Real-time AI-curated news from 97,155+ articles across 50+ sources. Sentiment analysis, importance scoring, and key takeaways — updated every 15 minutes.
Researchers introduce LoRe, a training-free optimization method that dynamically routes computational resources to high-priority interactions in iterative graph solvers, achieving 8× speedup and 12× memory reduction on combinatorial optimization problems while maintaining solution quality.
AIRGuard is a runtime security framework that protects AI agents from authority confusion attacks, where attackers manipulate untrusted context to misuse authorized tool access. The system reduces attack success rates from 36.3% to 5.5% while maintaining 76% of benign functionality, outperforming existing defense mechanisms by enforcing least-privilege authorization at execution time.
MiraBench introduces a new evaluation framework for robotic world models that prioritizes action-conditioned reliability over visual fidelity. The benchmark reveals that current AI models struggle to faithfully follow commanded actions and exhibit persistent optimism bias when predicting outcomes of failure-inducing actions.
Researchers establish a mathematical framework connecting neural network training to Hamilton-Jacobi partial differential equations, showing that gradient descent searches through solutions to viscous PDEs. This theoretical unification applies across major architectures including residual networks and transformers, with implications for understanding generalization, adversarial robustness, and interpretability.
Researchers introduce Conf-Gen, a framework that extends conformal prediction—a formal uncertainty quantification method—to generative AI models like LLMs and image generators. The work bridges a gap between established machine learning safety techniques and modern unsupervised AI systems, enabling confidence guarantees on generative outputs across multiple domains.
Researchers propose Cross-Model Entropy (CME), a label-free reward signal for reinforcement learning that uses a separate verifier model's likelihood assessment instead of human labels or self-referential signals. The method successfully extends RL post-training to open-ended instruction following across multiple model families, achieving win rates of 52.5-71.4% in head-to-head comparisons.
A physicist supervised Claude AI models over 12 days to build CLAX-PT, a physics simulation module, documenting how AI agents struggle with architectural redesign and distinguishing symptom-fixes from root-cause solutions. The study reveals that supervision design and human domain expertise, rather than model capability alone, determine whether AI-generated scientific code produces trustworthy results.
Researchers present an empirical study examining whether Large Language Model agents with tool-calling capabilities produce consistent outputs when given identical inputs across multiple invocations. The study expands beyond prior ReAct-style research to measure behavioral reproducibility in structured tool-calling interfaces, revealing a fundamental reliability gap that could impact production deployment of LLM agents.
Researchers introduce HDPO, a method that uses hallucination detectors to guide iterative refinement of AI-generated clinical summaries, reducing factual errors by up to 48% in large language models. The approach combines inference-time detection with preference learning for model finetuning, demonstrating significant improvements in factual accuracy while maintaining summary quality for healthcare applications.
Researchers introduce BeliefTrack, a benchmark for evaluating how large language models manage contextual information over long interactions—deciding when to update beliefs, preserve state, or ignore noise. The study reveals vanilla LLMs fail significantly at this task, while reinforcement learning with belief-state rewards reduces failures by 71% on average.
Researchers introduce PokerSkill, a framework that enables large language models to play expert-level poker without training or computational solvers by combining rule-based poker skills with LLM reasoning. The approach achieves competitive performance against state-of-the-art GTO benchmarks, reducing losses by 49-61% compared to standard LLM prompting and outperforming established poker bots.
Researchers introduce Metacognitive Memory Policy Optimization (MMPO), a novel training method that improves how AI language model agents manage memory across long-horizon tasks. The approach uses Belief Entropy—a self-supervised metric measuring uncertainty about task state—to provide fine-grained supervision during memory summarization, enabling agents to maintain 97.1% performance even with 1.75M-token contexts.
Researchers introduce BioRefusalAudit, a framework using sparse autoencoders to evaluate the structural integrity of language model biosecurity refusals. The study reveals that five tested models fail to cleanly distinguish hazardous from benign biology, with refusals often disappearing under prompt formatting changes or output constraints, and some models refusing based on legality rather than actual biological hazard.
A comprehensive audit of three major AI models reveals that personalized user contexts significantly reshape brand recommendations in commercial AI assistants, with mid-market brands experiencing up to 75% recommendation volatility while category leaders maintain 80% consistency across personas. The study demonstrates that AI recommendation bias is strongly correlated with model architecture and retrieval strategies, with implications for fair evaluation and brand perception measurement.
Researchers introduce GrowLoop, a self-evolving evaluation system that continuously improves how AI models are assessed for human-like conversation quality. By combining human seed annotations with iterative LLM-driven rubric refinement, GrowLoop addresses the challenge that human-likeness criteria are implicit, subjective, and shift as model capabilities advance.
Researchers introduce dynamic symmetry as a design principle for robotics, where robots are optimized for uniform center-of-mass acceleration capabilities rather than just geometric form. The Argus family of spherical robots demonstrates that achieving extreme dynamic isotropy significantly improves trajectory tracking, robustness, and energy efficiency, with a physical 20-leg prototype exhibiting omnidirectional locomotion and resilience to actuator failures.
Researchers have developed AutoformBot, a multi-agent AI system that automatically translates informal mathematics textbooks into machine-verified formal proofs in Lean 4. The team successfully formalized 26 open-access textbooks into a library called Atlas containing over 45,000 declarations and 500,000 lines of verified code, demonstrating that large-scale automated mathematics formalization is now economically viable.
Researchers introduced Compass, an LLM agent framework that extracts marine lead data from 230,000+ academic papers without fine-tuning, successfully creating the largest integrated marine lead database with 3,751 previously uncatalogued records and 92% accuracy. The expert-guided approach demonstrates how domain-specific knowledge can overcome LLM hallucinations in high-stakes scientific applications.
Researchers introduce Moment-KV, a momentum-based compression technique that optimizes Key-Value cache usage during LLM decoding phases. The method improves long-generation task performance by 2.3-3.2% while maintaining latency by dynamically tracking token importance through temporal attention patterns rather than static heuristics.
Researchers introduce PRAIB, a benchmark framework that evaluates how Large Language Models perform peer review compared to human reviewers. Analysis of 11,000 LLM-generated reviews across major AI conferences reveals significant behavioral divergences: LLM ratings show less variability, positive bias, overconfidence, and frequently miss atomic weaknesses that human reviewers catch.
OmniRetrieval is a new framework that enables unified retrieval across heterogeneous knowledge sources—including unstructured text, relational databases, knowledge graphs, and property graphs—by translating natural language queries into source-native queries rather than forcing all data into a homogenized format. The system demonstrates superior performance compared to single-source retrievers across 13 datasets and 309 knowledge bases, positioning it as a general-purpose interface that preserves the structural advantages of each knowledge source.
Researchers introduce CFGzip, a token space compression technique that dramatically accelerates constrained decoding for large language models using context-free grammars. The method achieves up to 100x latency reduction and 7.5x total speedup, making complex grammar-constrained generation feasible at scale.
Researchers introduce polynomial representations as a quantitative measure of neural network simplicity, demonstrating that the effective degree of these representations predicts generalization better than existing metrics. The approach yields a differentiable regularizer that improves performance across image classification, text tasks, vision-language models, and reinforcement learning.
Researchers introduce AgentDoG 1.5, a lightweight AI safety framework designed to protect open-world agents like OpenClaw from emerging security risks. The framework uses only ~1k training samples to create efficient models (0.8B-8B parameters) that match closed-source alternatives while reducing deployment overhead by 100x, with all resources released openly.
Researchers introduce DOMINO, a framework that synthesizes domain-specific training data for large language models by learning from reference examples rather than explicit domain descriptions. The approach combines prompt tuning with contrastive learning to generate diverse, high-quality synthetic data without manual prompt engineering, improving coding task performance by up to 4.63%.
