AI News Archive: July 27, 2026 — Part 21
Sourced from 500+ daily AI sources, scored by relevance.
- Uncensored Image Generator - Imagify
Generate your wildest desire
- Clarivnt
Restore old family photos naturally with AI
- GeoSpy.tech – Find Location from Images
Find where any photo was taken with AI
- AXIS 3D Foundry — Studio Redirect
Engine-ready 3D meshes from text or image
- 1Capture
Double Your Trial-to-Paid Conversion with AI
- HelloShop - AI Shop Builder
Scan any location. See real gaps. Build the right shop.
- QlimbUp Free AI Career Toolkit
Build AI Resumes, cover letters, portfolios & jobs, all free
- Iliad by AXIS Studio
turns raw codebases into canonical, agent-ready artifacts
- ZGI.AI
A self-hostable runtime for business AI agents
- Photo2Video
The AI photo to video generator with every top model
- Car Background AI - Easy-to-use AI Car Photo Editor
AI car photo background removal and replacement, built for dealerships
- Airo Mail - Superpowered Email for Solo Founders
Superpowered email for Solo Founders. Available on iOS / Android, and Web
- ScamCheck Grok
AI-powered scam detection for photos and documents.
- ScamCheck Claude
AI-powered scam detection for photos and documents.
- aipricesearch.com
Retail price comparison shopping companion
- Decentralised Consensus Learning Networks: SME Rotation Without Centralised Reward
Centralised reward signals dominate modern AI learning systems, but they impose a single external definition of correct or valuable knowledge. We present a decentralised, consensus-based multi-agent learning framework in which expertise emerges through peer validation rather than prescribed reward. ...
- Moral Hazard in Multi-Agent Language Models
Cooperation can fail when socially valuable effort is costly, weakly observable, and mainly benefits others. Drawing on Holmström's team moral-hazard model, we introduce the Dialogue Moral Hazard Game, a controlled textual game that operationalizes this hidden-action structure for language agents. I...
- EEGForceFusion: Joint Tokenised-Continuous Representation Learning for Subject-Independent Grasp Force Decoding
Brain-machine interfaces provide a link between neural activity and external devices, enabling restoration of motor function and advancing human-machine interaction using non-invasive electroencephalography (EEG). However, continuous grasp force decoding remains challenging due to complex temporal d...
- Leveling the Playing Field: Temporal Video Segmentation for Individuals with ADHD in Computing Education
Individuals with Attention-Deficit/Hyperactivity Disorder (ADHD) often face significant barriers in computing education. In asynchronous learning environments, instructional videos can impose high extraneous cognitive load, often relying on assumptions about sustained attention and working memory th...
- Mapping the Reddit Bot Ecosystem: Taxonomy and Evolution
Automated agents increasingly participate in online communities, yet their population structure and roles remain poorly understood. Using a dataset of 3,389 identified bots and their full activity histories, we construct a taxonomy of bot "species" on the news aggregation and social media platform R...
- Just Testing, Move Along: Evasion of LLM-based System Log Interpretation by Prompt Injection
Large Language Models (LLMs) are increasingly integrated into Security Operations Center (SOC) workflows, where they support analysts in tasks such as the interpretation of system logs. However, the ability of LLMs to directly process untrusted textual input also introduces new attack surfaces. In p...
- A Cybersecurity MLPS Large Language Model with Multi-Path Retrieval Fusion
The Multi-Level Protection Scheme (MLPS) is a foundational system in China's cybersecurity governance framework. Therefore, accurate analysis and understanding of MLPS requirements are essential. At present, MLPS analysis still relies mainly on manual interpretation of standards and rule-based tools...
- ContainmentBench: Trace-Based Evaluation of Post-Injection Containment in Tool-Using LLM Agents
Tool-using LLM agents process untrusted content, maintain memory, delegate across agents, and invoke side-effecting tools. Existing prompt-injection evaluations typically summarize security with terminal attack or policy outcomes, but equal endpoints can conceal different post-exposure traces and di...
- Development of Vision-Language Model-based GNSS Spoofing Detection for Autonomous Vehicle Navigation
Autonomous vehicles (AVs) depend on Global Navigation Satellite Systems (GNSS) for localization and navigation, making them vulnerable to spoofing attacks that can covertly redirect vehicles or induce unsafe maneuvers. In this paper, we develop the first Vision-Language Model (VLM)-based framework f...
- EXE-Bench: Ranking the Tradeoffs of AI-based Windows Malware Detectors for Real-World Usability
Due to the lack of systematic evaluations, we are not yet able to determine which AI-based Windows malware detector to deploy in production, since existing evaluations (i) differ in terms of data used for both training and testing; (ii) do not consider temporal analysis to showcase whether models wi...
- Verification-Conditioned Use: A Qualitative Study on How Generative AI Reshapes Learning, Autonomy, and Market Entry for Junior Software Developers
Objective: to investigate how the use of generative Artificial Intelligence (AI) tools affects the early stages of a career in software development, from the perspective of the newcomers themselves. Method: thirteen interns and junior developers were interviewed individually, by videoconference. Int...
- Industrial Practice of LLM-Based Test Case Carving and Assertion Generation (Experience Paper)
Enterprise regression testing for microservice systems is often constrained by incomplete or outdated documentation. In practice, QA engineers frequently rely on real execution traffic to reconstruct business scenarios; however, turning raw traffic into replayable regression tests with stable valida...
- RESTOR: Automated Test Oracle Generation for RESTful APIs via Reinforcement Learning
Modern REST API testing faces a critical challenge in defining reliable test oracles, particularly in agile industrial environments where formal specifications (e.g., OpenAPI) are frequently missing or outdated, and historical execution logs are unavailable for newly deployed endpoints. In this pape...
- Four Years of GenAI: How Educators and Industry Adapted Their Assessment Strategies
GenAI's ability to solve a wide range of software engineering tasks is reshaping the software industry, raising the question of what an entry-level software engineer looks like today. In this work, we investigate whether the core skills for entry-level engineers have changed with GenAI and how the a...
- Leveraging Gradient Reversal Loss and Multitask Learning for Datasets-Aware Audio Deepfake Detection
Recent advances in speech synthesis and voice conversion, which pose threats to security and privacy, have underscored the need for deepfake detection technology. Although existing detection systems achieve strong performance on individual datasets, they often fail to generalize across diverse datas...
- Qwen-Audio-3.0-TTS: Freely Controllable and Highly Robust Speech Synthesis with Multi-Stage Training Paradigm
In this report, we present Qwen-Audio-3.0-TTS, a production-oriented speech synthesis system that jointly advances content consistency, speaker similarity, prosodic naturalness, audio quality, controllability, multilingual coverage, efficiency, and robustness. It combines a 12.5~Hz low-frame-rate sp...
- Revisiting Vocos: That Phasiness Business in Time-Frequency Neural Vocoding
Recently, time-frequency neural vocoders have been approaching the state-of-the-art quality of time-domain neural vocoders. Vocos is a notable example due to its efficiency, but its audio quality lags behind the time-domain vocoders and the reasons remain debated. Thus, in this study, we revisit Voc...
- LaRec: Unleashing LLM-based Latent Reasoning for Generative Recommendation
Large Language Models (LLMs) have shown great promise in recommendation due to superior reasoning abilities. However, existing methods mainly rely on explicit Chain-of-Thought (CoT), resulting in verbose reasoning texts and inefficient response times. latent reasoning aims to balance efficiency by t...
- One Graph, Multiple Gains: Single High-Quality Item-Item Graph for Multimodal Recommendation
Multimodal recommendation leverages item multimodal features alongside collaborative signals to capture user preferences. While item-item graphs have become a key building block in advanced models, existing methods typically construct them with noisy similarity edges and limit their role to a single...
- Robust Interpretation of Historical Documents in Knowledge Graphs Through Query Inference and Execution
The emergence of Large Language Models (LLMs) has redefined how users interact with information in digital environments. However, their widespread and often indiscriminate integration has raised significant concerns regarding reliability and trustworthiness issues that are particularly critical when...
- Strategy-Aware Parameter-Efficient Adaptation for LLM-based Auto-Bidding
Advertising bidding has evolved from manual strategies to auto-bidding systems better adapted for large-scale, dynamic auction environments. While recent advances in Large Language Models (LLMs) offer strong reasoning for auto-bidding, existing methods suffer from shallow trajectory-text interaction...
- Unifying Generative Recall and Multi-Objective Ranking in a Single Decoder-Only Sequence
Modern industrial recommendation systems typically separate recall and ranking into two independent stages. Although this cascade supports corpus-level retrieval and fine-grained multi-objective scoring, it causes objective inconsistency, information loss at the candidate hand-off, and redundant use...
- CORE: A Unified Cascaded Ordinal Relevance Estimation Framework for E-commerce Search
Ranking relevance is a fundamental task in e-commerce search, directly affecting ranking quality and consumer experience. Although inherently an ordinal classification problem, it is commonly formulated as conventional multi-class classification, which overlooks the natural order among relevance lev...
- CogRec: Structure-Cognitive Fast-and-Slow Reasoning for Generative Recommendation
Semantic-ID-based generative recommendation represents each item as a hierarchical discrete token sequence and reformulates next-item prediction as constrained sequence generation. Existing methods, however, mainly use Semantic IDs as target sequences to be memorized, leaving the hierarchy, intra-la...
- OxygenREC-v2: Internalizing Discrimination into Generative Recommendation
Generative recommendation unifies retrieval and ranking within a single model by autoregressively decoding semantic identifier (SID) sequences. Yet reliably incorporating behavior signals from clicks, cart additions, and orders remains challenging. Existing approaches either jointly optimize generat...
- Integrating Factual and Normative Industrial Knowledge via Constraint-Aware Graph Attention for Process Plan Recommendation
Integrating heterogeneous industrial knowledge, including factual relations and decision constraints, remains a core challenge in industrial information systems. Machining process planning exemplifies this problem because engineers must select operations by combining material properties, feature cha...
- Secrecy Energy Efficiency for IRS-Assisted Low-Altitude Communications: A D3QN-PER Based Approach
To address the security and energy efficiency challenges in low-altitude economy (LAE) wireless communications, we develop a secure synergistic network integrating unmanned aerial vehicle (UAV) and intelligent reflecting surface (IRS), with an emphasis on maximizing secrecy energy efficiency (SEE) f...
- ConAlign: Conditional Alignment Framework for Balancing Biased and Unbiased Recommendation
Industry recommender systems trained on observational data suffer from various biases that create filter bubbles, causing user interests to collapse into narrow categories and severely degrading long-term engagement. While utilizing unbiased uniform data for debiasing has shown promise, existing met...
- SpecFormer: Mitigating Embedding and Attention Collapse via Spectral-Aware Transformer for Recommendation
Transformer architectures have achieved remarkable success across diverse domains; however, directly applying their standard self-attention mechanism to recommendation often yields suboptimal performance, sometimes even trailing behind well-designed simple recommendation models. In this paper, we re...
- Mosaic: A Fleet of User Embedding Specialists for Recommendation at Meta
User representation is one of the highest-leverage modeling problems in industrial recommendation systems: a single advancement in how users are encoded can propagate across retrieval, ranking, and integrity tasks at platform scale. Prior industrial user representation work builds either a single us...
- B-SMART-Former: An Explainable Transformer-Based Deep Learning Model for Predicting Drug-Drug Interactions Between Biotech and Small-Molecule Drugs
Drug-drug interactions between biotech and small-molecule drugs play a critical role in medication safety and therapeutic efficacy. However, most existing computational DDI prediction methods focus primarily on interactions between small-molecule drugs, leaving biotech-small-molecule interactions comparatively underexplored. In this study, we propose B-SMART-Former, an explainable deep learning framework for predicting interaction types between biotech and small-molecule drugs. The proposed framework integrates ChemBERTa embeddings and Morgan molecular fingerprints for small molecules with ProtBERT embeddings for biotech drugs, eliminating the need for similarity-based features while leveraging complementary molecular representations. These multimodal features are processed by a hybrid architecture that combines Transformer-based self-attention, residual convolutional learning, and a multi-layer perceptron classifier to capture both global contextual dependencies and local discriminative patterns. The model is formulated as a multi-class classification task and evaluated using stratified 10-fold cross-validation. To improve model transparency, Integrated Gradients is employed as a post-hoc explainability method to identify the molecular features that contribute most strongly to each prediction. Experimental results demonstrate that B-SMART-Former achieves a micro-averaged AUROC of 0.9978 and an AUPR of 0.9682 while relying solely on intrinsic molecular representations, remaining competitive with similarity-based approaches. The proposed framework offers an effective and explainable solution for biotech-small-molecule DDI prediction and provides a practical foundation for future computational drug interaction studies.
- Multimodal Phenotyping of Myofascial Pain Syndrome Using Rotational Shear Wave Elastography and Clinical Network Analysis
Myofascial pain syndrome (MPS) is characterized by increased muscle stiffness, trigger points, and functional limitations, yet clinical diagnosis remains largely subjective. Shear wave elastography (SWE) provides quantitative assessment of muscle mechanical properties, but its value for identifying biomechanical and clinical phenotypes of MPS is not fully established. This study evaluated whether stiffness parameters derived from multi-angle SWE can reliably characterize upper-trapezius anisotropy, and whether integrating SWE with bioimpedance spectroscopy (BIS), range of motion (ROM), and patient-reported outcomes (PROs) improves differentiation of MPS subgroups. Seventy-one adults completed upper-trapezius SWE, BIS, ROM assessments, and PRO measures. Clinically, 18 were classified as active MPS, 36 as latent, and 17 as normal. Shear wave speed measurements were modeled to estimate longitudinal (uL), transverse (uT), and anisotropy (uE) components. Reliability was examined using intraclass correlation coefficients. Unsupervised clustering and partial-correlation network analysis were applied to biomechanical and clinical variables. uT showed the strongest associations with BIS frequency parameters and ROM measures, indicating sensitivity to fascial composition, and mobility. Multimodal clustering incorporating uT with Fc or ROM identified subgroups with distinct tissue-level and functional characteristics. Network analysis demonstrated a progression in connectivity patterns, shifting from localized mechanical relationships to broader symptom-level coupling involving pain interference, sleep disturbance, emotional distress, and physical function. These findings indicate that SWE-derived stiffness parameters provide reliable, direction-specific quantification of trapezius mechanical properties. Combining SWE with impedance and mobility measures yields physiologically coherent MPS phenotypes that differ in both biomechanical features and clinical network structure, supporting more objective framework for characterizing MPS.
- Diagnosing Rejection Collapse via Uncertainty Decomposition
Standard uncertainty-informed rejection can unexpectedly trigger severe performance collapse, exposing localized vulnerabilities that common machine learning metrics typically do not show. We systematically diagnose this failure dynamic using Levodopa-Induced Dyskinesia prediction in Parkinson's Disease as a proof-of-concept. By training a heterogeneous ML ensemble, decomposing Aleatoric and Epistemic uncertainty and applying unsupervised subgroup discovery, we isolated the precise drivers of these atypical errors. Stratified error analysis revealed two divergent predictive regimes previously hidden by a global evaluation. While the models successfully extracted a predictive signal for one subgroup, the baseline features of a second subgroup lacked discriminative capacity, resulting in a high rate of confident misclassifications. Operating entirely below rejection thresholds, this single subgroup flatlined predictive metrics, driving the collapse of the global rejection curve. Ultimately, we demonstrate that atypical rejection failures stem from subgroup-specific data ambiguity rather than algorithmic deficiencies, making localized uncertainty-aware evaluation a critical methodological requirement prior to real-world deployment.
- Hemoglobin-to-Red Cell Distribution Width Ratio Predicts Three-Year Major Adverse Cardiovascular Events After Percutaneous Coronary Intervention: A Vietnamese Multicenter Cohort Study
Introduction Percutaneous coronary intervention (PCI) is a widely adopted strategy for managing coronary artery disease (CAD), leading to improved survival rates, particularly in developing countries. However, the increased survival of these patients imposes a substantial burden on long-term management, especially at the primary healthcare level. Recently, the hemoglobin-to-red cell distribution width ratio (HRR) has emerged as a potentially valuable prognostic biomarker for post-PCI patients. Despite its accessibility and cost-effectiveness, HRR has not been extensively investigated in resource-limited settings. Aim This study aimed to evaluate the prognostic value of the HRR in predicting 3-year major adverse cardiovascular events (MACE) among patients undergoing PCI who were managed at the primary healthcare level. Methods We conducted a multicenter prospective cohort study in Vietnam. A total of 626 post-PCI patients were ultimately included in the final analysis. The study commenced in October 2019 and concluded in October 2025. The association between HRR and 3-year MACE was evaluated using Cox proportional hazards regression models. Results The MACE incidence decreased progressively across the ascending HRR quartiles (p < 0.001). According to the unadjusted Cox model, each unit increase in HRR was associated with a lower risk of MACE (HR = 0.756; 95% CI, 0.703-0.814; p < 0.001). This association persisted after adjustment for age, sex, comorbidities (Model I: HR = 0.810; 95%CI: 0.750-0.878; p < 0.001). HRR outperformed its individual components, hemoglobin and red cell distribution width. Subgroup analyses confirmed the consistency of the association across clinically relevant strata. Calibration and decision-curve analyses further suggested acceptable risk estimation and potential clinical utility of HRR for 3-year MACE risk stratification. In the discriminative analysis for predicting 3-year MACE, the HRR had the highest area under the curve outperforming other inflammation-based indices. Conclusion A lower HRR was independently associated with a greater 3-year MACE risk in post-PCI patients. HRR outperforms commonly used leukocyte- and platelet-derived indices, highlighting its potential utility as a simple, cost-effective prognostic marker in resource-constrained healthcare settings.
- Multimorbidity and Frailty in Older Adults: A Comparative Study of Care Models and Policy Strategies in Spain, Italy, Chile, and Colombia
Background Population aging is rapidly increasing the burden of multimorbidity and frailty, yet there is limited evidence on how healthy aging policies are implemented across different demographic and health system contexts. Methods We conducted a mixed-methods comparative case study in Spain, Italy, Chile, and Colombia, integrating a Targeted review of policy documents and relevant literature, quantitative indicators, and expert interviews. Results Chile and Colombia showed faster aging dynamics (compound annual growth rates of 4.18% and 4.21%) than Italy (1.96%) and Spain (1.39%), with estimated doubling times of 17 versus 36-50 years. All four countries have established policy frameworks; implementation remains uneven. Key barriers included fragmented services, limited community care, workforce constraints, and misaligned financing. Conclusions The central challenge lies not in the absence of policy frameworks but in their effective implementation. Strengthening primary care, embedding frailty assessment, and improving health-social care integration are key priorities for systems facing rapid demographic transition.