4 16

Krrish Agarwalla

kkkamur07

https://krrish-a07.github.io/

AI & ML interests

Anything DL

Recent Activity

liked a model 16 days ago

deepseek-ai/DeepSeek-V4-Pro

reacted to anakin87's post with ❤️ 17 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 2 months ago

nvidia/personaplex-7b-v1

View all activity

Organizations

None yet

liked a model 16 days ago

deepseek-ai/DeepSeek-V4-Pro

Text Generation • 862B • Updated 4 days ago • 1.17M • • 3.78k

reacted to anakin87's post with ❤️ 17 days ago

Post

10390

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