Cloudflare introduces Worker Previews, isolated preview environments for every change made by an agent, allowing for pre-production testing and validation before deployment to production.

Coverage Cat is a licensed insurance brokerage that helps users compare umbrella and home coverage side by side, using AI-guided intake and a licensed brokerage team. The company has launched an API for personal AI agents to shop for umbrella insurance, including support for MCP.

Australia's Online Safety Amendment (Social Media Minimum Age) Act 2024 requires social media platforms to take 'reasonable steps' to prevent under-16s from having accounts, with enforcement being technology-neutral. This has raised concerns about the practical implementation and ambiguity around account deletion.

A new proactive defense against partial speech manipulation via self-embedding steganography is proposed, which can detect and restore partially deepfaked audio without any training. This method embeds a clean speech signal with a compressed representation of itself, enabling post-hoc extraction of reference content.

PaCo-VLA is a new approach to regulate Vision-Language-Action models in contact-rich environments by decoupling high-level semantic reasoning from low-level motor commands. It introduces a passivity-shielded compliance prior that prevents invalid model predictions from bypassing contact physics, enabling safe deployment of foundation models in these domains.

Researchers identified two fundamental gaps in Large Language Models (LLMs) decision-making: an observation-belief gap and a belief-action gap. These gaps lead to systematic vulnerabilities that can be exploited in strategic domains without robust guardrails.

Researchers introduce BEAT-Net, a supervised biomimetic framework for ECG diagnosis using deep learning. It integrates QRS-centered biological tokenization with a hierarchical architecture, achieving comparable diagnostic accuracy to CNN baselines while reducing parameters by 95%. This framework demonstrates an efficient and interpretable alternative to massive pre-training for clinical deployment.

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.

This article evaluates the use of Large Language Models (LLMs) for generating Entity-Relationship (ER) diagrams from natural language requirements. The study found that LLMs perform well in simple scenarios but struggle with complex ones, leading to inconsistencies and failures in representing constraints. This indicates that LLMs are not yet mature for reliable use in complex conceptual database modeling.

Researchers introduced Bayesian Chronicle Agents (BCA), a belief layer that separates what an agent believes from how it speaks. This allows for more control over opinion dynamics in LLM agents, enabling three canonical regimes of opinion dynamics on demand. The BCA layer also makes simulation auditable, surfacing systematic stance biases.

A new benchmarking framework is proposed for evaluating the explanatory quality of open-weight vision-language models in face recognition. The framework focuses on explanation quality, including relevance and faithfulness, and provides a structured explanation format for automated querying and auditing. The authors benchmark several families of open-weight VLMs and highlight the need for explanation quality metrics in face recognition systems.

This paper introduces a novel transfer-learning mechanism for Neural Cellular Automata (NCAs) that injects a pretrained teacher's hidden states into a student model to guide early optimization. This results in superior generalization with a minimal parameter budget. The hidden channels of NCAs decouple feature extraction from uniform classification consensus, allowing for robust, decentralized computational substrates for parameter-efficient transfer learning.

Researchers introduced CIPL, a framework for evaluating privacy leakage in LLM (Large Language Model) agents. CIPL assesses how sensitive information is exposed through various channels, providing a common framework for comparing internal sensitive dependence to externally recoverable leakage across different agent pipelines.

Researchers proposed a method to link source code to scholarly articles in Wikidata, creating a knowledge graph of software heritage. The method harvests and validates publication-to-repository pairs from specific sources, yielding a corpus of 4,397 pairs. The approach enables rule-based reconciliation of software and articles, with a read-only lookup against Wikidata showing 4,182 new software items created.

Researchers introduced SynthDemo-RL, a framework for fine-tuning Vision-Language-Action models using synthetic demonstrations generated by an automated teacher. This approach improves the performance of reinforcement learning with sparse rewards and enables successful task execution without human demonstrations. The framework demonstrates significant improvements on various benchmarks and validates its effectiveness on a physical robot.

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.

VidOmni-Bench is a new benchmark for evaluating fine-grained video understanding in Video Large Language Models (Video-LLMs). It assesses whether models can accurately verify events in video captions, revealing weaknesses in current Video-LLMs that generate hallucinated descriptions and struggle with detecting incorrect event descriptions.

Researchers propose HE-Guardrail, a framework for protecting against jailbreak attacks on encrypted large language model inference. The framework evaluates guardrail mechanisms over encrypted data, allowing servers to control the return of model responses without revealing the input or output.

A study analyzed human-AI and human-human dialogues to investigate the function of morality, politeness, and alignment in cooperative communication. The results suggest that AI systems may simulate cooperative communication without the underlying social architecture, and that mechanisms sustaining human accommodation can run in reverse with AI. This has implications for the design and evaluation of conversational AI systems.

Researchers propose WS-NeRF, a novel neural radiance field for 3D reconstruction and novel view synthesis that can handle image degradation and 3D inconsistency. The method, which uses a mixture-of-experts mechanism and temporal memory, improves blurry radiance field reconstruction quality.