AI Wiki — Knowledge Base

Introducing 𝐒𝐩𝐞𝐜𝐮𝐥𝐚𝐭𝐢𝐯𝐞 𝐃𝐞𝐜𝐨𝐝𝐢𝐧𝐠: 𝐇𝐨𝐰 𝐈𝐭 𝐄𝐯𝐨𝐥𝐯𝐞𝐝, 𝐖𝐡𝐞𝐧 𝐈𝐭 𝐒𝐭𝐚𝐲𝐬 𝐋𝐨𝐬𝐬𝐥𝐞𝐬𝐬, 𝐚𝐧𝐝 𝐖𝐡𝐚𝐭'…

Introducing 𝐒𝐩𝐞𝐜𝐮𝐥𝐚𝐭𝐢𝐯𝐞 𝐃𝐞𝐜𝐨𝐝𝐢𝐧𝐠: 𝐇𝐨𝐰 𝐈𝐭 𝐄𝐯𝐨𝐥𝐯𝐞𝐝, 𝐖𝐡𝐞𝐧 𝐈𝐭 𝐒𝐭𝐚𝐲𝐬 𝐋𝐨𝐬𝐬𝐥𝐞𝐬𝐬, 𝐚𝐧𝐝 𝐖𝐡𝐚𝐭'𝐬 𝐍𝐞𝐱𝐭. An interactive tutorial @Madisonkanna and I built for the NeurIPS Education Track. Every LLM you use generate

tutorialsswyx--x

6 Sept 2026

tutorials

Towards a universal language of concepts: A survey

Training Large Language Models for Small-Molecule Design with Synthetic Task Scaling

TSMini: A Simple Yet Highly Effective Trajectory Similarity Learning Model

Ultrasound-Based Prediction of Cirrhosis Decompensation Using Large-Scale Computer Vision Models

WeakMCN: Multi-task Collaborative Network for Weakly Supervised Referring Expression Comprehension and Segmentation

Who Should Grade My Work? Student Perspectives on Transparent AI-Assisted Writing Assessment in Higher Education

Why We Care About Understanding: Competence through Predictive Compression

Introducing 𝐒𝐩𝐞𝐜𝐮𝐥𝐚𝐭𝐢𝐯𝐞 𝐃𝐞𝐜𝐨𝐝𝐢𝐧𝐠: 𝐇𝐨𝐰 𝐈𝐭 𝐄𝐯𝐨𝐥𝐯𝐞𝐝, 𝐖𝐡𝐞𝐧 𝐈𝐭 𝐒𝐭𝐚𝐲𝐬 𝐋𝐨𝐬𝐬𝐥𝐞𝐬𝐬, 𝐚𝐧𝐝 𝐖𝐡𝐚𝐭'…

Loops and graphs. Sam Altman is now the ultimate advocate for (neuro)symbolic AI and he doesn’t even know it.

Today we're releasing data on models accelerating research at OpenAI. Recursive self-improvement could be the most important contributor to …

Beyond Retrieval: Progressive Latent Memory Evolution for Streaming Video Understanding

Customizing your knowledge base on Amazon Bedrock for large and complex documents using Amazon Textract

Dancing with REEM-C: A robot-to-human physical-social communication study

Data Market Design through Deep Learning

Deploy a multimodal WhatsApp ordering assistant with Amazon Bedrock AgentCore

EPIC: Explicit Posterior Item Conditioning for Semantic ID Diffusion Recommendation

From Nowcasting to Forecasting: Adapting a Reanalysis-Trained

Guide, Not Bind: Why Defeasible Priors Fail in Augmented Lagrangian Causal Discovery

Knowledge Acquisition During Pre-training? Large Language Models Learn Better With Auxiliary Views

Listen to the Latents: Self-Correcting Speech Recognition in Large Audio Language Models Through Hidden-State Interactions

LLMs Learn Better In-Context from Rules than from Examples

OBER+: Continuity-Aware Reporting and Traceable Continuous Improvement in Outcome-Based Education

P-CORE: Self-Supervised Surface Consistency for Point-Based Neural Editing

RuleMem: Active Rule Memory for Long-Term Conversational Agents

Sparse auto-regressive modeling for scene generation from multi-view images

Statistical Feature Augmentation for Anomaly Detection in Dynamic Graphs

The Shape of Time: Video-Token Contrast for Temporal Understanding in VideoLMs

TokenMatch: 3D Mesh Correspondence Transformer with Curvature-Guided Tokenisation

TraveL: Transformer-based Multi-view Path Distributional Representation Learning

Zero-Shot Novel Depth Synthesis Using 3D Foundation Models Scene Representations

Achieving More with Less: A Tensor-Optimization-Powered Ensemble Method

An example of useful knowledge work: I assigned GPT-6 to read through tens of thousands of my emails, my writings, my calendar appointments …

AutoCompass: Accurate Visual Localization on Public Maps by Learning from Weak Labels

Compositional Spectral Prompts for LLM-based Online Time Series Forecasting

Consistency as Regularization for Unsupervised Shadow Removal

Differentiable Electricity-Market Clearing for Gradient-Based Planning

Do Tabular Foundation Models Know Physics? Contamination, Units, and the Deterministic Limit

Embed Quick Sight visuals using Cognito user authentication

EmoStance: Response-Side Affective-Orientation Control for Empathetic Response Generation via Emoji Weak Supervision

From topology learning to graph generation: A unifying perspective

Generative Diffusion Surrogates with Analytical Variance Schedule

GitHub Copilot app for Beginners: Run several agents at once

Induction and Inquiry via Probabilistic Reasoning over Language and Code

Kirin: Animal Motion Generation from In-the-Wild Video

Latent unified smooth Hamiltonians for excited state chemistry

Learning and extrapolating scale-invariant processes

No Data Wasted: A Semi-supervised Generative Model for Incomplete Multi-view Data Integration with Missing Labels

Oracle, will I ever learn? A study of prediction convergence and complementarity across link prediction models