Put an idea to the test.

Understand importance sampling, PPO clipping and CISPO through intuitive examples, equations and the policy updates used in this lab.

Follow GRPO from group sampling and rewards to policy updates on GSM8K, with runnable PyTRIO code and a comparison of three loss functions.

Build a minimal on-policy distillation loop on DeepMath-103K: student sampling, teacher scoring and reverse KL, with code and evaluation notes.

Start from medical SFT and study SAR-OPD and IDT-OPD, with specialist and general-task evaluations and explicit experimental limitations.

Put online search inside the rollout loop and train a model to retrieve evidence across turns, with switchable search backends and evaluation notes.

Study self-distillation with a fixed step-0 teacher and trajectories sampled by the student, including the training loop and mathematical reasoning results.

Connect a local Python sandbox and study reinforcement learning with interleaved reasoning and code execution, including rollout handling and results.

Examine DAPO’s four core changes and the practical time cost of dynamic sampling through training code, equations and recorded experiments.

Implement sequence-level importance ratios and clipping, and distinguish the core GSPO loss reproduced here from a complete training reproduction.

Train a household agent in a real TextWorld environment, with long trajectories, environment feedback, group-relative advantages and evaluation notes.

Connect image inputs, geometry reasoning and verifiable rewards on GeoQA, with multimodal training code and evaluation details.

Explore macro-step optimization and a generative critic in ALFWorld, with clear boundaries between the public TEMPO description and this reproduction.

Use a skill-guided self-teacher to provide turn-level learning signals, then examine multi-turn agent training and paired evaluation in ALFWorld.

Move from SFT to GRPO for astronomical candidate review, interleaving spectrum images, reasoning and tools with explicit evaluation settings.

Fine-tune Qwen3.5-4B on 2,160 records with LoRA, then run restricted-softmax inference, temperature calibration and five benchmark evaluations.