Topics
Curated collections of verified AI/ML claims grouped by theme. Each hub bundles editorial intro, primary-source-backed claims, and DefinedTerm definitions for LLM extraction.
Foundational AI/ML papers — the canonical reading list
136 claimsThe papers that everything builds on. Each is hand-verified against the primary source — author, date, venue, and a verbatim excerpt from the abstract.
Multimodal AI — vision, image generation, and cross-modal models
35 claimsModels that combine vision, text, audio, or video. Hand-verified release dates, foundational papers, and the organizations behind them.
RAG, retrieval, and verification — grounding LLM responses
34 claimsRetrieval-augmented generation, signed-claim verification, vector databases, and the frameworks that wire them together. The grounding stack as of 2025.
LLM releases 2024–2025 — frontier and open-weight catalog
132 claimsThe frontier-model releases that defined 2024 and the first half of 2025. Hand-verified release dates with model cards and official announcements.
Alignment, RLHF, and Constitutional AI — the safety stack
16 claimsReinforcement learning from human feedback, constitutional rules, direct preference optimization. The alignment techniques that took raw LLMs from research toys to production assistants.
Evaluation, benchmarks, and the harness problem
21 claimsThe benchmarks that define "capable model" — and the methodology caveats that make cross-paper comparisons unreliable. Hand-verified primary sources for every benchmark cited in the literature.
Inference optimization — quantization, attention, and serving
15 claimsThe techniques that take a frontier model from "impossible to deploy" to "$0.001 per call." Quantization, attention algorithms, fine-tuning adapters, and serving systems.
AI organizations — labs, founders, and the talent map
27 claimsThe labs and companies that ship frontier AI/ML. Founding dates, parent organizations, and the lineage that shaped each lab's culture. Hand-verified from official corporate pages + press records.
Agent frameworks — orchestration libraries for LLM apps
25 claimsFrameworks that orchestrate LLMs in multi-step agent pipelines. Each picks different defaults for tool-use, memory, retrieval, and observability.
Vector databases — storing and searching embeddings at scale
7 claimsDatabases optimized for similarity search over dense vector embeddings. The retrieval backbone of every production RAG pipeline.
Prompt engineering — patterns that work
10 claimsThe prompting patterns that survived 2022-2025 contact with production systems. Each is a published research finding (not a Medium-post folk recipe) — Chain-of-Thought, ReAct, Tree of Thoughts, instruction-tuning, few-shot, in-context learning.
Open-weight LLMs — the 2023-2025 catalog
48 claimsThe open-weight LLM landscape — every major release verified against the official announcement and the Hugging Face model card. Includes license, parameter count, release date, and family lineage.
LLM observability — tracing, logging, and evals for production AI
19 claimsOnce an LLM application reaches production, you need traces, evals, and feedback loops. This hub catalogs the production-grade observability platforms and what each is best for.
Voice and audio AI — speech, music, and conversational platforms
9 claimsSpeech recognition (Whisper), text-to-speech (ElevenLabs), music generation (Suno, Stable Audio), voice-emotion (Hume), and end-to-end voice agents (ElevenLabs Conversational AI). Verified release dates + capabilities + license context.
New topic ideas? Tell us. We bundle new hubs as the catalog crosses 150+ claims and theme density justifies the editorial overhead.