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 MELLA, a multimodal dataset addressing culturally 'thin' descriptions from MLLMs in low-resource languages by focusing on native visual-textual alignments.
Research addresses challenges in voice anonymization for long-form audio, noting current methods struggle with speaker privacy over multiple utterances.
Researchers introduced WebCoach, a framework for self-evolving web agents with cross-session memory guidance to reduce repetitive errors and improve learning.
Research proposes StackingNet, a method for collective inference across independently developed, black-box AI foundation models to share capabilities.
Research finds LLM political bias depends on evaluation instrument, with abstract questionnaires showing left-leaning bias but concrete policy questions showing centrist views.
ARCO proposes adaptive, co-evolving, natural-language rubrics for multi-step LLM agent rewards, addressing limitations of scalar and static rubric-based methods.
New benchmark, DataPrep-Bench, introduced to evaluate LLMs as end-to-end training data preparators, focusing on data construction and quality evaluation.
Research shows LLMs hallucinate more when forced to fill structured forms like JSON or function arguments, even when the input lacks an answer.
Research demonstrates scaling closed-loop LLM-based channel configuration for neural networks by optimizing widths via executable code generation.
Researchers propose Multimodal CoLRAG-TF, a quadruple-fusion RAG architecture combining text, keywords, knowledge graphs, and image similarity for complex PDFs.
Research explores 'looped Transformers' with adaptive depth, analyzing how learned halting gates influence inference computation and trajectory formation.
Research proposes Generative Bayesian Filtering, a new method for state estimation in dynamic systems with complex observation models.
Research proposes a method for uncertainty-aware trust estimation in multi-LLM systems using structured expert judgment to improve reliability.
Research proposes a shift from reactive, error-patching AI maintenance to proactive, test-driven development for deployed systems to improve generalization.
Research proposes using neural predicates to formalize and scale view generation in the Black-Litterman model for portfolio construction, reducing subjectivity.
Research introduces a Graph Neural Network (GNN) framework for 'Zero-Shot Digital Twins' that adapts to changing physical domains without retraining.
ReliableTableQA is a framework training LLMs to annotate the statistical reliability of tabular QA results, assessing if answers are statistically meaningful.
Codec-Gauge introduces a post-training cache-coordinate layer to improve KV-cache compression and quantization in long-context Transformer models.
Research explores using biokinetic knowledge priors for deep learning in biomanufacturing to address data scarcity with ODE models.
Research proposes a statistical framework for optimal noise-level allocation in diffusion model training, moving beyond heuristic schedules.
Researchers introduced HypNO, a graph-based neural operator using physics-informed message passing for hyperbolic conservation laws, benchmarked on traffic-flow models.
Research finds Reinforcement Learning with Verifiable Rewards (RLVR) can improve one-sample accuracy while decreasing distinct problem-solving ability.
Research uses multi-agent reinforcement learning (MARL) for decentralized conflict resolution in air corridors under degraded surveillance.
Research details higher-order optimizers like Muon and SOAP for faster LLM pretraining convergence, addressing stability issues at scale.
SevDiff is a new diffusion model for generating rare, high-severity conflict trajectories in ADAS evaluation, addressing bias in real-world data.
Research identifies 'attention specific heat' as a reliable predictor for Grokking, improving neural network generalization and training efficiency.
Research explores reliable extraction of "strong lottery tickets"—sparse, performant subnetworks in large neural networks prior to training.
A new research paper introduces Monkey King Bang, a multimodal foundation model designed for unified scientific reasoning across diverse domains and data types.
StabilityBench is a new research benchmark evaluating LLM instability and context dependence, particularly in high-stakes environments.
A new deep learning framework, JuGAAD, uses geospatial and census data to downscale socioeconomic indicators in India, addressing data resolution mismatches.
© 2026 OneBench: AI Insights. All rights reserved.
Evidence before opinion