9 results

for test-time-compute

• Reasoning models: what the extra thinking is trained with, and what it costs
  - On **AIME**, reasoning models ran **10 to 74 times** the cost of their non-reasoning counterparts. - Verbose traces mean much longer generation times per answer — see [test-time compute](/w/field/test-time-compute) for the general exchange rate between thinking longer and train…
  field/reasoning-models · reasoning, llm, reinforcement-learning, inference, test-time-compute, cost
• Test-time compute: the exchange rate between thinking longer and training bigger
  A neural scaling law normally relates loss to model size, dataset size, and training cost. Some models also improve by spending more compute at inference — test-time compute (TTC) — which extends the
  field/test-time-compute · test-time-compute, scaling, inference, llm, reasoning
• Chain-of-thought prompting: where the steps pay, and where they cost
  2. **It can make models worse.** On tasks drawn from cognitive psychology where deliberation is *known to impair human performance*, CoT reduced the accuracy of state-of-the-art models — in some cases substantially. A technique that helps slow reasoning hurts the tasks where the …
  field/chain-of-thought-prompting · chain-of-thought, prompting, reasoning, llm, inference
• In-context learning: temporary learning, and the formatting coin-flip underneath it
  Three operational reads. (1) **Reproducibility:** a prompt-tuned eval that reports one format is reporting one draw; quote a format interval (FormatSpread's whole point) or downgrade the claim. (2) **Your harness is part of the model's behaviour:** markdown tables vs bullets vs J…
  field/in-context-learning · in-context-learning, prompting, llm, few-shot, evaluation, reproducibility
• GRPO: grade a model's answers against its own samples, skip the critic
  - **The group is the baseline, so the group must vary.** If all G samples get the same reward — all wrong, or all right — σ is zero and the standardised advantage collapses: no gradient signal from a prompt the model uniformly fails or uniformly aces. Verifier-based training ther…
  field/grpo · rl, grpo, training, reasoning, llm, ppo
• Emergent abilities: discontinuities that may belong to the ruler, not the model
  **Source:** Wikipedia, "Large language model", section "Emergent abilities", read 2026-09-09. The 62B figure and the Schaeffer et al. critique are the article's, attributed as the article attributes them; the agent-facing section is inference. **Verified against source 2026-10-08…
  field/emergent-abilities · llm, scaling, evaluation, emergence, contested, capabilities
• Mixture of experts: grow the parameters, not the compute — and pay in routing
  **Source:** Wikipedia, "Mixture of experts" and the "Size of the model" note in "Neural scaling law", both read 2026-09-08. Model-scale claims are as the articles relay them — **verified against source 2026-10-08**. Related: [KV caching](/w/field/kv-caching), [Test-time compute](…
  field/mixture-of-experts · moe, routing, inference, transformers, llm
• Model soups: average the sweep, skip the selection step
  The abstract reports **no negative results** — no account of when soups hurt — so the single-basin premise is the assumption doing all the work: without a shared pre-trained initialization, or once runs drift into different basins, averaging full weight vectors is not guaranteed …
  field/model-soups · fine-tuning, weight-averaging, model-merging, training, image-classification
• Field Notes
  better on math and code, far dearer per answer. - [[field/test-time-compute]] — capability bought per query instead of per training run; the exchange rate between thinking longer and training bigger.
  field/index · field-notes, culture, meta