Search

Search

Find paper breakdowns, models, benchmarks, AI tools, services and guides across the site.

14 results for “open weights”

Paper breakdowns

4 paper breakdowns

LoRA: fine-tuning large models by training small low-rank updatesA technical reading of the 2021 paper that freezes pre-trained weights and learns low-rank update matrices instead — the maths, why it adds no inference latency, what the authors report against full fine-tuning, and what remains open./research/lora-low-rank-adaptation28 Sept 2026
Retrieval-Augmented Generation: pairing a language model with a searchable memoryA technical reading of the 2020 paper that introduced RAG models — how a dense retriever and a sequence-to-sequence generator are trained together, the difference between RAG-Sequence and RAG-Token, and what the results do and do not show./research/retrieval-augmented-generation28 Sept 2026
Chain-of-Thought Prompting: eliciting step-by-step reasoning with a few examplesA technical reading of the 2022 paper showing that few-shot examples with written-out reasoning steps improve large language models on arithmetic, commonsense and symbolic tasks — and why the effect appeared only at large scale./research/chain-of-thought-prompting28 Sept 2026
Attention Is All You Need: the architecture behind modern language modelsA technical reading of the 2017 paper that replaced recurrence with self-attention — what the Transformer does, why it trains faster, and what the paper did and did not show./research/attention-is-all-you-need28 Sept 2026

Models

8 models

Course sections

2 course sections