HNHacker News
TopNewBestAskShowJobs

E-Reverance

808 karma · joined November 11, 2020

submissionscomments

Chess, but it looks like doodles on paper

twitter.com·2 pts·E-Reverance·
0

Self-Play Pretraining with Zero Data

arxiv.org·4 pts·E-Reverance·
0

Learning to Discover Interesting Mathematics

arxiv.org·1 pts·E-Reverance·
0

Log-Depth Recurrent Language Modeling

arxiv.org·1 pts·E-Reverance·
0

Realtime Pathtracing in Octane

twitter.com·1 pts·E-Reverance·
0

94% on AIME with 1B Params

paradigma.inc·2 pts·E-Reverance·
1

86" 3D Display

lookingglassfactory.com·3 pts·E-Reverance·
0

Fluid Simulation in One Tweet

davidar.io·1 pts·E-Reverance·
0

Solving rubiks cubes "without search"

arxiv.org·4 pts·E-Reverance·
0

Jev might be Qwen finetune

nitter.click·2 pts·E-Reverance·
0

DeepSeek's Parent Company

high-flyer.cn·1 pts·E-Reverance·
0

Replicating Jev with Qwen-2.5

nitter.perennialte.ch·3 pts·E-Reverance·
0

Dart: Denoising Autoregressive Transformer

arxiv.org·1 pts·E-Reverance·
0

Discrete Beckmann Transport Models for One-Step Language Modeling and Reasoning

arxiv.org·1 pts·E-Reverance·
0

Grid-Free Monte Carlo for Time-Dependent Diffusion

arxiv.org·2 pts·E-Reverance·
0

An Algorithm for Proving Combinatorial Infeasibility

arxiv.org·2 pts·E-Reverance·
0

Thinking with Looped Flows

arxiv.org·9 pts·E-Reverance·
1

Revisiting Spectral Representations in Generative Diffusion Models

arxiv.org·2 pts·E-Reverance·
0

Semi-Implicit Pairwise Descent for Nonlocal Continuum Mechanics

arxiv.org·2 pts·E-Reverance·
0

Kalman Delta Networks

arxiv.org·3 pts·E-Reverance·
0

Efficient Exploration Is Enough

arxiv.org·1 pts·E-Reverance·
0

Astra replicates Adobe Lightroom

xcancel.com·3 pts·E-Reverance·
12

The Foundations of Modern AI: Generalization, Data Selection, and Epiplexity

youtube.com·2 pts·E-Reverance·
0

Diffusion as a Training Curriculum for Timestep-Free Iterative Reasoning

arxiv.org·3 pts·E-Reverance·
0

Unlocking Lossless Speedups in LLMs via Discrete Diffusion (5000 Tk/S)

s-sahoo.com·3 pts·E-Reverance·
0

Matching Features, Not Tokens: Energy-Based Fine-Tuning of Language Models

energy-based-fine-tuning.github.io·1 pts·E-Reverance·
0

Language Models Can Control Their Own Attention

arxiv.org·2 pts·E-Reverance·
0

Inverse Rendering for Modeling with Line Primitives (Good for Fur)

kenji-tojo.github.io·1 pts·E-Reverance·
0

Reward Guided Speculative Decoding

arxiv.org·1 pts·E-Reverance·
0

Renderformer V2 (Full neural rendering)

renderformer.github.io·3 pts·E-Reverance·
0
Page 1 of 6Next →