{"slug":"reasoning-models","topic":"reasoning-models","title":"Reasoning Models & Test-Time Compute: State of the Field (October 2026)","summary":"An expert briefing on reasoning models and test-time compute as of late October 2026, covering reinforcement learning with verifiable rewards, distillation, parallel, sequential and search-based inference scaling, process and generative verifiers, latent reasoning, chain-of-thought faithfulness and monitorability, and evaluation practice. The main takeaway is that reinforcement learning on automatically checkable tasks, combined with large inference budgets, has produced gold-threshold olympiad results in several recent preprints and company reports, although those results differ in grading and verification. The open questions concern mechanism and measurement: whether RL creates new capability or sharpens what pretraining supplied, how to reward work that cannot be checked automatically, and whether visible reasoning faithfully reflects the computation behind an answer.","version":1,"published_at":"2026-10-29","updated_at":"2026-10-30","stats":{"words":5658,"citations":111,"claims":39,"timeline":20,"benchmarks":15,"open_problems":7,"key_papers":15,"glossary":20},"verification":{"status":"verified","method":"automated figure and citation-relevance triage against cited abstracts, targeted AI review with web checks of flagged statements, and a close-out pass hedging or removing unconfirmed statements; automated citation resolution","claims_checked":80,"issues_fixed":2,"verified_at":"2026-10-30"},"ai_assisted":true,"price":{"usd":"0.20","protocol":"x402","network":"eip155:84532","asset":"USDC"},"url":"https://space-kitty.com/dossiers/reasoning-models","preview_url":"https://space-kitty.com/api/v1/dossiers/reasoning-models/preview","purchase_url":"https://space-kitty.com/api/v1/dossiers/reasoning-models","toc":[{"level":2,"text":"Executive summary","anchor":"executive-summary"},{"level":2,"text":"Background and core concepts","anchor":"background-and-core-concepts"},{"level":2,"text":"Timeline of key milestones","anchor":"timeline-of-key-milestones"},{"level":2,"text":"State of the art","anchor":"state-of-the-art"},{"level":3,"text":"Reinforcement learning with verifiable rewards","anchor":"reinforcement-learning-with-verifiable-rewards"},{"level":3,"text":"Distillation and small-data fine-tuning","anchor":"distillation-and-small-data-fine-tuning"},{"level":3,"text":"Test-time scaling: parallel, sequential and search","anchor":"test-time-scaling-parallel-sequential-and-search"},{"level":3,"text":"Verifiers and process rewards","anchor":"verifiers-and-process-rewards"},{"level":3,"text":"Olympiad mathematics and competitive programming","anchor":"olympiad-mathematics-and-competitive-programming"},{"level":3,"text":"Latent, recurrent and efficient reasoning","anchor":"latent-recurrent-and-efficient-reasoning"},{"level":3,"text":"Tools, agents, multimodality and other domains","anchor":"tools-agents-multimodality-and-other-domains"},{"level":3,"text":"Quantitative benchmarks","anchor":"quantitative-benchmarks"},{"level":2,"text":"Open problems and active debates","anchor":"open-problems-and-active-debates"},{"level":2,"text":"Major programs and players","anchor":"major-programs-and-players"},{"level":2,"text":"Outlook: what to watch (November 2026 to April 2028)","anchor":"outlook-what-to-watch-november-2026-to-april-2028"},{"level":2,"text":"Key papers","anchor":"key-papers"},{"level":2,"text":"Glossary","anchor":"glossary"}],"sample_claims":["DeepSeek-R1-Zero was trained with reinforcement learning rather than human-labeled reasoning trajectories, and the peer-reviewed DeepSeek-R1 paper reports that advanced reasoning patterns emerged from this pure-RL setup.","Repeated sampling raised coverage, the fraction of problems solved by at least one sample, roughly log-linearly across four orders of magnitude of sample count."],"includes":["markdown","claims","timeline","benchmarks","open_problems","key_papers","glossary","bibliography"],"citation_count":111,"how_to_buy":"GET https://space-kitty.com/api/v1/dossiers/reasoning-models with an x402 client ($0.20, USDC on Base Sepolia (testnet))."}