Researchers from the University of Maryland and Google DeepMind developed a method to identify AI-generated web content from its structure alone. A new study ported this method to commercial web pages, achieving 98% accuracy in distinguishing between AI-generated and human-written content. The study found that AI-generated content often exhibits a 'tidy, self-announcing' structure, where the main point is repeated in the title, intro, and ending.

The Planetary Prediction Engine (PPE) is an autonomous AI system that enables rapid, customized, expert-level deployment of planetary-scale analytics. PPE synthesizes multimodal datasets on the fly, fuses them with geospatial foundation model embeddings, and searches over task-tailored model architecture families. This reduces the technical barrier to planetary-scale analytics, addressing critical global challenges such as food security and disaster risk.

Researchers analyzed 73,093 Reddit posts about using OpenClaw, an AI agent, to understand how users prioritize values in AI delegation. The study found that values are typically met when users describe the agent's outputs, but often unmet when describing supervision. This highlights the importance of considering the conditions users set around delegation, including cost, access, and oversight, to support human values.

This study investigates the effect of personality-aware fine-tuning on LLMs for social simulation. Researchers fine-tuned two LLMs using a corpus of personality-labelled social media posts and dialogues, but found that fine-tuning did not improve role-playing abilities. The study suggests that the quality and domain alignment of training data are crucial for accurate personality role-playing.

Researchers propose a reinforcement learning framework, PA-RL, that uses artificial potential fields as an action representation for contact-rich robotic manipulation tasks. This approach allows the policy to adapt task strategy and low-level motion generation, reducing learning burden and improving performance. PA-RL achieves 100% evaluation success rate in simulation and demonstrates real-robot deployment feasibility.

A new risk-aware occupancy representation for end-to-end autonomous driving is proposed, improving candidate generation and ranking. This could inform AI agent development in autonomous driving systems.

TinyCeNN-LM proposes a quality-gated conversion framework for replacing attention in pretrained language models with CeNN-inspired cellular-recurrent layers. This approach has implications for the development and deployment of AI agents, as it allows for more efficient and flexible model architectures. The results show that the proposed method maintains representation fidelity and improves performance in certain scenarios.

This research introduces KinshipQA, a procedurally-generated benchmark for multi-hop kinship reasoning that tests large language models' (LLMs) ability to reason across different cultures and reasoning hops. The study found that LLMs struggle to adapt to culturally-marked classification, with a 40.9% accuracy drop compared to biological multi-hop reasoning. The results suggest that LLMs require additional training or rules to overcome cultural biases and improve performance.

Researchers propose a framework for using small language models to translate natural language queries to Kusto Query Language (KQL) for efficient and accurate threat hunting in security operations centers. They evaluate nine models and find that a two-stage approach with a low-cost LLM judge achieves high syntax and semantic accuracy.

Researchers studied Large Language Model (LLM) agents' ability to collaborate with each other under information asymmetry. They developed a fine-tuning-plus-verifier framework to enhance LLM agents' communication and verification capabilities, leading to safer and more interpretable collaboration in AI systems.

TripScore is a benchmark and evaluation framework for language models (LLMs) in travel planning. It uses real user logs and expert judgments to assess LLMs' ability to understand and respond to user inputs, and it compares various fine-tuning approaches for LLMs in this domain.

A new method for improving the evaluation of AI systems, particularly in deployed agents, by using prediction-powered smoothing and validation. The method improves point and interval estimation, especially when labels are few, and provides a new cross-validation score for choosing among estimators.

A new simulation platform, SPAR, is developed for AUV fault recovery using large language models. The platform evaluates LLM-assisted mission management on low-power AUVs, demonstrating improved diagnosis and mitigation of unanticipated faults. The architecture extends recovery from detection to mitigation, and the methodology enables ensemble testing of LLMs for fault recovery.

A new approach to emotion recognition in speechLLMs improves Macro F1 and reduces hallucinations by adapting the model to read the final prompt token's hidden state through a classification head, providing controlled comparison of generative and discriminative inference.

Researchers propose a new approach to retrieval for coding agents, focusing on recovering minimal sufficient evidence to support decision-making. They introduce State-Conditioned Minimal Sufficient Evidence Recovery (SERBench) and evaluate its effectiveness on a dataset of 500 agent states, showing significant improvements over existing methods.

A new framework, agentic robotic additive manufacturing (A-RAM), uses a large language model to generate execution-ready plans for robotic additive manufacturing. The framework integrates pre-execution evaluation of planning decisions and kinematic realization, improving planning outcomes and reducing jerk and extrusion times.

A research paper presents a deployed pipeline for semantic user profiling using LLM inference, reducing the inference budget from user-count to transaction-pattern count, achieving a three-order-of-magnitude reduction in LLM inference targets in a major Japanese bank's deployment.

Researchers studied the web search behaviors of four conversational LLM agents (ChatGPT, Claude, Grok, and DeepSeek) using real-world user interactions and controlled experiments. They found that web-search decisions vary across platforms and models, and that more frequent invocation of web search does not necessarily lead to better response quality. The study highlights the importance of optimizing web search tools for conversational retrieval and raises concerns about attribution and reliability in AI agent responses.

A new vision-language model framework for embodied navigation is proposed, which combines global contextual information with local perception to enhance spatial reasoning and decision-making in long-horizon tasks.

A study evaluates the feasibility of using a large language model (LLM) to score teacher-child interactions in early childhood classrooms. The results show that the LLM can capture some aspects of teacher-child interactions, but struggles with procedural or context-dependent interactions. This suggests that AI-assisted observation may be useful as a preliminary screening tool, but not a replacement for trained observers.