Reasoning Models
Test-time compute, process reward models, the o-series, DeepSeek-R1 and contamination.
15concepts
196flashcards
122minutes of reading
- 01 Chain-of-Thought Faithfulness Whether a model's stated reasoning is the reasoning that produced its answer, how the hint-injection test measures it, and why unfaithful CoT undermines monitoring more than it undermines accuracy.
- 02 Controlling Reasoning Length The mechanisms that set how long a model thinks, from s1's crude budget forcing to length-targeted RL, and why overthinking makes length control a quality problem rather than just a cost problem.
- 03 DeepSeek-R1 and the Open Reasoning Recipe How DeepSeek's R1 pipeline produced o1-class reasoning with an open paper, an open model, and a recipe other labs could replicate within weeks.
- 04 Deliberative Alignment Teaching a model the text of its own safety policy and training it to reason over that text before answering, which improves jailbreak robustness and overrefusal at the same time.
- 05 Inference-Time Scaling Laws and Thinking Budgets How accuracy trades against tokens spent at inference, why the optimal strategy depends on question difficulty, and what a thinking budget actually buys.
- 06 Process Reward Models and Verifiable Rewards Why scoring every step of a reasoning trace beats scoring only the final answer, and how Ai2 and DeepSeek replaced PRMs entirely with programmatic correctness checks.
- 07 Reasoning Distillation and Budget Forcing A thousand carefully chosen reasoning traces and a decoding trick can move a 32B model past o1-preview on competition maths, which says something uncomfortable about what reasoning training is actually teaching.
- 08 The Limits of Self-Correction Why asking a model to review its own answer often makes it worse, what separates intrinsic from extrinsic correction, and the oracle-label leak that inflated the early results.
- 09 Tool-Integrated Reasoning Interleaving code execution with natural-language reasoning removes an entire class of error the model cannot fix by thinking harder, and creates a new class of error it cannot see.
- 10 Verifier-Guided Search and Best-of-N Sampling many candidate solutions is only useful if you can pick the right one, and the gap between what sampling can reach and what selection can find is the central constraint on test-time scaling.