THE INTERNET IS THINKING
Earlier sources: Language models
Language models learn patterns in text and other representations to predict or generate sequences. Follow model releases, training methods, reasoning, evaluation, and deployment research; a benchmark result is specific to its test conditions.
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 ↗440 classified signals in the fourteen-day window. 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
Build and Train a 25M Parameter LLM from Scratch on Your CPU ↗
Hacker News discussion.
- GitHub · Repository created · First archived
rasidi3112/pondok-ai ↗
One-command private AI stack: Ollama + Open WebUI + Qdrant + an OpenAI-compatible API gateway, with hardware-aware model recommendations. Runs on your machine via Docker Compose.
- Hacker News · Posted on Hacker News · First archived
3JSBench: A benchmark for LLM-generated 3D objects ↗
Hacker News discussion.
- Hacker News · Posted on Hacker News · First archived
Show HN: Anvil, offline LLM chat and SDXL image generation on iPhone ↗
Hacker News discussion.
- Hacker News · Posted on Hacker News · First archived
Show HN: LLM Motion Graphics Benchmark ↗
Given that agents are getting good at generating motion graphics I wanted to understand what the differences are in cost and time using say Opus 5.5 and the cheaper models. Let me know if you want other LLMs added and feel free to click around the rest of the website. All videos are editable in place and aim to give engineers a simple way of discovering and remixing motion graphics without spending too much on compute.
- Hacker News · Posted on Hacker News · First archived
Ontological Language Model (OLM) + LLM: Deterministic AI Advancement ↗
This was recently announced by Dr. Berkan from New York. LLMs meeting ontologies, quite exciting to see such new advancement. OLMApp https://olmapp.ai/more.html
- arXiv · Published · First archived
One Block, Multiple Depths: Recurrent Vision Transformers with Depth-Programmed Experts ↗
In this work, we show that a single Transformer block, applied recurrently, can match the accuracy of a full-depth vision encoder at comparable inference FLOPs without intermediate feature distillation. reViT restores depth-specific transformations by representing the FFN at each recurrent depth as a convex combination of a small shared expert bank. A continuous normalized-depth coordinate programs this mixture, defining a resampleable trajectory through FFN parameter space. We evaluate this design in two regimes: supervised ImageNet-1k training and distillation from a DINOv2 teacher. Across b…
- arXiv · Published · First archived
SpaceCast-Bench: Evaluating Predictive Spatial Reasoning in Vision-Language Models ↗
Existing spatial reasoning benchmarks mainly test spatial perception: reading off relations already visible in the input. Yet real-world spatial intelligence demands predictive spatial reasoning: constructing a scene from observations, anticipating how an intervention changes it, and reasoning about the unseen outcome. We introduce SpaceCast-Bench, the first benchmark to directly and diagnostically evaluate this capability. Built around an observe-transform-infer framework, its 3,862 questions from 182 real-world scenes span 16 task types at three levels: static perception, local prediction, a…
- arXiv · Published · First archived
Could LLM Watermark Detection be Public? ↗
Watermarking large language models is popular for tracing chatbot and agentic outputs, yet detectors remain unreleased since exposing them could let attackers do targeted edits with the detector's feedback. However, watermarks are already vulnerable to uninformed tampering attacks. We thus first quantify whether a public detector would be an additional liability in a deployment setting at varying levels of access, from token-level scores to a binary verdict. Second, we introduce a split-key public-private watermarking method that exposes one key through a public detector while keeping the othe…
- Hacker News · Posted on Hacker News · First archived
Show HN: Teaching a 30ms local model, play browser games with LLM written rules ↗
Hacker News discussion.
Live window recorded . This is the ingestion observation time, not the publication or discovery date of each source.
How to assess evidence about Language models
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 ↗
Related threads
Explore related regions
These navigation suggestions are part of the map’s editorial structure, rather than measured evidence of a connection.