BigCodeBench: Benchmarking Large Language Models on Solving Practical and Challenging Programming Tasks Jun 18 • 43
📈 Scaling Laws with Vocabulary Collection Increase your vocabulary size when you scale up your language model • 5 items • Updated Aug 11 • 6
🧬 RegMix: Data Mixture as Regression Collection Automatic data mixture method for large language model pre-training • 10 items • Updated Jul 26 • 8
Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Multimodal Models Paper • 2409.17146 • Published Sep 25 • 104
Molmo Collection Artifacts for open multimodal language models. • 5 items • Updated 27 days ago • 289
OLMoE Collection Artifacts for open mixture-of-experts language models. • 13 items • Updated 27 days ago • 27
OpenDevin: An Open Platform for AI Software Developers as Generalist Agents Paper • 2407.16741 • Published Jul 23 • 68
BRIGHT: A Realistic and Challenging Benchmark for Reasoning-Intensive Retrieval Paper • 2407.12883 • Published Jul 16 • 8
Scaling Laws with Vocabulary: Larger Models Deserve Larger Vocabularies Paper • 2407.13623 • Published Jul 18 • 53
RegMix: Data Mixture as Regression for Language Model Pre-training Paper • 2407.01492 • Published Jul 1 • 35
BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions Paper • 2406.15877 • Published Jun 22 • 45
SEACrowd: A Multilingual Multimodal Data Hub and Benchmark S Collection SEACrowd is a community movement project aimed at centralizing and standardizing AI resources for Southeast Asian languages, cultures, and/or regions. • 3 items • Updated Jun 18 • 6
DataComp-LM: In search of the next generation of training sets for language models Paper • 2406.11794 • Published Jun 17 • 50
view article Article BigCodeBench: Benchmarking Large Language Models on Solving Practical and Challenging Programming Tasks Jun 18 • 43
SEACrowd: A Multilingual Multimodal Data Hub and Benchmark Suite for Southeast Asian Languages Paper • 2406.10118 • Published Jun 14 • 30
Language models scale reliably with over-training and on downstream tasks Paper • 2403.08540 • Published Mar 13 • 14