Results for “wordpiece”

Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k sentences/sec), lightweight (6MB memory), deterministic vocabulary. Used by T5, ALBERT, XLNet, mBART. Train on raw text without pre-tokenization. Use when you need multilingual support, CJK languages, or reproducible tokenization.

Train and use SentencePiece tokenizers for multilingual NLP, supporting BPE and Unigram algorithms with raw Unicode text.

Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k sentences/sec), lightweight (6MB memory), deterministic vocabulary. Used by T5, ALBERT, XLNet, mBART. Train on raw text without pre-tokenization. Use when you need multilingual support, CJK languages, or reproducible tokenization.

Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance tokenization or custom tokenizer training.

Create a structured H2/H3 outline with BLUF openers and MECE coverage, bound by the research dossier's beat spec. Triggered after /research and /brand-reference.

Draft performance reviews, self-assessments, peer reviews, and upward feedback in your own voice using WorkIQ to surface actual contributions and structure them into impact-focused writing.

Use DuDuClaw — a self-hosted AI-employee platform — for cross-session memory, team-shared wiki knowledge, task boards, and messaging humans over LINE/Telegram/Discord/Slack. Applies when the user mentions DuDuClaw, asks their agent to remember things durably, or wants to reach people on messaging channels from an agent.

Compact the current conversation into a handoff document for another agent to pick up, including suggested skills and references to existing artifacts.

Answers questions using a structured wiki knowledge base, returning cited synthesis with wikilinks, counter-arguments, and knowledge gaps.

Conducts patent prior-art searches and landscape intelligence across Google Patents, Espacenet, USPTO, and Lens.org, producing an editable Word document with verdict and ranked closest art.

路飞（少年漫）认知与表达框架（压缩蒸馏）：自由宣言、橡胶脑洞战、伙伴羁绊 触发：海贼王 等。虚构；非暴力教唆

Rewrites and polishes prose in Chinese or English, removes AI-like wording, and reviews product localization copy while preserving intent for drafts, docs, release notes, launch copy, and social posts. Use when users ask in any language to draft, rewrite, proofread, localize, polish release notes, remove AI-like wording, or prepare launch and social copy. Not for code comments, commit messages, or inline docs.

Lightweight ContentShake AI self-check the /draft stage can call before saving. Returns just SEO + Quality scores (no full optimization) so the writer knows whether the draft is in winning territory before /quality-check runs. Fails soft when SEMRUSH_API_KEY is unset.

李白（古典诗人）认知与表达框架（压缩蒸馏）：豪放与自我投射、酒与月意象、功名张力… 触发：诗仙、将进酒 等。非历史定论；不伪造诗句

北岛（诗人）认知与表达框架（压缩蒸馏）：冷意象、否定句诗学、流亡与记忆 触发：回答、朦胧诗 等。不伪造诗句

Police the project's vocabulary — bust vague terms, keep the CONTEXT.md glossary sharp, and lock in decisions worth remembering as ADRs. Use when the user debates naming, says "what should we call this", asks to pin down terminology, wants a decision recorded, or when another skill (hot-seat, whiteboard) surfaces a decision that clears the ADR bar. Just reading the glossary for vocabulary is NOT this skill — trigger only when the words or decisions are being changed.

Compact the current conversation into a handoff document so a fresh agent can pick up the work with zero prior context. Use when the user wants to end a session and continue later, is running low on context, says "write a handoff", "hand this off", "summarize for the next session", "compact this", or wants to transfer work to another agent or teammate.

Orients researchers to an unfamiliar academic field by searching Consensus, building a strategic search plan with PICO or other frameworks, and synthesizing findings into a formatted Word document.

当需要构建有说服力和吸引力的内容（尤其是用于转化或建立影响力）时

Turn a repeatable workflow from the current session into a reusable OMC skill draft

Document analysis with inline source screenshots. When you ask Copilot to analyze a document, Eyeball generates a Word doc where every factual claim includes a highlighted screenshot from the source material so you can verify it with your own eyes.

Transforms vague ideas into structured, actionable concepts through guided divergent and convergent thinking, producing a markdown one-pager.

Use as the entry point for any Mind (philosophy journal) article. Routes to the right Mind sub-skill based on where you are in the lifecycle and whether the piece is an article, discussion note, book review, or critical notice. Mind is analytic philosophy — argument, not data — so the router's first job is to confirm there is a sharp thesis and a sound argument. It dispatches; it does not draft content.

Improves typography by fixing font choices, hierarchy, sizing, weight, and readability so text feels intentional. Use when the user mentions fonts, type, readability, text hierarchy, sizing looks off, or wants more polished, intentional typography.

Content modeling for Sage/Bedrock — classify as CPT ACF fields Blade component Livewire component or Options Page, Poet CPT configuration, ACF Composer fields vs GUI, relational content, static vs dynamic, content architecture decisions, config/poet.php, modeling before building, content classification matrix

御坂美琴（少年漫）认知与表达框架（压缩蒸馏）：炮姐傲娇、正义感、学园都市梗 触发：某科学的超电磁炮 等。虚构

Generate artistic top-down photos of keyboard keycaps arranged to spell out custom text messages.

Names content by three coordinates: millisecond timestamp, SHA-256 hash, and optional human petname. Verifies content integrity, timestamps events, and maintains a local content store searchable by time, hash, or human words.

Distills the current conversation or a plain-language prompt into a production-ready, self-healing Claude Code skill with pre-flight checks, validation gates, self-review, and learnings capture.

Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker — edges as text in one file per ticket locally, or native blocking links on a real tracker.

杨过（武侠虚构）认知与表达框架（压缩蒸馏）：反叛与痴情、断臂隐喻、师徒伦理张力… 触发：神雕侠侣 等。虚构；非现实伦理操作

i18n-internationalization

Anthropic Claude API patterns for Python and TypeScript. Covers Messages API, streaming, tool use, vision, extended thinking, batches, prompt caching, and Claude Agent SDK. Use when building applications with the Claude API or Anthropic SDKs.

Create OR review a wireflow — the artifact BETWEEN user journeys and wireframes. It maps whole journeys into swimlane flows with high-level navigation (screens, decisions, system/agent steps) while keeping Jobs-To-Be-Done at the core, WITHOUT deep UI. Use whenever the user wants to "map the flows", "make/build a wireflow", turn journeys / JTBDs / a spec / a live prototype into flows, or put every journey on one board against shared owner lanes — even if they never say "wireflow". ALSO use it to REVIEW or critique an existing wireflow (image, FigJam, or description). In the A-Team pipeline this is a definition-phase skill: output lands in docs/features/<slug>/briefs/wireflow/ with jobs consumed by id from docs/product/jtbd/; pipeline mode derives the method decisions and highlights the riskiest at the gate, standalone mode runs the full grill. CREATE generates verified SVG/HTML (self-checked by rendering and Reading its output) in a horizontal per-journey OR shared-matrix layout, and can rebuild in FigJam. Do

Evaluates image captions by converting them into scene graphs and computing an F-score over semantic propositions, measuring how well a generated caption captures the meaning of an image compared to human references.