Models, papers, tools. 39,862 articles with AI-powered sentiment analysis and key takeaways.
FunctionEvolve is a new evolutionary framework that combines expression trees with LLM guidance to recover exact mathematical equations from data, achieving 82.9% accuracy on synthetic benchmarks—significantly outperforming prior symbolic regression methods by making the search process structure-aware rather than structure-blind.
Researchers introduce Stage-Aware Dynamic Weighting (SAW), a novel mechanism for multi-objective reinforcement learning in large language models that addresses the asynchronous nature of reward learning across different objectives. By using coefficient of variation as a real-time informativeness proxy, SAW dynamically reweights objective contributions to improve training efficiency and final performance with minimal computational overhead.
Researchers introduce a new cross-view urban traffic dataset combining synchronized drone and bicycle-mounted camera footage from real intersections. The benchmark enables two computer vision tasks: matching identical objects across street and aerial views, and predicting bird's-eye-view layouts from ground-level cameras with drone supervision.
Researchers introduce Rosetta Memory, an adaptive memory system designed to work seamlessly across different large language models. The system uses profile-conditioned operators to optimize how memory is stored and retrieved, enabling users to switch between models like Claude and GPT without degrading performance.
Researchers introduce IDS-Anta++, an enhanced machine learning framework that defends intrusion detection systems against adversarial attacks through ensemble learning and multi-layer defensive mechanisms. The system achieves over 99% detection accuracy on clean data while demonstrating improved robustness against sophisticated attacks like FGSM and ZOO on standard cybersecurity datasets.
Researchers propose a lightweight 2D-U-Net framework for segmenting abdominal organs in 3D CT scans by combining multi-planar analysis with spatial occurrence maps. The two-stage approach achieves approximately 4% Dice improvement over baseline models and demonstrates practical viability for medical imaging applications.
Researchers present a quantum-classical hybrid system for material classification using polarimetric data, employing quantum SWAP-test circuits to measure similarity between high-dimensional embeddings. The approach achieves competitive accuracy on 23 materials while demonstrating potential for open-set discrimination, positioning it as a practical near-term quantum computing application.
Researchers benchmarked seven uncertainty quantification (UQ) methods on the AION-1 astronomical foundation model for galaxy property prediction, finding that conformal prediction methods—particularly the Locally Valid and Discriminative (LVD) framework—significantly outperform traditional approaches by providing reliable, adaptive confidence intervals. This work establishes best practices for deploying foundation models in scientific inference where uncertainty estimates are as critical as point predictions.
Researchers propose a new governance framework addressing how AI systems can gradually disempower human culture by shaping values and preferences—a threat they argue existing AI policy largely ignores. The Cultural Pluralistic Governance Framework combines cultural influence metrics, democratic assemblies, and deployment standards to prevent "memetic capture" while emphasizing that monocultural AI governance itself accelerates the disempowerment it aims to prevent.
Researchers introduce ACUTE, a protocol that uses language model activations to improve confidence calibration and trustworthiness across multiple LLM tasks. The approach balances calibration accuracy with informativeness through a new EURO metric, addressing the persistent problem of overconfident AI systems.
A researcher demonstrates that AI-paired software engineering, combined with executable specifications and parallel implementations as safeguards, enabled a single developer to port a vector illustration application across five platforms (Rust, Swift, OCaml, Python, browser) in 120 hours. This approach revives N-version programming, a 1980s technique previously abandoned due to cost, making it economically viable by leveraging AI assistance.
Researchers applied process mining techniques to red team attack logs against large language models, revealing that standard attack success rate metrics mask critical differences in how models defend themselves. GPT-OSS 120B exhibits a near-absorbing refusal state, while Llama 3.3 70B shows multiple escape routes from refusal, with mutator effectiveness varying significantly across models.
Researchers introduce an agent-guided multi-fidelity machine learning framework that corrects numerical instabilities in GW-Bethe-Salpeter calculations for simulating electronic and optical properties of strained MoS2-WS2 bilayers. The approach uses confidence-weighted structural agents and Gaussian process corrections to improve accuracy of quasiparticle gaps and exciton binding energies while preserving physical strain dependence.
Researchers tested whether large language models assigned distinct personas could simulate a live concert audience experience through real-time chat during K-pop video playback. While persona-conditioned LLM agents produced more natural and higher-quality chat messages than baseline models, the study found no measurable improvement in user engagement, social connectedness, or emotional response, suggesting that algorithmic personas alone cannot replicate the cultural and social depth of authentic fandom experiences.
Researchers present a cost-aware method for optimizing speculative execution in LLM-agent workflows, addressing the challenge of reducing idle time while managing per-token billing costs. The approach combines five design decisions—including predictive execution, dual-rate pricing, Bayesian probability estimation, and a configurable latency-cost tradeoff—with safeguards ensuring only side-effect-free operations proceed speculatively.
Researchers introduce ClinicalBr, the first bilingual clinical benchmark using 2,892 real Brazilian Portuguese-English case reports to evaluate large language models. The study reveals that English-language advantages in clinical AI are task-dependent, with Portuguese performing comparably in differential diagnosis, exam recommendations, and treatment planning.
Researchers propose a novel defense mechanism called model multiplicity to detect poisoning attacks in distributed small language model training on edge devices. Instead of maintaining a single global model, the system trains multiple independent models on different device subsets, using divergence between them to identify adversarial behavior—outperforming traditional single-model defenses.
Researchers introduce FineSightBench, a benchmark testing vision-language models' ability to perceive and reason about fine-grained visual details at pixel scales of 4-48px. The study reveals that VLMs' visual perception saturates around 12px while reasoning capabilities remain limited even at larger scales, exposing fundamental deficiencies in current multimodal AI systems.
Researchers propose 'instrumented data' as a new paradigm for scientific machine learning, where each data point carries its mechanistic model, uncertainty estimates, and executable counterfactuals. This approach bridges observational data and synthetic data by creating sensor-backed simulations with explicit parameters and causal intervention capabilities, with applications across computational biology, climate modeling, materials science, and medical imaging.
Researchers discovered that thirteen different vision neural networks, despite being trained for distinct tasks (classification, contrast learning, image-text matching), converge on the same sixteen-dimensional geometric structure called the 'cross-architecture substrate.' This invariant structure persists across multiple visual domains and survives calibration testing, suggesting a universal representational principle in modern vision encoders that could enable new transfer learning and distillation techniques.
Researchers improved a deep learning framework for 3D oral reconstruction by introducing Hungarian matching and Repulsion Loss to achieve more uniform vertex distribution across predicted dental models. While numerical accuracy decreased from 77.49% to 68.02%, the trade-off eliminates vertex clustering in sparse regions, producing more clinically useful reconstructions from intraoral images.
Larch is a new optimization framework that improves the efficiency of semantic SQL queries by reducing token usage and computational costs when processing unstructured data with Large Language Models. The framework uses two approaches—reinforcement learning and supervised learning—to optimize the order of filter evaluation, achieving 3x-19x token cost reductions compared to existing solutions.
Researchers present a training-free Video RAG (Retrieval-Augmented Generation) system that decouples semantic retrieval from logical reasoning to improve cross-lingual video comprehension and reduce hallucinations. The two-stage pipeline uses dense retrieval with clean visual data followed by LLM-powered cognitive reranking, achieving strong precision in information retrieval and persona-conditioned generation.
Researchers introduce PartitionSel, a minibatch selection algorithm that optimizes training of large language models on diverse datasets by balancing convergence speed with domain coverage. The method uses partition-matroid constraints and gradient-matching utilities to reduce redundancy across domains while maintaining computational efficiency, demonstrating improvements over existing approaches on Qwen2.5 and Llama-3 benchmarks.
Researchers introduce RecurGuard, a runtime monitoring system that defends reasoning-capable large language models against prompt injection attacks designed to exhaust generation budgets on decoy tasks. The defense detects 99% of such attacks while maintaining minimal false positives, though adaptive adversaries can partially evade detection by using topical rather than semantic attacks.
