Hacker Papers
- 151
LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale
Hacker News threads
- Jun 2023LLM.int8(): 8-Bit Matrix Multiplication for Transformers at Scale
- Aug 2022LLM.int8(): 8-Bit Matrix Multiplication for Transformers at Scale
- Sep 2022Hot Chips 34: AMD’s Instinct MI200 Architecture
- Feb 2023How to Train Large Models on Many GPUs?
- Mar 2023Show HN: Llama-dl – high-speed download of LLaMA, Facebook's 65B GPT model
- Apr 2023Web LLM – WebGPU Powered Inference of Large Language Models
- May 2023Billion-Scale Approximate Nearest Neighbor Search [pdf]
- May 2023QLoRA: Efficient Finetuning of Quantized LLMs
- Jul 2023Transformer Attention is off by one
- Jun 2024Optimizing AI Inference at Character.ai
- 152
Towards Understanding Grokking: An Effective Theory of Representation Learning
Hacker News threads
- 153
Multiplying Matrices Without Multiplying
Hacker News threads
- Sep 2021Multiplying Matrices Without Multiplying
- Jun 2021Multiplying Matrices Without Multiplying
- Jun 2022Bolt: Faster matrix and vector operations that run on compressed data
- Jun 2022Bolt: Faster matrix and vector operations that run on compressed data
- Nov 2023Show HN: Stella Nera – Maddness Hardware Accelerator
- May 2024Reproducing GPT-2 in llm.c
- 154
Denoising Diffusion Probabilistic Models
Hacker News threads
- 155
Factoring integers with sublinear resources on a superconducting quantum processor
Hacker News threads
- Jan 2023Factoring integers with sublinear resources on a quantum processor
- Jan 2023Factoring integers with sublinear resources on a quantum processor
- Jan 2023Factoring integers with sublinear resources on superconducting quantum processor
- Jan 2023Breaking RSA with a Quantum Computer
- Jan 2023Scott Aaronson weighs in on purported quantum factoring breakthrough
- Jan 2023Factorization (DCQF) of a 48-bit integer using 10 trapped-ion qubits
- Aug 2023
- May 2024Advancing Our Bet on Asymmetric Cryptography
- 156
Locating and Editing Factual Associations in GPT
Hacker News threads
- Feb 2023We Found an Neuron in GPT-2
- Feb 2023ChatGPT is falsely recommending us for a service we don't provide
- May 2023How does data removal from GPT *work
- Aug 2023PMET: Precise Model Editing in a Transformer
- Oct 2023MemGPT – LLMs with self-editing memory for unbounded context
- Nov 2023Is the reversal curse in LLMs real?
- Dec 2023What Is Retrieval-Augmented Generation a.k.a. RAG?
- May 2024Llama3 implemented from scratch
- 157
Graph Pattern Matching in GQL and SQL/PGQ
Hacker News threads
- 158
Validating UTF-8 In Less Than One Instruction Per Byte
Hacker News threads
- Oct 2020Validating UTF-8 in Less Than One Instruction per Byte
- Oct 2020Validating UTF-8 in Less Than One Instruction per Byte
- Sep 2022Ask HN: Why are older processors slower at the same frequency?
- Jan 2024How fast can you validate UTF-8 strings in JavaScript?
- May 2024You probably don't need to validate UTF-8 strings
- 159
Predictive Coding Approximates Backprop along Arbitrary Computation Graphs
Hacker News threads
- May 2021Predictive Coding Approximates Backprop Along Arbitrary Computation Graphs (2020)
- Feb 2021Artificial Neural Nets Finally Yield Clues to How Brains Learn
- Apr 2021Predictive coding has been unified with backpropagation
- Apr 2021Cerebras’ new monster AI chip adds 1.4T transistors
- Apr 2021The Fourier transform is a neural network
- Jun 2023Artificial brains help understand real brains
- Sep 2023What Is an Emotion? William James’s Theory of How Our Bodies Affect Our Feelings
- May 2024Energy, not compute, will be the #1 bottleneck to AI progress – Mark Zuckerberg [video]
- 160
Character-Aware Models Improve Visual Text Rendering
Hacker News threads
- Dec 2022Improving Text Rendering in Image Generation Models
- Dec 2022What ChatGPT Can't Do
- Feb 2023Illusion Diffusion: Optical Illusions Using Stable Diffusion
- Apr 2023A Baby GPT
- Apr 2023Stability AI releases DeepFloyd IF, a powerful text-to-image model
- Oct 2023DALL-E 3: Improving image generation with better captions [pdf]
- Mar 2024Ask HN: Why can't image generation models spell?
- 161
Merkle-CRDTs: Merkle-DAGs meet CRDTs
Hacker News threads
- 162
Vectorized and performance-portable Quicksort
Hacker News threads
- Feb 2024Vectorized and performance-portable Quicksort (2022)
- Jun 2022Vectorized and performance-portable Quicksort
- Jun 2022Vectorized and performance-portable Quicksort
- Jun 2022Apple M2 Die Shot and Architecture Analysis – Big Cost Increase and A15 Based IP
- Dec 2022Sorting with SIMD
- Jan 2023Efficient and performance-portable vector software
- Mar 2024JPEG XL and the Pareto Front
- 163
Efficiently Scaling Transformer Inference
Hacker News threads
- 164
Attacking the DeFi Ecosystem with Flash Loans for Fun and Profit
Hacker News threads
- Jan 2023Attacking the DeFi Ecosystem with Flash Loans for Fun and Profit
- Jan 2021Attacking the DeFi ecosystem with flash loans for fun and profit (2020)
- May 2022Billion-Dollar Crypto Loan Is Easy to Get, but Gone in a Flash
- Feb 2024Two Ethereum DeFi traders just made $120M using a strategy called 'looping'
- 165
Assessing Game Balance with AlphaZero: Exploring Alternative Rule Sets in Chess
Hacker News threads
- Sep 2020Assessing Game Balance with AlphaZero: Exploring Alternative Rule Sets in Chess
- Sep 2020Exploring Alternative Rule Sets in Chess with AlphaZero
- Sep 2020Assessing Game Balance with AlphaZero: Exploring Alternative Rule Sets in Chess
- Dec 2020Ideal Chess – drop chess perfected?
- Aug 2021Fairer Chess: A Reversal of Two Opening Moves in Chess Creates Balance
- Nov 2022Acquisition of chess knowledge in AlphaZero
- Jan 2024I realized chess pieces can be redesigned to be geometric attack directions
- 166
Student of Games: A unified learning algorithm for both perfect and imperfect information games
Hacker News threads
- 167
GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers
Hacker News threads
- May 2023GPTQ Accurate Post-Training Quantization for Generative Pre-Trained Transformers
- Mar 2023LLaMA Text-Generation-Webui
- Mar 2023Large language models are having their Stable Diffusion moment
- Mar 2023Alpaca- An Instruct Tuned Llama 7B. Responses on par with txt-DaVinci-3. Demo up
- Mar 2023LLaMa running at 5 tokens/second on a Pixel 6
- Mar 2023Apple: Transformer architecture optimized for Apple Silicon
- Apr 2023Matrix Calculus: Calculate derivatives of matrices
- May 2023The tiny corp raised $5.1M
- Dec 2023Do large language models need all those layers?
- 168
egg: Fast and Extensible Equality Saturation
Hacker News threads
- Oct 2020egg: Fast and Extensible E-graphs
- Mar 2021Reflections on IDEA vs VS Code
- Jul 2022Ask HN: What are some cool but obscure data structures you know about?
- Jul 2022Ask HN: What are some cool but obscure data structures you know about?
- Dec 2023Atree: A simple and efficient pointer-free tree implementation
- 169
GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding
Hacker News threads
- Jul 2020GShard: Scaling giant models with conditional computation and automatic sharding
- Jul 2020600B Parameters Model – GShard: Scaling Giant Models
- Jul 2020GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding
- Jul 2020OpenAI's GPT-3 may be the biggest thing since Bitcoin
- Dec 2021How to Go Beyond Data Parallelism and Model Parallelism: Starting from GShard
- Dec 2023Google's Gemini: MIT Technology Review
- 170
Multi-Agent Reinforcement Learning is a Sequence Modeling Problem
Hacker News threads
- 171
ZeRO-Offload: Democratizing Billion-Scale Model Training
Hacker News threads
- Feb 2021Zero-Offload: Democratizing Billion-Scale Model Training
- Mar 2021Zero-3 Offload: Scale DL models to trillion parameters without code changes
- Mar 2021Zero-3 Offload: Scale DL models to trillion parameters without code changes
- May 2023Ask HN: Is it just me or GPT-4's quality has significantly deteriorated lately?
- Oct 2023Why async gradient update doesn't get popular in LLM community?
- 172
Addressing "Documentation Debt" in Machine Learning Research: A Retrospective Datasheet for BookCorpus
Hacker News threads
- 173
Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
Hacker News threads
- 174
MIP*=RE
Hacker News threads
- Jan 2020MIP*=Re
- Aug 2021MIP* = Re
- Jan 2020Recent groundbreaking result in complexity theory: MIP*=RE
- Jan 2020'Remarkable' MIP*=RE proof describes how to solve seemingly impossible problem
- Mar 2020Landmark computer science proof cascades through physics and math
- Mar 2020Landmark computer science proof cascades through physics and math
- Apr 2020Mathematicians try to understand how CS proof solves Connes embedding conjecture
- Mar 2023Ask HN: What's new in theoretical CS these days?
- Aug 2023Complexity Theory’s 50-Year Journey to the Limits of Knowledge
- 175
Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding
Hacker News threads
- Aug 2022Imagen: An AI system that creates photorealistic images from input text
- Feb 2023Google Invests Almost $400M in ChatGPT Rival Anthropic
- Mar 2023Visual ChatGPT
- Apr 2023Stability AI releases DeepFloyd IF, a powerful text-to-image model
- May 2023Governance of Superintelligence
- Jul 2023Stability AI releases its latest image-generating model, Stable Diffusion XL 1.0
Data from the Hacker News archive (ODC-By), arXiv and OpenAlex. Not affiliated with Hacker News or Y Combinator.