12,864 AI articles curated from 50+ sources with AI-powered sentiment analysis, importance scoring, and key takeaways.
Researchers introduce PolyGLU, a new transformer architecture that enables dynamic routing among multiple activation functions, mimicking biological neural diversity. The 597M-parameter PolychromaticLM model shows emergent specialization patterns and achieves strong performance despite training on significantly fewer tokens than comparable models.
Researchers propose Latent Entropy-Aware Decoding (LEAD), a new method to reduce hallucinations in multimodal large reasoning models by switching between continuous and discrete token embeddings based on entropy states. The technique addresses issues where transition words correlate with high-entropy states that lead to unreliable outputs in visual question answering tasks.
Researchers introduce FL-I2MoE, a new Mixture-of-Experts layer for multimodal Transformers that explicitly identifies synergistic and redundant cross-modal feature interactions. The method provides more interpretable explanations for how different data modalities contribute to AI decision-making compared to existing approaches.
Researchers introduce Flare, a new AI fairness framework that ensures ethical outcomes without requiring demographic data, addressing privacy and regulatory concerns in human-centered AI applications. The system uses Fisher Information to detect hidden biases and includes a novel evaluation metric suite called BHE for measuring ethical fairness beyond traditional statistical measures.
Researchers propose Evi-DA, an evidence-based technique that improves how large language models predict population response distributions across different cultures and domains. The method uses World Values Survey data and reinforcement learning to achieve up to 44% improvement in accuracy compared to existing approaches.
Researchers introduce Pragma-VL, a new alignment algorithm for Multimodal Large Language Models that balances safety and helpfulness by improving visual risk perception and using contextual arbitration. The method outperforms existing baselines by 5-20% on multimodal safety benchmarks while maintaining general AI capabilities in mathematics and reasoning.
Research reveals that LLM query rewriting in RAG systems shows highly domain-dependent performance, degrading retrieval effectiveness by 9% in financial domains while improving it by 5.1% in scientific contexts. The study identifies that effectiveness depends on whether rewriting improves or worsens lexical alignment between queries and domain-specific terminology.
Researchers propose FedTreeLoRA, a new framework for privacy-preserving fine-tuning of large language models that addresses both statistical and functional heterogeneity across federated learning clients. The method uses tree-structured aggregation to allow layer-wise specialization while maintaining shared consensus on foundational layers, significantly outperforming existing personalized federated learning approaches.
Researchers introduce Truncated-Reasoning Self-Distillation (TRSD), a post-training method that enables AI language models to maintain accuracy while using shorter reasoning traces. The technique reduces computational costs by training models to produce correct answers from partial reasoning, achieving significant inference-time efficiency gains without sacrificing performance.
Researchers developed PREBA, a retrieval-augmented framework that uses PCA-weighted retrieval and Bayesian averaging to improve surgical duration prediction accuracy by up to 40% using large language models. The system grounds LLM predictions in institution-specific clinical data without requiring computationally intensive training, achieving performance competitive with supervised machine learning methods.
Researchers introduce SPLARE, a new method that uses sparse autoencoders (SAEs) to improve learned sparse retrieval in language models. The technique outperforms existing vocabulary-based approaches in multilingual and out-of-domain settings, with SPLARE-7B achieving top results on multilingual retrieval benchmarks.
Researchers discovered that transformer language models process factual information through rotational dynamics rather than magnitude changes, actively suppressing incorrect answers instead of passively failing. This geometric pattern only emerges in models above 1.6B parameters, suggesting a phase transition in factual processing capabilities.
Researchers introduce a structural taxonomy and unified evaluation framework for Audio Large Language Models (ALLMs) to assess fairness, safety, and security. The study reveals systematic differences in how ALLMs handle audio versus text inputs, with FSS behavior closely tied to acoustic information integration methods.
Researchers developed a framework to assess public summaries of AI training data required by EU's AI Act Article 53(1)(d), evaluating transparency and usefulness for stakeholder rights enforcement. The study analyzed 5 public summaries from GPAI model providers as of January 2026, creating guidelines for compliance and a public resource website.
Researchers propose autonomous editorial systems that use AI to continuously process, analyze, and organize large volumes of news and information. The system treats stories as persistent state that evolves over time through automated updates and enrichment, while maintaining human oversight and traceability.
A new study reveals that standard algorithmic metrics used to evaluate AI counterfactual explanations poorly correlate with human perceptions of explanation quality. The research found weak and dataset-dependent relationships between technical metrics and user judgments, highlighting fundamental limitations in current AI explainability evaluation methods.
A research paper examines how AI-generated visual content is transforming society's relationship with reality and representation, intensifying visual media's dominance in shaping public consciousness. An experiment in Bolzano, Italy revealed people's strong preference for visually striking AI-generated urban development scenarios over practical solutions, highlighting how AI accelerates image commodification and deepens societal alienation.
Researchers propose a new computational concept for modeling the human psyche as an operating system for artificial general intelligence. The approach treats the psyche as a decision-making system that operates in a state space including needs, sensations, and actions to optimize goal achievement while minimizing risks.
Researchers propose a new framework that uses LLMs as code generators rather than per-instance evaluators for high-stakes decision-making, creating interpretable and reproducible AI systems. The approach generates executable decision logic once instead of querying LLMs for each prediction, demonstrated through venture capital founder screening with competitive performance while maintaining full transparency.
Researchers propose a new AI learning architecture inspired by human and animal cognition that integrates observational learning and active behavior learning. The framework includes a meta-control system that switches between learning modes, addressing current limitations in autonomous AI learning.
Researchers developed a resource-efficient framework for compressing large language models using knowledge distillation and chain-of-thought reinforcement learning. The method successfully compressed Qwen 3B to 0.5B while retaining 70-95% of performance across English, Spanish, and coding tasks, making AI models more suitable for resource-constrained deployments.
Researchers developed an information-theoretic framework to explain 'Aha moments' in large language models during reasoning tasks. The study reveals that strong reasoning performance stems from uncertainty externalization rather than specific tokens, decomposing LLM reasoning into procedural information and epistemic verbalization.
Researchers propose a priority graph model to understand conflicts in LLM alignment, revealing that unified stable alignment is challenging due to context-dependent inconsistencies. The study identifies 'priority hacking' as a vulnerability where adversaries can manipulate safety alignments, and suggests runtime verification mechanisms as a potential solution.
Researchers have introduced Prompt Readiness Levels (PRL), a nine-level maturity framework for evaluating and governing AI prompt assets in production environments. The system includes a multidimensional scoring method (PRS) designed to ensure prompt engineering meets operational, safety, and compliance standards across organizations.
Researchers introduce VTC-Bench, a comprehensive benchmark for evaluating multimodal AI models' ability to use visual tools for complex tasks. The benchmark reveals significant limitations in current models, with leading model Gemini-3.0-Pro achieving only 51% accuracy on multi-tool visual reasoning tasks.
