llm-post-training

Here are 8 public repositories matching this topic...

Richard-zrx / ADS

Learning at the Right Pace: Adaptive Data Scheduling improves LLM reinforcement learning with semantic-cluster and policy-boundary sampling.

reinforcement-learning reasoning semantic-clustering adaptive-sampling mathematical-reasoning large-language-models grpo rlvr data-scheduling llm-post-training llm-reinforcement-learning adaptive-data-scheduling

Updated Jul 5, 2026
Python

AgriciDaniel / reinforcement-learning-second-brain

Star

Your RL second brain: 34 source-cited topics from Q-learning to GRPO and agentic RL, kept fresh monthly. Obsidian vault + AI agent layer, Brainstein SSS+.

machine-learning reinforcement-learning deep-reinforcement-learning knowledge-base obsidian ppo dpo second-brain obsidian-vault rlhf rlaif grpo llm-post-training

Updated Sep 11, 2026
Python

xcyao00 / HTPO

Star

A novel RL algorithm that designs hierarchical token-level optimization objective for exploration-exploitation balanced policy optimization

reinforcement-learning reasoning large-language-models llm-post-training

Updated May 17, 2026
Python

John-Li-A / Mr.When-to-GRPO

Star

Paired-intervention toolkit for testing when to switch from on-policy distillation to GRPO.

reinforcement-learning grpo verl qwen3 on-policy-distillation llm-post-training

Updated Jul 31, 2026
Python

YiFanWangSCU / skill-annealing

Star

Controlled LLM post-training study of Agent robustness under reduced skill prompts.

agent evaluation lora sft prompt-robustness grpo llm-post-training

Updated Aug 8, 2026
Python

huanggeyu717-beep / regulated-agent-posttraining-lab

Star

金融工单 Agent 的数据工程与后训练实验：数据/意图/工具/政策四张注册表 → SFT / DPO / tool-use SFT 导出 → Qwen2.5-3B LoRA；held-out 政策评测 88/91，加推理时 guard 后 91/91。

python lora sft dpo tool-use qwen llm-evaluation agent-safety dataset-engineering llm-post-training

Updated Aug 21, 2026
Python

Linwei-Chen / llm-post-training-research-atlas

Star

按优化机制与反馈闭环导航 2017—2026 年关键方法、证据和失效边界｜56 项记录

survey literature-review academic-research llm-post-training research-atlas

Updated Aug 16, 2026
JavaScript

hxm2023 / GRPO-Guard

Star

Machine-verifiable evidence chain for online LLM post-training (TRL GRPO + vLLM): trajectory contracts, fault injection F1-F8, guarded updates, paired gradient replay

fault-injection trl rlhf vllm llm-observability grpo llm-post-training training-infrastructure

Updated Aug 26, 2026
Python

Add this topic to your repo

To associate your repository with the llm-post-training topic, visit your repo's landing page and select "manage topics."

Learn more