Full-text search across 12627 articles. Combine with topic and date filters; results sorted by relevance.
A developer built Collection Radar, a local AI tool that ingests a wholesale business's existing Excel invoice spreadsheet and converts it into a prioritized collections list. The tool uses TabPFN, a tabular-data model, …
Tencent Zhuque Lab contributed RogueHandoff-20, a 20-scenario benchmark added via GitHub PR to Tencent's AI-Infra-Guard project, which shows that injecting unsafe intent into the transition between agents raises harmful-…
A developer built Inkwell, an open-source tool that converts messy plain-text lecture notes into structured study guides with KaTeX-rendered math and vector PDF export. The tool uses Gemma open-weight models with a STEM-…
A technical deep dive explains how vLLM's PagedAttention and continuous batching address the memory fragmentation and GPU starvation that bottleneck production LLM inference. PagedAttention partitions each sequence's KV …
A developer released Agentic-Local-Prompt-Semantic-Diffuser, a zero-dependency Python CLI tool that statically quantifies semantic drift between two LLM prompts in under 10 seconds. The tool combines term-frequency vecto…
A developer built Harmony AI, a browser-based design concierge for their father's home textile wholesale shop, using Vite, React, and Tailwind CSS with the open-weight Mistral-7B-Instruct-v0.3 model served through the se…
A developer built GlucoPulse, an open-source, privacy-first metabolic digital twin that forecasts post-meal glucose spikes using Prior Labs' TabPFN tabular foundation model and Google Gemma 2 for plain-English coaching. …
Anthropic's Claude Opus 5.5, released September 22, 2026, shows a 19% lower word-distribution divergence from human writing than Opus 5, while OpenAI's GPT-6 Astra's divergence is 8% higher than GPT-5.6 Sol's, according …
Researchers posted a paper on arXiv on 1 October 2026 introducing LoopCD, a training-free contrastive decoding framework for looped Transformers that contrasts the final prediction with an earlier recurrent pass. LoopCD-…
The University of Cambridge became the first university to refuse to sign Turnitin's new licence agreement in July 2026, according to the Cambridge Students' Union, after the proposed Terms of Use indicated student work …
LLVM 23.1.0 was released on August 26, 2026, with almost 1300 patches contributed by Arm teams across architecture support, libraries, tools, performance and ML workloads. The release adds support for the Arm AGI CPU, Ar…
HackerRank's fourth Orchestrate hackathon, which concluded September 12, drew 35,700 registrations and 3,062 participants, up from 12,871 registrations and 1,349 participants in the first May edition. HackerRank reported…
A developer published a census of their homelab, documenting 41 Docker containers and 9 LXC containers spread across four machines, including a single 6 GB RTX 2060 that handles most AI routing decisions. The writeup det…
OpenTrainDNN, an open-source client-side web application from developer MatiwosKebede, runs real backpropagation, Adam optimizer weight updates, and gradient clipping entirely in the browser to visualize neural network t…
Prime Intellect launched Prime Inference, a serverless and reserved-capacity serving platform for frontier open-source models that processed nearly a trillion tokens per day internally before public release. The platform…
A developer built "The Silent Killer," an adversarial evaluation harness on the Kaggle Model Proxy that tests whether frontier LLMs can audit ML pipelines for fatal methodological flaws rather than syntax errors. In the …
A new arXiv paper (2610.00102v1) proposes an uncertainty-aware learning method that preserves individual expert intervals instead of collapsing them to a consensus, separating within-label imprecision from between-expert…
Amazon Mechanical Turk closed permanently on 2026-09-30, after it stopped accepting new requesters on 2026-07-30, leaving teams that used the MTurk public workforce inside SageMaker Ground Truth labeling jobs and Amazon …
Justine Tunney wrote 84 new matrix multiplication kernels for llamafile that make prompt evaluation 30% to 500% faster than llama.cpp on CPU with F16 and Q8_0 weights, according to benchmarks published March 31, 2024. On…
DynamicTune, a Show HN project, reports that closed-form weight surgery can transfer trajectory dynamics from a 4B-parameter Qwen3.5 (d=2560) into a 0.8B student (d=1024) and from GPT-2 XL (d=1600) into GPT-2 small (d=76…