THE INTERNET IS THINKING
Earlier sources: Artificial Intelligence
Artificial intelligence covers systems that learn from data, generate content, perceive their surroundings, or act toward a goal. Follow model research, agent tooling, evaluation, and the infrastructure used to build and run these systems.
SYMTRI organizes observations from public sources into a living map. Connections reflect shared subject matter; they do not establish causation or endorse a source.
Open in the map ↗5181 classified signals in the fourteen-day window. Recent momentum: steady. These counts include signals awaiting readable evidence.
Earlier sources
Up to ten sources with readable evidence, ordered by their feed timestamp. These archive sources can be older than the live fourteen-day window. This list is a sample, not a measure of total activity. Dates distinguish paper publication, repository creation, model/dataset activity, Hacker News posts, and KEV catalog additions. “First archived” records when SYMTRI first stored the signal.
- Hacker News · Posted on Hacker News · First archived
AI has changed my engineering leadership role ↗
Hacker News discussion.
- arXiv · Published · First archived
BiShield-TEE: On the (In-)Security of Unilateral Weight Obfuscation in On-Device TEE-Shielded LLM Partition ↗
On-device deployment of large language models (LLMs) is increasingly adopted due to latency and privacy requirements, but exposes proprietary models to reverse engineering and intellectual property theft. Trusted Execution Environments (TEEs) provide hardware-enforced isolation for on-device inference, yet most TEE-based parameter obfuscation relies on unilateral matrix transformations that obfuscate only output side. We identify a vulnerability in this design, termed the \emph{Solution Determinacy Property (SDP)}: the obfuscation secret is uniquely determined, exposing the structure of the tr…
- Hacker News · Posted on Hacker News · First archived
Even Without AI, the Risk of a Biological Attack Is Terrifying ↗
Hacker News discussion.
- Hacker News · Posted on Hacker News · First archived
Show HN: Reproducible failure drills for AI coding agents ↗
Hacker News discussion.
- Hacker News · Posted on Hacker News · First archived
Some interesting internal documents from Anthropic litigation ↗
Hacker News discussion.
- arXiv · Published · First archived
Federated Art Chase: Privacy Leakage and Data Poisoning in Federated Fine-Tuning ↗
Federated learning (FL) is increasingly used to adapt a pretrained, centrally fine-tuned model to data that remain on user devices, beyond training a model from scratch. In this report, prepared for the 4th Federated Machine Learning International Summer School, we study the privacy and integrity of this approach with a reproducible pipeline. An ImageNet-pretrained MobileNetV4-Conv-Small model is fine-tuned centrally on ten Quick, Draw! classes (97.50% test accuracy) and then adapted with head-only FedAvg over ten non-IID clients, which raises client training accuracy from 90.34% to 91.49% whi…
- Hugging Face · Model / dataset date · First archived
jakeatx/llamampere-evalcache — dataset release ↗
# llamAmpere eval caches for Swift 1.5 (IQ4_XS-M) at SJ-KVaRN 4/4 Long-context benchmarks send the same huge prompts every run, and the model has to read ("prefill") each one before it writes a word. At 100K tokens that is minutes per prompt. This dataset holds the finished result of that reading step for several well-known benchmarks, so your server can load it from disk instead of recomputing it. It is made for the `--eval-cache` option of llamAmpere v0.5.5 (see `tools/eval-cache/README.md` in that release). Every folder was built with llamAmpere commit `c305c699f` (the v0.5.5 release code;
- Hugging Face · Model / dataset date · First archived
serialhex/gemma4-26b-biblical-qlora-gt — model update ↗
# Model Card for gemma4-26b-biblical-qlora-gt This model is a fine-tuned version of [unsloth/gemma-4-26B-A4B-it](https://huggingface.co/unsloth/gemma-4-26B-A4B-it). It has been trained using [TRL](https://github.com/huggingface/trl). ## Quick start ```python from transformers import pipeline question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?" generator = pipeline("text-generation", model="serialhex/gemma4-26b-biblical-qlora-gt", device="cuda") output = generator([{"role": "user", "content": question}], max_n
- Hugging Face · Model / dataset date · First archived
Bharathikannan/brain-tumor-classification — model update ↗
# NeuroLens — brain tumor MRI classifier Selected model of the NeuroLens project (NUS-ISS PRS Practice Module): **Fibonacci-Net** (lr=0.001). - Classes (output order): glioma, meningioma, notumor, pituitary - Input: 224 x 224 RGB, pixels / 255 - Explanation method: Grad-CAM - Test metrics: accuracy 0.972, macro_f1 0.971 - Training data: Kaggle Brain Tumor MRI dataset (masoudnickparvar, version 1) Research prototype — **not a diagnostic device**. Full details in `model_info.json`.
- GitHub · Repository created · First archived
weisgnosis/weisgnosis-agents ↗
Build tool-using AI agents and collaborating teams in Scala 3 and ZIO 2. Typed tools, scoped execution, explicit application control.
Live window recorded . This is the ingestion observation time, not the publication or discovery date of each source.
How to assess evidence about Artificial Intelligence
Check each original source for its methods, release status, limitations, and supporting data. A repository announcement, a preprint, and a published study provide different kinds of evidence. SYMTRI’s subject classification helps discovery; it does not independently validate the source’s claims.
See the collection and counting methodology ↗ · Read the weekly synthesis ↗
Follow a thread
Explore related regions
These navigation suggestions are part of the map’s editorial structure, rather than measured evidence of a connection.