Nlp Alignment

LLM Alignment Best Practice

Methods:

• RLHF: Train reward model → PPO fine-tuning (complex but powerful)
• DPO: Direct preference optimization (simpler, no reward model needed)
• GRPO: Group relative policy optimization
• SFT: Supervised fine-tuning as alignment baseline

Training recipe:

• Start with SFT on high-quality instruction data
• DPO: lr=5e-7, beta=0.1, batch_size=64
• PPO: lr=1e-6, clip=0.2, KL coeff=0.02
• Use reference model for KL penalty
• Evaluate on safety benchmarks (TruthfulQA, BBQ, etc.)

Common pitfalls:

• Reward hacking: model finds shortcuts to high reward
• Mode collapse: model generates repetitive outputs
• Catastrophic forgetting: loses general capabilities

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: thada2402
• Source: thada2402/AutoResearchClaw
• License: MIT

Install and usage instructions live in the source repository linked above.