「how to use」の検索結果

Identifies the real underlying study design used in a medical or biomedical paper, distinguishes primary and secondary design components when papers are hybrid, and converts the paper into an evidence-aware design label suitable for literature appraisal, evidence grading, and downstream review workflows. Always identify the actual design from what the study did, not from how the authors describe it. Never fabricate references, metadata, or study features.

日本語の概要は準備中です。原文の説明を表示しています。

Use when evaluating markets through Howard Marks-style second-level thinking, credit cycles, risk premiums, market psychology, and defensive versus aggressive posture.

日本語の概要は準備中です。原文の説明を表示しています。

Reference for the TRL (Transformer Reinforcement Learning) library codebase. Use proactively before reading or editing any file under `trl/` so you have the intended contracts and invariants in mind, not just what the current code says. Covers trainer hierarchy (SFT, DPO, GRPO, KTO), shared utility functions (selective_log_softmax, decode_and_strip_padding, padding helpers), configuration system, model wrappers, and how data flows through any TRL trainer.

日本語の概要は準備中です。原文の説明を表示しています。

Diagnostic guide for RL-based post-training of language models (GRPO, PPO, REINFORCE, DPO). Use proactively when debugging a training pipeline that shows no improvement, loss anomalies, reward stagnation, NaN gradients, or other unexpected behavior during reinforcement learning fine-tuning. Work through all pipeline stages — reward, advantages, log-probs, loss, generation/decoding — before concluding the diagnosis is complete; stopping after one or two fixes is a common failure mode. Covers log-probability math, advantage estimation, numerical stability, reward processing, and generation/decoding pipeline issues.

日本語の概要は準備中です。原文の説明を表示しています。

Worked qupai-system (曲牌体) opera scripts, act by act. Guan Hanqing's three surviving zaju (窦娥冤／救风尘／金线池 — twelve acts, 126 tunes, one mode and one rhyme per act, sung/spoken ratios); Tang Xianzu's 牡丹亭 (55 scenes, ~420 tunes, 集唐 closing quatrains, the 圆驾 south-north composite suite); Kong Shangren's 桃花扇 (44 scenes plus four unnumbered, the 副末 frame line, upper/lower halves, a fan's seven transformations, his sixteen 凡例); 长生殿 (Wang Jilie's fifty-scene staging table); Wei Minglun's 巴山秀才 (1983 Sichuan 高腔, 43 tunes, nine 帮腔 passages). Every case states its edition and collation status, then structure, tunes, stage conventions and transferable devices. Method is in sw-chinese-opera-qupai; this skill holds the worked scripts. Use when you want a real example of how a 传奇 or 杂剧 is assembled, need actual scene counts and tune counts to imitate, are judging where a famous script breaks prosody, or need models for 副末开场, 集唐下场诗, 南北合套 or 帮腔 (曲牌体全本／牡丹亭／桃花扇／窦娥冤／救风尘／长生殿／巴山秀才／唱段表／结构表).

日本語の概要は準備中です。原文の説明を表示しています。

Use when designing how one canonical agent core runs across Codex, Claude Code, Gemini CLI, Cursor, and AGENTS.md-compatible tools.

日本語の概要は準備中です。原文の説明を表示しています。

Run a fast, honest competitive scan — the dimension table built from public evidence (sites, docs, pricing pages, changelogs, reviews), the claims-vs-observed discipline, and the so-what synthesis that ends in moves, not a landscape mural. Use when asked what are competitors doing, quick scan of these three rivals, how does our pricing/feature set compare, or prep the competitive slide honestly. Produces the evidence-based comparison table, the marketing-vs-reality flags, the so-what synthesis, and the staleness date.

日本語の概要は準備中です。原文の説明を表示しています。

Write a one-page CV for management consulting, investment banking or private equity, in the dense format those screeners expect: education first, quantified bullets, leadership and interests that show drive. Use when asked to write my consulting CV, banking resume, PE CV, make my CV for McKinsey, BCG, Bain, Goldman or a boutique bank, or fit my experience on one page for finance. Produces the one-page CV, a bullet-by-bullet strength check, and the three lines a screener will read first.

日本語の概要は準備中です。原文の説明を表示しています。

Compute what a degree will actually cost — sticker minus real aid, inflated per year, split into cash and loans, with the loan's decade-long monthly tail made visible before enrollment instead of after. Use when asked what will college really cost, compare these two offers' real prices, how much loan payment after graduation, or is this school affordable. Produces the all-in number from the script, the offer-letter decode (grants vs loans untangled), the monthly-tail reality check, and the two-school comparison.

日本語の概要は準備中です。原文の説明を表示しています。

Plan a co-marketing partnership — two brands reaching each other's audiences for mutual gain. Use when asked to plan a partnership, joint campaign, co-branded content/webinar, integration launch, or partner outreach. Produces the partner fit rationale, a fair value exchange, the joint campaign plan, the partner pitch, and how success is split and measured.

日本語の概要は準備中です。原文の説明を表示しています。

Make a genuinely good drink from the bottles already on your shelf — no special trip, no 12-ingredient recipe. Use when asked what can I make with [spirits], cocktail from what I have, I've got [bottles] what can I drink, or make me a drink without buying anything. Produces two or three cocktails you can build right now with ratios, a method, sensible substitutions for what you're missing, and a zero-proof version — scaled to how many you're making.

日本語の概要は準備中です。原文の説明を表示しています。

Use when asked 医保怎么报销, 异地就医怎么备案, 住院能报多少, 门诊报销比例, 手工报销要什么材料, 惠民保值不值得买, or work out what China's basic medical insurance (医保) will pay for a treatment. Produces the person's scheme (职工医保 or 居民医保), how a claim works for this case (direct settlement or manual reimbursement), an out-of-pocket estimate built from the city's own deductible, rate and cap, the cross-region registration (异地就医备案) steps, and the documents to keep.

日本語の概要は準備中です。原文の説明を表示しています。

Keep citations honest in business documents — every load-bearing claim sourced, links that actually contain the claim, the as-of dates that keep numbers honest, and the internal-vs-external sourcing rules for decks and memos. Use when asked check the sourcing in this deck, add citations to this doc, our slide says 'studies show' — which studies, or set citation norms for the team. Produces the claim-by-claim audit, the fixes (source found, claim softened, or cut), the citation format for the venue, and the team norm card.

日本語の概要は準備中です。原文の説明を表示しています。

Assess physical and transition climate risk for a site, product, or portfolio with scenario-based structure. Use when asked to run a climate risk assessment, evaluate physical or transition risk, prepare TCFD/ESRS-style climate risk analysis, or assess how climate scenarios affect an asset or business. Produces a hazard-exposure-vulnerability matrix across 2030/2040/2050 horizons and labelled scenarios, with financial-impact ranges, confidence levels, and adaptation options.

日本語の概要は準備中です。原文の説明を表示しています。

Prepare for China's annual individual income tax reconciliation (个税年度汇算): check income, deductions and special additional deductions, compare the two treatments of a year-end bonus, and estimate the refund or amount to pay before filing in the 个人所得税 app. Use when asked 个税汇算怎么弄, 年终奖单独计税还是合并, 我能退多少税, 专项附加扣除, or how does China's annual tax reconciliation work. Produces a filing checklist, the deductions the person may be missing, a bonus comparison, the estimate with workings, and deadlines. Not tax advice.

日本語の概要は準備中です。原文の説明を表示しています。

Show or change the Cassette editing model and thinking level for the current media session. Invoke only when the user explicitly asks for the model picker, wants to inspect the current Cassette model, or requests a model/thinking change.

日本語の概要は準備中です。原文の説明を表示しています。

Coordinate several workers: what idle agents do next, which finished work gets checked first, how to recover a dead one. Use when: managing multiple agents.

日本語の概要は準備中です。原文の説明を表示しています。

Use when packaging AISTATS code, data, proofs, simulation scripts, notebooks, random seeds, and logs as anonymous supplementary evidence or public post-acceptance artifacts, even when there is no separate artifact badge. Covers what statistically minded AISTATS reviewers inspect first and how to make Monte Carlo studies turnkey.

日本語の概要は準備中です。原文の説明を表示しています。

Use when preparing AISTATS supplementary material, appendices, proof details, code/data archives, simulation scripts, additional tables, and anonymized artifacts under deadline, size, anonymity, and reviewer-discretion constraints, including how to split a theory-plus-experiments paper between body and supplement.

日本語の概要は準備中です。原文の説明を表示しています。

Use when deciding what characterization and experimental detail an Advanced Materials manuscript must show to prove its structure–property claim, and what belongs in the Experimental Section and Supporting Information. Partitions and stress-tests methods; does not design figures or run analysis.

日本語の概要は準備中です。原文の説明を表示しています。

Use when an American Economic Journal: Applied Economics (AEJ: Applied) manuscript's headline estimate must be shown to survive specification, sample, and inference choices before submission or in an R&R. Builds the robustness suite a sophisticated referee expects; it does not establish the primary identification (aeja-identification) or format the exhibits (aeja-tables-figures).

日本語の概要は準備中です。原文の説明を表示しています。

Use when articulating what an Accounting, Organizations and Society (AOS) manuscript contributes — converting findings into a claim that changes how the field understands accounting's social, organizational, behavioral or institutional operation. Frames the payoff; it does not produce new analysis (aos-data-analysis).

日本語の概要は準備中です。原文の説明を表示しています。

Use when strengthening AAMAS reproducibility evidence for interaction claims, including proofs and game definitions, opponent and population sets, self-play protocols, random seeds, compute, uncertainty on strategic outcomes, baselines, and consistency between what the paper claims about agents and what the artifact can actually show.

日本語の概要は準備中です。原文の説明を表示しています。

Use when preparing AAMAS supplementary material - proofs, extended game and mechanism definitions, extra multiagent experiments, opponent-set details, and the anonymized artifact zip - under the size cap, anonymity, and reviewer-discretion limits, including how to split a game-theory-plus-experiments paper across the 8-page body and the supplement.

日本語の概要は準備中です。原文の説明を表示しています。