Qwen-VL Paper Notes: Visual Adapter, Cross-Attention, 2D Position Encoding, and RoPE

Training and evaluation of the slime Search-R1-trained Qwen2.5-3B model on NQ and HotpotQA.

Comparing PPO and GRPO for post-training Qwen2.5-0.5B-Instruct on GSM8K with veRL, focusing on a small-model, verifiable-reward setting where the reward is rule-based correctness rather than a learned reward model.

Classic RL Methods including DP, MC, and TD

RL Preliminary

From CartPole-v1 to CartPole Swing-Up

slahmr examples

Kogge-stone scan and Brent-Kung scan

reduction trees

GPU architecture and CUDA programming model