3 75 79

Shyam Sunder Kumar

theainerd

AI & ML interests

Natural Language Processing

Recent Activity

upvoted a paper 1 day ago

Gold-medalist Performance in Solving Olympiad Geometry with AlphaGeometry2

upvoted an article 5 days ago

Welcome to Inference Providers on the Hub 🔥

reacted to chansung's post with 👍 6 days ago

Simple Paper Review #5 I briefly reviewed the paper "SFT Memorizes, RL Generalizes," which compares SFT and RL in post-training of LLM/VLM from HKU, UC Berkeley, Google DeepMind, and New York University The conclusion suggests SFT excels in memorization, while RL is better for generalization. However, since LLM/VLM should benefit humans beyond just generalization, a mix of SFT and RL is advisable. Typically, some SFT is followed by RL to understand prompt formats and enhance generalization through trial and error. The study focused on one model, Llama-3.2-Vision-11B, using environments like General Points for arithmetic reasoning and V-IRL for spatial reasoning. Training data was used for both SFT and RL, with evaluations on in-distribution and out-of-distribution data to assess memorization and generalization. I want to apply RL extensively, but it requires building a similar simulation environment. For domain-specific models, significant investment in creating a "playground" for the model is crucial, as the effort will directly influence the outcomes. https://arxiv.org/abs/2501.17161

View all activity

Organizations

theainerd's activity

upvoted a paper 1 day ago

Gold-medalist Performance in Solving Olympiad Geometry with AlphaGeometry2

Paper • 2502.03544 • Published 4 days ago • 28

upvoted an article 5 days ago

Article

Welcome to Inference Providers on the Hub 🔥

13 days ago

• 294

upvoted an article 7 days ago

Article

Open-R1: Update #1

and 7 others •

8 days ago

• 258

upvoted a collection 8 days ago

🧠 Reasoning datasets

Collection

Datasets with reasoning traces for math and code released by the community • 7 items • Updated 3 days ago • 42

upvoted a paper 9 days ago

Thoughts Are All Over the Place: On the Underthinking of o1-Like LLMs

Paper • 2501.18585 • Published 10 days ago • 51

upvoted an article 12 days ago

Article

Open-R1: a fully open reproduction of DeepSeek-R1

13 days ago

• 679

upvoted a paper 13 days ago

Humanity's Last Exam

Paper • 2501.14249 • Published 16 days ago • 54

upvoted a collection 14 days ago

Qwen2.5-1M

Collection

The long-context version of Qwen2.5, supporting 1M-token context lengths • 2 items • Updated 14 days ago • 99

upvoted an article 14 days ago

Article

We now support VLMs in smolagents!

17 days ago

• 74

upvoted 3 papers 14 days ago

Reasoning Language Models: A Blueprint

Paper • 2501.11223 • Published 20 days ago • 31

Kimi k1.5: Scaling Reinforcement Learning with LLMs

Paper • 2501.12599 • Published 18 days ago • 90

DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Paper • 2501.12948 • Published 18 days ago • 305

upvoted 2 papers 20 days ago

Towards Best Practices for Open Datasets for LLM Training

Paper • 2501.08365 • Published 26 days ago • 53

Evolving Deeper LLM Thinking

Paper • 2501.09891 • Published 23 days ago • 105

upvoted a collection 21 days ago

IndicLLMSuite

Collection

Largest Collections of Pretraining and Instruction Finetuning datasets for 22 Indic languages. • 4 items • Updated Nov 5, 2024 • 15

upvoted an article 23 days ago

Article

MiniMax-01 is Now Open-Source: Scaling Lightning Attention for the AI Agent Era

•

25 days ago

• 40

upvoted an article 24 days ago

Article

The Large Language Model Course

•

24 days ago

• 89

upvoted a paper 24 days ago

MiniMax-01: Scaling Foundation Models with Lightning Attention

Paper • 2501.08313 • Published 26 days ago • 273

upvoted an article 25 days ago

Article

AI Agents Are Here. What Now?

28 days ago

• 60

upvoted a paper 27 days ago

Search-o1: Agentic Search-Enhanced Large Reasoning Models

Paper • 2501.05366 • Published about 1 month ago • 92