eoe's picture

eoe

eoe

·

AI & ML interests

None yet

Recent Activity

upvoted a collection 12 days ago

2026 April 🐝 - China Open Source Highlights

reacted to anakin87's post with ❤️ 15 days ago

How LLM training with RL Environments works? It all starts with 𝗥𝗲𝗶𝗻𝗳𝗼𝗿𝗰𝗲𝗺𝗲𝗻𝘁 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗩𝗲𝗿𝗶𝗳𝗶𝗮𝗯𝗹𝗲 𝗥𝗲𝘄𝗮𝗿𝗱𝘀 - question asked - model generates reasoning + answer - answer checked against ground truth - reward drives RL training In this setup, the environment is simple: fixed questions and answers, rollout logic, reward(s) Consider a more complex tic-tac-toe env ❌⭕ It adds: - dynamic game generation/handling - tunable opponent skill - multi-turn interactions (envs can also include tools) --- What happens at training? We use 𝗚𝗿𝗼𝘂𝗽 𝗥𝗲𝗹𝗮𝘁𝗶𝘃𝗲 𝗣𝗼𝗹𝗶𝗰𝘆 𝗢𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻 with a tic-tac-toe env No critic model needed, the group is the baseline Simpler than PPO 1️⃣ Rollout generation: from the same board, model plays N games via sampling 2️⃣ Each game scored with deterministic rewards (win, format, ...) 3️⃣ Mean score computed across the group 4️⃣ Each rollout's advantage = its score minus the group mean 5️⃣ Model updated to favor trajectories above baseline 🔁 Repeat For a deep dive, check out 🌱 https://github.com/anakin87/llm-rl-environments-lil-course a free hands-on course on RL environments for LLMs

liked a model 22 days ago

DunnBC22/vit-base-patch16-224-in21k_Human_Activity_Recognition

View all activity

Organizations

None yet

upvoted a collection 12 days ago

2026 April 🐝 - China Open Source Highlights

32 items • Updated 7 days ago • 3

upvoted a paper 8 months ago

UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning

Paper • 2509.02544 • Published Sep 2, 2025 • 127

upvoted a collection about 1 year ago

AMD-RyzenAI-Deepseek-R1-Distill-Hybrid

3 items • Updated Feb 19 • 5

upvoted a collection over 1 year ago

Qwen2

Qwen2 language models, including pretrained and instruction-tuned models of 5 sizes, including 0.5B, 1.5B, 7B, 57B-A14B, and 72B. • 37 items • Updated Mar 2 • 376