Full-text search across 70962 articles. Combine with topic and date filters; results sorted by relevance.
A developer has released MPEdb, a serverless database designed as a drop-in replacement for SQLite3 with PostgreSQL-like concurrency support, claiming it is faster on most benchmarks and passes 100% of internal unit test…
Alibaba launched Qwen3.8, a foundation model with 2.4 trillion total parameters, focused on coding and professional office tasks, with its API now available on Alibaba's Qwen AI platform. The model is integrated into Ali…
Embabel released version 1.0 of its JVM agent framework for Java and Kotlin developers, reaching general availability in 2026. The framework, built on Spring AI, models agents as typed domain objects and uses Goal-Orient…
Researchers introduced LARA (Lightweight Additive Residual Adaptation), a method that adapts frozen AI models by adding low-rank corrections to the residual stream instead of modifying weights, matching LoRA's performanc…
Researchers propose SAF, a Stable Advantage Fusion framework that combines reinforcement learning with verifiable rewards (RLVR) and on-policy distillation (OPD) for training language models, addressing magnitude and tem…
Researchers from Southeast University introduced a training-free paradigm for attributing AI-generated videos to their source generative models, achieving a Rank-1 accuracy of 20.5% and mean Average Precision of 16.6% on…
Researchers from the Applied Machine Learning Lab propose GenCDSR, a generative framework for cross-domain sequential recommendation that improves average accuracy by 1.5% and reduces inference latency by 85.1% over stat…
Researchers introduced Task-Aware Prompt Rewriter (TAPR), a model that reformulates user prompts into task-optimized prompts to improve downstream LLM performance, trained via reinforcement learning with Group Relative P…
Alibaba released Qwen3.8-Max-Preview, a flagship AI model with 2.4 trillion parameters, claiming it trails only Anthropic's Claude Fable 5 in performance, and its shares rose 5.4% in Hong Kong. The model, available on Al…
Alibaba Group Holding Ltd. released Qwen3.8-Max, a flagship AI model with 2.4 trillion parameters that ranks higher on some benchmarks than Moonshot's Kimi K3, and plans to release its weights for public download next we…
OpenUK, a UK-based non-profit advocating for open technology, has urged the British government to establish a national foundation to hold code, governance, and standards built with public money, warning that the UK is lo…
Onton, a San Francisco-based search and discovery company, released Ontology 1, a neurosymbolic model for complex, conversational, multimodal product search, which achieved a mean precision@10 of 0.630 on a 90-query benc…
Semsearch, an embedding-based search engine bootstrapped from indieblog.page's blog list, launched at semsearch.blog with roughly 6,000 unique blog sites, 154k pages, and 1.4B tokens, ranking results by semantic relevanc…
A 16-year-old developer, Aarav, released Sprocket, an AI agent that creates hardware designs and writes code, and can autonomously purchase parts or SaaS subscriptions from any website. The tool, available via npm as @sp…
The 2026 AI Security & Cybersecurity Expert Bundle, a training package with six courses and 88 hours of instruction, is available for $20.99 (reg. $120) through Aug. 2, with code COURSE30. The bundle, offered by StackSoc…
Open Interpreter, a fork of OpenAI's Codex, now emulates the harnesses of Claude Code and Kimi for low-cost models, enabling AI agents to read, edit, and render Office documents through a single binary called OfficeCLI, …
A new research paper by Choi et al. (2026) finds that multi-agent LLM systems often fail to explore their teammates, sticking with one peer and missing better collaborators. The paper introduces MACE (Multi-Agent Context…
Arclat, a studio focused on developer tools for AI agent teams, has released Relay, an AST-based latency auditor that scans Python AI agent codebases to identify async anti-patterns such as missing awaits, sequential LLM…
Automation bias, the tendency to over-trust automated systems, undermines human oversight of AI agents, turning approval processes into rubber stamps. Research on AI coding agents shows that even when humans see a proble…
Open Interpreter released a fork of Codex that emulates Claude Code and Kimi's harness for cheap models, while OfficeCLI introduced a single binary enabling AI agents to read, edit, and render Word/Excel/PowerPoint files…