Science – Page 80

Beyond Connections: Reclaiming Knowledge from Fragmented Data

11.01.2026 by qfx

Orion-RAG addresses the challenge of knowledge retrieval by constructing hierarchical navigation paths from fragmented text-a process facilitated by dual-layer labeling-and then leveraging these paths to guide a hybrid search integrating both sparse and dense retrieval methods, ultimately aiming for answers that are not only accurate but also demonstrably interpretable as the system ages.

A new retrieval framework, Orion-RAG, offers a path to effective knowledge access even when data isn’t neatly organized into traditional graphs.

The Persuasion Game: Can You Spot the AI?

11.01.2026 by qfx

New research explores whether persuasive text crafted by artificial intelligence is becoming indistinguishable from human writing.

Seeing is Understanding: AI Gains Ground in Farm Diagnostics

11.01.2026 by qfx

$A two-stage framework leverages deep language models to synthesize reasoning data for agricultural disease identification, first transforming question-answer pairs into reasoning exemplars via a generative-filtering process-utilizing [latex]\tau = 8.0/10.0[/latex] as a threshold-and then employing a group relative policy optimization (GRPO) approach, incorporating a five-tier fuzzy matching system to address linguistic variation and a three-component reward function-focused on format, answer accuracy, and reasoning quality-to achieve stable learning with a 3B parameter model.$

New research demonstrates a significant step towards building AI systems that can accurately reason about agricultural challenges directly from images and natural language.

Beyond Pixels: Giving Self-Driving Cars Common Sense

11.01.2026 by qfx

Augmenting deep learning with automated reasoning capabilities significantly improves the reliability of object detection in autonomous vehicles, especially when faced with ambiguous or complex scenarios.

Can Machines Truly Write Like Humans?

11.01.2026 by qfx

The study introduces MAGA-Bench, a dataset comprising 20 distinct domains and leveraging 12 generative models to rigorously evaluate the performance of detection algorithms, with decoding strategies detailed elsewhere in the supplementary material (§E).

A new benchmark and training framework aims to improve the accuracy of detectors that distinguish between text written by people and generated by artificial intelligence.

AI Pilots: Training Agents for Safer Skies

10.01.2026 by qfx

$The transformer network’s performance, quantified by [latex]\lambda\lambda[/latex]-returns and entropy during training, demonstrates that configurations with 11, 22, and 33 encoder layers-each tested with three random seeds-converge with predictable variance, as indicated by the smoothing achieved using an exponential moving average with [latex]\alpha = 0.05[/latex].$

A new reinforcement learning framework, powered by transformer networks, demonstrates robust aircraft separation in both structured and unpredictable airspace environments.

Taming Time’s Uncertainty with Deep Generative Models

10.01.2026 by qfx

A new framework leverages stochastic dynamics to capture and quantify uncertainty in complex, evolving data streams.

Smart Networks: AI Learns to Choose the Best Connection

10.01.2026 by qfx

Heterogeneous wireless networks necessitate intelligent access selection mechanisms, as systems adapt to fluctuating conditions rather than resisting inevitable change.

A new approach leverages deep reinforcement learning to intelligently select the optimal wireless network in complex, heterogeneous environments.

Unlocking Reasoning: How to Train Smarter AI with Less Data

10.01.2026 by qfx

A new framework, Miner, boosts the reasoning capabilities of large language models by cleverly repurposing existing data and turning uncertainty into a powerful learning signal.

Making AI Decisions Understandable: A New Approach to Transparency

10.01.2026 by qfx

Following fine-tuning, the system demonstrates an enhanced capability for both expert explanation and prediction, suggesting improved performance and understanding of underlying principles.

Researchers have developed a novel framework for fine-tuning large language models to generate clear, accurate explanations for complex financial decisions, building trust in AI-driven systems.