VulcanAMI Might Help
I open-sourced a large AI platform I built solo, working 16 hours a day, at my kitchen table, fueled by an inordinate degree of compulsio...
Text understanding and language tasks
I open-sourced a large AI platform I built solo, working 16 hours a day, at my kitchen table, fueled by an inordinate degree of compulsio...
I built an experimental UI and visualization layer around Meta’s open brain-response model just to see whether this stuff actually works ...
The BDH (Dragon Hatchling) paper (arXiv:2509.26507) describes a Hebbian synaptic plasticity mechanism where model weights update during i...
Abstract page for arXiv paper 2603.20510: Grounded Chess Reasoning in Language Models via Master Distillation
Abstract page for arXiv paper 2603.20470: DiffGraph: An Automated Agent-driven Model Merging Framework for In-the-Wild Text-to-Image Gene...
Abstract page for arXiv paper 2603.20425: Leveraging Natural Language Processing and Machine Learning for Evidence-Based Food Security Po...
Abstract page for arXiv paper 2603.20270: FactorSmith: Agentic Simulation Generation via Markov Decision Process Decomposition with Plann...
Abstract page for arXiv paper 2603.20213: AgenticGEO: A Self-Evolving Agentic System for Generative Engine Optimization
Abstract page for arXiv paper 2603.20260: ProMAS: Proactive Error Forecasting for Multi-Agent Systems Using Markov Transition Dynamics
We’ve been working on a probabilistic interpretation of causal self-attention where token embeddings are treated as latent variables. In ...
Most discussions around AI safety focus on what models know or whether outputs are correct. But since 2019, I’ve been working on somethin...
Paper: https://arxiv.org/abs/2603.18280 TL;DR: Current alignment evaluation measures concept detection (probing) and refusal (benchmarkin...
A neat blog post by Mayank Pratap Singh with excellent visuals introducing ViTs from the ground up. The post covers: Patch embedding Posi...
I've been building no-magic — a collection of 47 single-file Python implementations of the algorithms behind modern AI. No PyTorch, no Te...
Abstract page for arXiv paper 2510.15520: Discovering Intersectional Bias via Directional Alignment in Face Recognition Embeddings
Abstract page for arXiv paper 2603.17246: On the Cone Effect and Modality Gap in Medical Vision-Language Embeddings
Abstract page for arXiv paper 2509.08625: An upper bound on the silhouette evaluation metric for clustering
Abstract page for arXiv paper 2502.05709: Flow-based Conformal Prediction for Multi-dimensional Time Series
Abstract page for arXiv paper 2603.20048: Structured Latent Dynamics in Wireless CSI via Homomorphic World Models
Abstract page for arXiv paper 2603.20025: Graph-Informed Adversarial Modeling: Infimal Subadditivity of Interpolative Divergences
Abstract page for arXiv paper 2603.19862: IsoCLIP: Decomposing CLIP Projectors for Efficient Intra-modal Alignment
Abstract page for arXiv paper 2603.19840: Explainable cluster analysis: a bagging approach
Abstract page for arXiv paper 2603.19439: Subspace Projection Methods for Fast Spectral Embeddings of Evolving Graphs
Get the latest news, tools, and insights delivered to your inbox.
Daily or weekly digest • Unsubscribe anytime