Full-text search across 9702 articles. Combine with topic and date filters; results sorted by relevance.
Flathub announced on May 29, 2026, that it will no longer accept LLM-generated submissions and codebases due to overwhelming spam from automated tools like OpenClaw. The volunteer review team, consisting of only three me…
Residents in east London are fighting a datacentre planned for Brick Lane, saying the site should hold homes, not servers. The facility would serve high-frequency trading, not AI, and the government will rule on the plan…
Researchers have developed a fault-tolerant LLM pipeline that autonomously produced a novel physics research paper on altermagnetic piezomagnetism, from corpus analysis to manuscript writing, with minimal human intervent…
A study comparing aligned and refusal-ablated (abliterated) large language models from the Gemma and Qwen families found that abliterated models outperformed aligned models in vulnerability analysis tasks, including patc…
Researchers from Meta AI benchmarked the energy costs of inference for LLaMA models on NVIDIA V100 and A100 GPUs across up to 32 GPUs, finding that inference energy consumption is significant and often overlooked compare…
Researchers introduced agent data injection attacks (ADI), a new category of indirect prompt injection that disguises malicious data as trusted data to trick AI agents into executing unintended actions. The attacks bypas…
Lilian Weng, cofounder of Thinky, published a summary of 35 papers on harness engineering for recursive self-improvement (RSI), arguing that harness improvements will remain essential even as core models evolve. The post…
Manifest Cloud will introduce paid plans starting July 8, 2026, with a Pro plan at $19/month offering unlimited routed requests, while the Free plan retains all features but is limited to 10,000 routed requests per month…
Meta's Muse Video model debuted at third place on arena.ai's Text-to-Video Arena leaderboard with an Elo score of 1459, behind Google's gemini-omni-flash and ByteDance's dreamina-seedance-2.0-720p. The model features nat…
Researchers have developed a $3,000 open-source maskless photolithography stepper that uses a DLP projector and deep learning computer vision to achieve sub-2-micron resolution, enabling low-cost microfabrication of devi…
DeepSpec, a GitHub repository with over 6,378 stars, provides a comprehensive codebase for training and evaluating speculative decoding algorithms. Built in Python, it offers a full-stack platform to improve computationa…
By mid-2026, the agentic AI ecosystem has coalesced around a layered protocol stack—including MCP, A2A, WebMCP, and OSI—that converts the quadratic N×M integration problem into a linear N+M problem. A developer tracking …
A developer built a 2ms real-time AI inference pipeline in .NET for moderating a Discord chat environment with over 50,000 concurrent members. The pipeline abandons generative AI and third-party LLMs in favor of constant…
Researchers found that language models in coding agents linearly encode properties of evolving programs, including correctness, up to 25 steps before edits are made, a phenomenon they call the latent programming horizon.…
Researchers at an undisclosed institution have developed a black-box attack framework called Zombie Agent that can covertly implant persistent payloads into self-evolving LLM agents through indirect exposure to poisoned …
WILDLABS and Arm announced the winners of the WILDLABS Awards 2026 on July 7, 2026, distributing $360,000 across 16 conservation technology projects. The awards support innovations such as AI-powered bioacoustic monitori…
GeoLambda GmbH released Oikoumene v0.3.1, an autonomous agent civilization simulator that models human history from 70,000 years ago to 2100 using JEPA world models and a Maslow-style needs hierarchy. The simulation, bui…
Researchers introduced Zero-Flow Encoders, a flow-inspired framework for representation learning that uses a zero-flow criterion to certify conditional independence and extract sufficient information from data. The metho…
The MiMo-V2.5 model family introduces Hybrid Sliding Window Attention (Hybrid SWA) to reduce KVCache storage to roughly 1/7 of Full Attention, sparse MoE activation to cut per-token compute, and multimodal encoders for c…
Researchers introduced Raven-Agent, the first autonomous trading agent for prediction markets, which achieved the only positive return and positive risk-adjusted return among tested policies in controlled replays over ar…