concept · created Apr 27, 2026 · updated Apr 27, 2026

post-training

#llm-training#post-training#alignment

Everything that happens to a model after pretraining and before deployment: instruction tuning, preference optimization, reasoning RL, alignment. By 2026 this is the layer where most of the perceived capability gap between modern models lives. 2026-04-27-llm-training-principles-paths-practices‘s reference point: a 1.3B InstructGPT model, post-trained with alignment + preference optimization, beat a 175B GPT-3 on human preference — two orders of magnitude in parameter count, the smaller model won.

The four routes

Modern post-training is rarely a single technique; the landscape is four overlapping methods:

MethodWhat it doesTrade
SFT (supervised fine-tuning)Train on (instruction, response) pairs. Teaches form: how to answer, how to format, how to be an assistant.Cheap, well-understood; teaches style as much as content.
RLHF (reinforcement learning from human feedback)First imitate high-quality responses (SFT), then use a reward model trained on human preference comparisons to do RL.The original recipe; expensive, hard to keep stable, vulnerable to reward overfitting.
DPO (direct preference optimization)Skip the separate reward model — learn directly from preference pairs against a reference policy.Simpler stack, no separate reward model, often comparable to RLHF on common tasks.
RFT (reinforcement fine-tuning)The productized interface ([[openaiOpenAI]]‘s framing): user supplies task, grader, reward signal. The training infrastructure is hidden.

These aren’t strict rivals; production recipes often chain them.

SFT teaches style, not just knowledge

A subtle but consequential point in the article: SFT data length, formatting choices, citation style, list-vs-prose preferences propagate directly into model output shape. Combined with the empirical bias of preference evaluations toward longer, well-structured responses, this means leaderboard differences between models are often style differences, not capability differences.

Practical implication: leaderboard wins from “improved post-training” sometimes mean “post-trained to produce the kind of output the eval rewards”. Real-task results, cost, and stability are the calibration set.

The DeepSeek-R1 four-stage recipe (the public reference)

DeepSeek-R1‘s technical report is the cleanest public account of a modern post-training pipeline, and the article walks through all four stages:

StageWhatWhy
1. Cold-start SFTSmall high-quality CoT dataset, light supervised fine-tune of the base modelDeepSeek-R1-Zero showed pure RL on a base model is feasible but produces repetition, language mixing, unreadable output. The cold start gives RL a stable starting point.
**2. Reasoning RL with [[grpoGRPO]]**RL on math / code / logic with verifiable correctness as the reward
3. Rejection-sampling fine-tuneFilter successful trajectories from stage 2 → new SFT dataset → another supervised passBridges RL → SFT: the good RL trajectories become high-quality training examples for the next round.
4. Alignment RLHelpfulness + safety preference reward, RL passFinal shaping into a release-ready assistant.

Stages 3 and 4 iterate. The article’s point: the gap between “just SFT” and the full four-stage recipe is visible in published results, not a theoretical improvement.

The cold-start stage is the article’s specific anti-pattern flag — R1-Zero proved cold start isn’t strictly required for RL to work, but skipping it gives you a model that’s strong on benchmarks and unusable for users.

Why RLHF is being supplemented by verified rewards

Within RL post-training, the field is shifting toward verified rewards — programmatic correctness checks (math, code, logic) instead of (or alongside) human preference labels. The benefits: cheaper, more consistent, less prone to label drift. The catch: works only where verification is automatable. Open-ended tasks (writing, advice, design) still depend on preference signals, with all the reward-hacking risks that entails.

Where this layer fits

  • Upstreampretraining decides what the model can learn to do; post-training decides what it will do.
  • Adjacenteval-grader-reward design is post-training, in the sense that the reward / grader / eval triple is the actual training target, not a separate evaluation step.
  • Downstreamreasoning-models (o1, R1) emerged from pushing reasoning RL hard within post-training. Agent training is the further generalization: post-training in environments rather than on static tasks.
  • Sidewaysconstitutional-ai and deliberative-alignment are alignment-specific post-training techniques that fold safety into the training target rather than tacking it on at the end.

Diagnostic implication

If a model “got better” between releases, post-training is the most common single source — instruction following, refusal behavior, tool-use shape, response style. Pretraining changes are slower, more expensive, and harder to attribute. Per llm-training-pipeline‘s closing checklist: when you can’t explain a release-to-release improvement by pretraining, look at post-training first.

Post-training as “RL with reasoning in the action space” (Yao 2025)

Shunyu Yao’s The Second Half reframes the recent post-training wave from a different angle. The author’s RL-side decomposition is algorithm + environment + priors. By 2024, two pieces had quietly fallen into place: language pretraining provided priors (pretraining), and adding reasoning to the action space (react) provided a way to use them generatively. Once those two were locked, the RL algorithm became the easy part — that’s why the o-series, R1, deep research, and computer-using agents arrived close together.

Read this way, the four-stage DeepSeek-R1 recipe above is one instantiation of the broader pattern; the GRPO / cold-start / verifiable-rewards details matter, but the load-bearing move was earlier — adding reasoning as an action so RL on top of language priors could generalize. The author’s own pre-2022 work on text-game agents (CALM, the first agent built via pre-trained LMs) needed millions of RL steps per game and didn’t transfer; the same model class with reasoning-as-action transfers across SWE / math / web / computer-use without retargeting.

Practical consequence for diagnosing post-training improvements: ask what’s changed at the action-space layer, not only at the algorithm-and-data layer.

  • Did the model learn to allocate inference compute over reasoning tokens differently? → reasoning-models axis.
  • Did the action space gain new tools or environments? → harness axis.
  • Did the cold-start / preference / verified-reward stages change? → the four-stage recipe axis above.

The three are coupled (the four-stage recipe is how the reasoning policy gets trained), but they fail differently and want different diagnostic moves.

Referenced by 18

2026-04-27-llm-training-principles-paths-practices 2026-04-27-the-second-half-of-ai chinchilla-scaling constitutional-ai data-engineering deliberative-alignment distillation eval-grader-reward grpo llm-training-pipeline pretraining react reasoning-models cursor deepseek openai shunyu-yao tw93
esc