MEVZU N° TAG / VOL. 133
#reasoning
0 blog · 0 news · 13 wiki
Wiki
Extended Thinking
Anthropic's user-visible long-thinking mode introduced for Claude models.
- EN
- Extended Thinking
- TR
- Uzun Düşünme (Extended Thinking)
Long Thinking
When a model performs extensive internal reasoning across many tokens before producing its final answer.
- EN
- Long Thinking
- TR
- Uzun Düşünme
Inference-time Reasoning
A paradigm where the model reasons step by step at response time to arrive at a more accurate answer.
- EN
- Inference-time Reasoning
- TR
- Çıkarım-Anı Akıl Yürütme
Test-time Compute
An approach where a model spends more compute at inference time, not just during training, to lift performance.
- EN
- Test-time Compute
- TR
- Test-Anı Hesaplama
Reasoning Models
A new generation of language models that engage in long-form thinking at inference time to tackle complex problems.
- EN
- Reasoning Models
- TR
- Akıl Yürüten Modeller
DeepSeek R1
DeepSeek's 2025 open-weight reasoning model, which shook the industry with low-cost long-form thinking.
- EN
- DeepSeek R1
- TR
- DeepSeek R1
OpenAI o3
The next-generation member of OpenAI's o-series reasoning models, focused on deeper reasoning and agent use.
- EN
- OpenAI o3
- TR
- OpenAI o3
OpenAI o1
OpenAI's 2024 release introducing inference-time long thinking — the first major reasoning model.
- EN
- OpenAI o1
- TR
- OpenAI o1
Chain-of-Thought (CoT)
Prompting the model to reason step-by-step before producing a final answer.
- EN
- Chain-of-Thought (CoT)
- TR
- Düşünce Zinciri
Reasoning
A model's capacity to work through a problem in multiple steps — now a central axis of LLM competition.
- EN
- Reasoning
- TR
- Akıl Yürütme (Reasoning)
Self-Critique
A technique in which a model evaluates its own output against explicit criteria to surface errors and weaknesses.
- EN
- Self-Critique
- TR
- Öz-Eleştiri
Planning
An agent's ability to design and order the steps it will take to reach a goal.
- EN
- Planning
- TR
- Planlama
Tree-of-Thought (ToT)
A reasoning approach that explores multiple branches in parallel instead of a single chain, then picks the best path.
- EN
- Tree-of-Thought (ToT)
- TR
- Düşünce Ağacı