental-scale pluvial flood damage prediction.
Research compares model merging to joint multi-task training for reinforcement learning agents, using Qwen3-8B on AppWorld to test performance.
Research on physics-based deep learning U-Net for high-resolution precipitation nowcasting (10-90 min period) for urban flood management.
Researchers propose DADiff, a diffusion-driven method for cross-domain policy adaptation in reinforcement learning, addressing dynamics mismatch with limited target domain interaction.
Research explores 'honest quorum problem' for agentic infrastructure, addressing how to guarantee reliability when agents may endorse faulty states.
Research from arXiv explores multi-agent system (MAS) advantages over single-agent systems (SAS) using an information bottleneck perspective.
Research explores embedding inferred behavioral structure into Neural Process-based models for adaptive residential short-term load forecasting.
Research finds the Muon optimizer significantly improves agentic reinforcement learning performance (+88% validation success) over AdamW in sparse-reward environments.
Physics-enhanced reinforcement learning (RL) improves sample efficiency for optimal control of complex dynamical systems, addressing a key RL limitation.
PagedWeight proposes dynamically quantizing MoE LLM weights to balance memory for model weights and KV cache, improving serving efficiency.
Research finds Chain-of-Thought (CoT) reasoning in LLMs complicates refusal control, making activation steering less effective than with non-CoT models.
AV-JEPA extends LeJEPA for audio-visual self-supervised learning, using an early-fusion Vision Transformer and modality dropout.
MLLM-DataEngine is a research proposal for a closed-loop system to iteratively generate multimodal instruction tuning data, train, and evaluate MLLMs.
Research explores using entropy-based features to enhance supervised network traffic classification for anomaly detection, addressing diverse traffic patterns.
Research explores Choquet-integral-based feature aggregation with Random Forest and XGBoost to enhance network anomaly detection.
Research proposes a framework using ML to proactively request inpatient beds for emergency department patients, aiming to reduce boarding time.
Research explores self-distillation for linear and logistic regression models when original training data is unavailable, using fresh unlabeled covariates.
Research demonstrates fingerprinting federated learning architectures using 5G PHY-layer side channels, even with encrypted payloads.
Research introduces a memory compression technique for Caputo fractional gradient descent, reducing its computational cost from quadratic to linear.
Research identifies critical vulnerability in facial recognition to intentional electromagnetic interference attacks, extending physical presentation risks.
New research explores performativity in predictive models, where model deployment influences future data, creating feedback loops during retraining.
New research explores distribution testing against bounded classes of distinguishers to assess if an unknown distribution matches a reference.
Research demonstrates a black-box adversarial framework, Lucid, to implant false visual memories in multimodal AI agents using image-bounded attacks.
Research details SpeechGuard, an online defense mechanism to detect and mitigate backdoor attacks targeting speech recognition models at runtime.
Research proves a finite-sample formulation gap for physics-informed learning in nonlinear multiscale elliptic equations, improving stability.
Research explores multi-modal foundation models for wireless localization using spatial signals, aiming for improved generalization in diverse environments.
Research demonstrates 'natural backdoor attacks' on speech recognition models using ordinary sounds as triggers, affecting model performance.
© 2026 OneBench: AI Insights. All rights reserved.
Evidence before opinion