Reasoning

Spend a configurable budget of internal reasoning tokens before producing a user-visible answer.

Elicit multi-step reasoning by prompting the model to produce intermediate steps before its final answer.

Reduce hallucination by drafting an answer, generating independent verification questions, answering them in isolation, and revising.

Route reasoning-heavy tasks to a reasoning-tuned model that trades inference time for deliberation, rather than to a fast LLM that exhibits premature-closure.

Drive the agent toward its goal by asking the user a sequence of strategic, open-ended questions that surface the user's own latent knowledge, goal, or context — rather than producing an answer direc…

Have the model emit explicit follow-up sub-questions, answer them (optionally via search), then compose the final answer.

Allocate more inference-time compute (samples, search, deeper thinking) instead of scaling parameters to improve quality.

Elicit step-by-step reasoning with a single trigger phrase rather than few-shot exemplars.

Allocate inference-time compute (thinking tokens, samples, depth, model size) per query based on input difficulty, rather than using a fixed budget across all queries.

Replace free-form chain-of-thought with a domain-tailored sequence of structured queries that re-anchor the model's attention to the critical instructions and prior decisions at the exact generation…

Generate multiple candidate solutions in parallel, then systematically test each against declared constraints rather than committing to the first plausible one — adapted from Langley & Simon's cognit…

Iterate generate → reward-filter → fine-tune to bootstrap reasoning capabilities without human-labelled data.

Bootstrap a model's reasoning by training it on its own correct chain-of-thought outputs.

Search over a tree of partial reasoning states with explicit lookahead, evaluation, and backtracking.

Decompose a hard problem into an ordered list of easier subproblems, then solve them sequentially with each answer feeding the next.

Model reasoning as an arbitrary DAG so thoughts can be merged, refined, and aggregated across branches.

Let the model reason in continuous hidden-state space instead of decoding each step to text, feeding the last hidden state back as the next input embedding, so one latent step can hold several contin…

Treat an over-long prompt as an environment the model navigates by code, letting it partition and recursively call itself over snippets, so it answers over inputs far larger than its context window.

When an agent's outputs are extended action-observation trajectories rather than short answers, scale test-time compute by compressing each rollout into a structured summary and selecting or reusing…