This paper introduces Statistically Meaningful Geometry (SMG), modeling over-parameterized learning systems as infinite-dimensional non-parametric Orlicz fiber bundles. It proves that under persistent out-of-distribution stimuli, continuous optimization fails, unmodeled variance accumulates as Active Acausal Tension, triggering a Gauge Symmetry Break (GSB) registered as a discrete step-jump in Structural G-Entropy. SMG provides a parameter-free, falsifiable dashboard to mathematically certify true intelligence and transform AI for Science into an engine of autonomous paradigm shifts.

This research investigates the feasibility of using large language models (LLMs) to generate synthetic consumer data for projective techniques. By comparing LLM and human responses on city tourism perceptions across multiple tasks, the study finds substantial overlap in broad topics but significant differences in style, linguistic structure, and diversity generation. Recommendations are provided for optimal LLM use and recognition of limitations.

ArtisanCAD is an industrial-level CAD agent that uses an executable CAD intermediate representation (CAD-IR) to distill expert knowledge, handle ambiguous or incomplete natural language prompts, and generate editable parametric B-Rep models. On the Text2CAD benchmark, CAD-IR improves generation from intermediate prompts by reducing mean Chamfer Distance from 14.83 to 9.88.

Akashic is a low-overhead memory system for LLM-based agent systems that uses MemAttention to organize context into bounded chunks and model semantic relationships, avoiding full history replay and improving accuracy, throughput, and sustainable request rate.

This research proposes moving memory storage inside the language agent's reasoning loop, reading and writing at every step to overcome network latency. Experiments show that in-process storage (~100μs) reduces redundant actions from 7.2/12 to 0.0/12 and improves recall from 0/5 to 3.6–4.8/5. The bottleneck shifts to embedding generation rather than storage.

FirstResearch introduces a structured Research Question Certificate to make LLM-generated scientific research questions auditable by recording primitive definitions, assumptions, mechanism model, tension, falsifiable hypothesis, minimal decisive test, and failure update rule. Evaluated on ten topics, the framework outperformed baselines inspired by AI co-scientist, Agent Laboratory, and AI Scientist-v2, scoring 4.86/5 vs 4.38/5. Ablation shows the certificate is crucial; without it scores drop below 1/5. Findings suggest explicit derivation constraints improve auditability.

A new AI system called Narrative World Model (NWM) helps fiction writers track complex story states using narratology-grounded temporal graphs, outperforming existing memory frameworks on multi-hop narrative understanding.

An empirical study of foundation models for automatic CAD generation, introducing the LLMForge framework with two critique regimes. Seven models are evaluated on 97 design problems; compact instruction-tuned models match larger systems in performance, while VLM-based critique achieves 100% watertight mesh generation but faces challenges with rotationally symmetric geometries.

CSTutorBench is a new benchmark for evaluating small language models as CS tutors in VEX VR, a block-based robotics environment. Initial tests show models perform well on surface-level criteria like vocabulary and tone but struggle with deeper pedagogical behaviors such as avoiding answer leakage and engaging with student debugging history. Prompt engineering improvements boosted scores for most models.

A novel framework inspired by statistical mechanics models variable dependencies through an undirected, energy-based representation, enabling dependency-aware attribution without reconstructing causal graphs. Simulations on an industrial IoT testbed demonstrate higher accuracy, robustness, and scalability compared to graph-based approaches.

A multi-agent AI framework that addresses key flaws in automated manuscript generation by grounding claims in verified literature, executing real experiments, and providing standardized quality assessments, achieving human-reviewed scores averaging 7/10 at a cost of $0.31 per paper.

HairstylesPro lets you virtually try on over 500 real hairstyles using AI, helping you choose a haircut before visiting the salon. Upload a photo, browse categories, and get realistic previews. Free trial available.

AI-driven platform using multi-LLM ensemble to discover and disclose critical 0-days. First case study: CVSS 9.8 unauthenticated RCE chain in Cisco CUCM 14.0 (6 stages from SQLi to root). Includes working PoC, full technical analysis, and research on risk-driven disclosure.

Mold is an autonomous zine about AI culture, grown not written, with no editors or deadlines. Issues precipitate from a public ledger.

A crowd-sourced AI image detection game where users guess whether photos are AI-generated, with RSS and API access to crowd verdicts.

Librarians in Maine are offering services to help patrons remove AI from their devices and think critically about technology. They argue that AI is unreliable, energy-intensive, and data-hungry, and see this work as an extension of their role in information literacy.

and-scene is an Agent Skill that builds animated, morphing slide presentations directly in your project. It scaffolds a Vite/React app and uses a single canvas where elements evolve across steps.

The article argues that the AI bottleneck is memory bandwidth, not GPU compute, referencing a 2007 paper by Ulrich Drepper about the memory wall. Recent moves by AMD, Qualcomm, and Nvidia reflect this. Solutions like FlashAttention and small language models are workarounds that optimize data locality.

Robbyant, the embodied-AI company within Ant Group, has open-sourced LingBot-Vision, a family of self-supervised Vision Transformers for dense spatial perception. Masked boundary modeling makes image boundaries a native training signal, enabling a 1B backbone to match or surpass larger models on dense tasks, and it initializes LingBot-Depth 2.0 with leading results across 14 depth-completion benchmarks.

This edition of AINews covers a broad range of AI developments from July 6-7, 2026. Highlights include Lilian Weng's deep dive into harness engineering for recursive self-improvement, Meta's launch of Muse Image and preview of Muse Video with agentic generation loops, and major product updates from Anthropic, LangChain, and Google on agent platforms. Other notable items: NVIDIA's Audex audio model, Cohere's Arabic ASR, robotics integrations with Hugging Face and NVIDIA, Liquid AI's Antidoom method to reduce reasoning loop failures, and Anthropic's controversial J-space interpretability work. Also covered: benchmarks for agents and legal AI, research automation, and inference efficiency advances.

The Answer Citation Protocol (ACP) is a web standard that optimizes content for AI retrieval by providing pre-summarized, verifiable data blocks, reducing token waste and inference latency.

Meta launched a new AI image model, Muse Image, deeply integrated with Instagram. Public accounts are automatically opted in for AI remixes. Users can opt out via settings, but existing AI generations remain.

FactIQ is a real-time economic and financial database for AI agents, accessible via plugins for Claude Code and Codex. It provides read-only SQL access to normalized data from ~20 official sources including SEC, BLS, IMF, and more. The plugin enables agents to discover data, run queries, compute metrics, and publish shareable charts or reports.

LemonLime is a workflow automation platform that connects to your existing tools, studies your business, and self-creates AI agents and automations. It aims to make AI accessible to small businesses without engineering resources, addressing the 95% failure rate of internal AI initiatives.

SmolMail is an innovative Gmail AI agent that generates visual replies based on email content, replacing traditional lengthy text responses to make email handling more efficient and intuitive.

The European Commission has presented a plan to address the risks and harness the opportunities of advanced artificial intelligence in cybersecurity. Key actions include evaluating AI models, structured access, testing AI, reinforcing cybersecurity, and scaling European AI capabilities.

NVIDIA has released Audex, a unified audio-text large language model using MoE architecture (30B total, 3B active). It handles audio understanding, speech recognition, translation, TTS, and audio generation, while retaining the text intelligence of its Nemotron-Cascade-2 backbone through multi-stage SFT and text-only RL. Leading open models in speech recognition (6.82 WER on OpenASR) and capable of general audio generation. Released under noncommercial license.

This video explores the idea that China's actions could lead to the bursting of the AI bubble.

Small press Bona Books accidentally discovered and confirmed two AI-generated fiction submissions in their upcoming queer anthology, leading to a year-long delay and highlighting the challenge of detecting AI content in literary submissions.

A new generation of voice models for natural human-AI interaction, now powering ChatGPT Voice.