SuperGPQA: Scaling LLM Evaluation across 285 Graduate Disciplines
Paper
•
2502.14739
•
Published
•
92
Table Research Lab
classroom
AI & ML interests
None defined yet.
Recent Activity
View all activity
TAPEX's activity
SivilTaram
authored
a
paper
7 days ago
dreamerdeo
posted
an
update
9 days ago
Post
2753
🚀 Excited to share our technical report on the Southeast Asian multilingual model Sailor2 and its latest updates!
Our 49-page report details Sailor2's development journey, including multilingual data cleaning, small model data mixture simulations, multi-stage continual pre-training, multi-stage post-training, and multi-cultural multi-lingual evaluations. Sailor2 aims to streamline the multilingual model pre-training process efficiently for the community.
🧭 We highlight Sailor2's impressive performance in low-resource language translation scenarios and its cultural understanding advantages in Southeast Asia, promoting practical applications for regional languages.
Model updates include:
💡 More precise outputs: Reduced redundancy in model outputs through refined post-training data and optimization techniques.
🌈 Handling longer texts: Expanded to handle up to 128K context length in Southeast Asian languages through long-text training.
⚡️ Faster inference: Achieved 2.5x faster inference speed with speculative decoding.
🌪️ More model sizes: Introduced new sizes of 3B and 14B through model pruning.
🌟 All models are Apache-licensed for commercial use; development tools (code, resources) are open-source.
📚 Technical report: Sailor2: Sailing in South-East Asia with Inclusive Multilingual LLMs (2502.12982)
🤖️ Models: sail/sailor2-language-models-674d7c9e6b4dbbd9a869906b
💬 Demo: sail/Sailor2-20B-Chat
📣 Sailor2 community: https://huggingface.co./sailor2
Our 49-page report details Sailor2's development journey, including multilingual data cleaning, small model data mixture simulations, multi-stage continual pre-training, multi-stage post-training, and multi-cultural multi-lingual evaluations. Sailor2 aims to streamline the multilingual model pre-training process efficiently for the community.
🧭 We highlight Sailor2's impressive performance in low-resource language translation scenarios and its cultural understanding advantages in Southeast Asia, promoting practical applications for regional languages.
Model updates include:
💡 More precise outputs: Reduced redundancy in model outputs through refined post-training data and optimization techniques.
🌈 Handling longer texts: Expanded to handle up to 128K context length in Southeast Asian languages through long-text training.
⚡️ Faster inference: Achieved 2.5x faster inference speed with speculative decoding.
🌪️ More model sizes: Introduced new sizes of 3B and 14B through model pruning.
🌟 All models are Apache-licensed for commercial use; development tools (code, resources) are open-source.
📚 Technical report: Sailor2: Sailing in South-East Asia with Inclusive Multilingual LLMs (2502.12982)
🤖️ Models: sail/sailor2-language-models-674d7c9e6b4dbbd9a869906b
💬 Demo: sail/Sailor2-20B-Chat
📣 Sailor2 community: https://huggingface.co./sailor2
SivilTaram
authored
a
paper
3 months ago
SivilTaram
authored
2
papers
4 months ago
SivilTaram
authored
2
papers
5 months ago
SivilTaram
authored
a
paper
7 months ago
dreamerdeo
authored
a
paper
7 months ago
SivilTaram
posted
an
update
8 months ago
Post
2673
Still following your human intuition to mix corpora from different sources for pre-training 🧠? Everyone says that data mixture has a big impact on model performance, but how - and why🕵️? Did you know that web corpora are actually highly impactful for downstream tasks 🏆?
Check out our preprint "RegMix: Data Mixture as Regression for Language Model Pre-training" 📄
🔬 In this paper, we've proposed an automatic data mixture method RegMix that achieves a 6.3% improvement over human selection on the widely used HellaSwag benchmark - and it only needs a 2% extra training FLOPs! 📈
📄 Paper: RegMix: Data Mixture as Regression for Language Model Pre-training (2407.01492)
💻 Code: https://github.com/sail-sg/regmix
📊 Collection: sail/regmix-data-mixture-as-regression-6682b6caab37b9442877f0ce
🎮 Demo: https://huggingface.co./spaces/sail/RegMix
Check out our preprint "RegMix: Data Mixture as Regression for Language Model Pre-training" 📄
🔬 In this paper, we've proposed an automatic data mixture method RegMix that achieves a 6.3% improvement over human selection on the widely used HellaSwag benchmark - and it only needs a 2% extra training FLOPs! 📈
📄 Paper: RegMix: Data Mixture as Regression for Language Model Pre-training (2407.01492)
💻 Code: https://github.com/sail-sg/regmix
📊 Collection: sail/regmix-data-mixture-as-regression-6682b6caab37b9442877f0ce
🎮 Demo: https://huggingface.co./spaces/sail/RegMix
dreamerdeo
authored
a
paper
8 months ago
SivilTaram
authored
9
papers
8 months ago
RegMix: Data Mixture as Regression for Language Model Pre-training
Paper
•
2407.01492
•
Published
•
36
Chain of Preference Optimization: Improving Chain-of-Thought Reasoning in LLMs
Paper
•
2406.09136
•
Published
BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions
Paper
•
2406.15877
•
Published
•
46
On Grounded Planning for Embodied Tasks with Language Models
Paper
•
2209.00465
•
Published
ARKS: Active Retrieval in Knowledge Soup for Code Generation
Paper
•
2402.12317
•
Published
Beyond Memorization: The Challenge of Random Memory Access in Language Models
Paper
•
2403.07805
•
Published
Self-Distillation Bridges Distribution Gap in Language Model Fine-Tuning
Paper
•
2402.13669
•
Published
•
1
MANTIS: Interleaved Multi-Image Instruction Tuning
Paper
•
2405.01483
•
Published
•
6
Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast
Paper
•
2402.08567
•
Published
•
2