Results for “forced-alignment”

Alignment

Embeds target and source image parquets, then mines nearest-neighbour source images for augmentation in VCN AOI workflows.

LoRA: Low-Rank Adaptation of Large Language Models

Skeptical pushback on visual placement — both density and quality. Reads the annotated outline plus the visuals manifest and asks (a) whether the article hits the density target from editorial-principles-visuals.md, (b) whether each [VISUAL:...] earns its place, (c) whether sections without one would benefit. One revision pass on FAIL (BLOG_AGENT_VISUALS_REVISION_BUDGET, default 1).

Visual Instruction Tuning

Fine-tune and align language models using reinforcement learning with TRL, including SFT, DPO, PPO, GRPO, and reward model training.

Matryoshka Representation Learning

CogVLM: Visual Expert for Pretrained Language Models

Rewrites free-form draft prompts into the conventions Claude Opus 4.6 or 4.8 responds best to, either inline or by overwriting a draft file.

Improve layout, spacing, and visual rhythm. Fixes monotonous grids, inconsistent spacing, and weak visual hierarchy. Use when the user mentions layout feeling off, spacing issues, visual hierarchy, crowded UI, alignment problems, or wanting better composition.

AutoAugment: Learning Augmentation Strategies from Data

Mantis: Interleaved Multi-Image Instruction Tuning

Validate whether an OSS tool is worth building through staged namespace, existing-solution, and feasibility review

Amend approved research artifacts through a bounded alignment-gated workflow

Fine-tune Cosmos Reason video QA models using supervised fine-tuning with FSDP parallelism, including dataset preparation, spec construction, and AutoML support.

Chameleon: Mixed-Modal Early-Fusion Foundation Models

Use this when a codebase or design effort needs consistency enforcement for tokens, spacing, typography, color usage, component variants, and reusable interaction patterns.

Analyze and automatically remediate MoonSpec artifact inconsistencies across spec.md, plan.md, tasks.md, and related design files. Use when the user asks to run `/moonspec.align`, identify uncertainty, weigh tradeoffs in project context, edit artifacts without asking follow-up questions, resolve coverage gaps, or align generated MoonSpec documents before implementation.

Diffusion-based molecular docking. Predict protein-ligand binding poses from PDB/SMILES, confidence scores, virtual screening, for structure-based drug design. Not for affinity prediction.

Generates award-level UI with GSAP motion, bento grids, and Python-driven randomization for layout variety.

Diffusion-based molecular docking. Predict protein-ligand binding poses from PDB/SMILES, confidence scores, virtual screening, for structure-based drug design. Not for affinity prediction.

Use when performing multiple sequence alignment of FASTA inputs, generating alignment ensembles, or calculating alignment confidence metrics.

Builds reusable specialized skills for creating research-paper figures from lawful source material, then uses those skills to design and render figures for target papers.

Hard Negative Mixing for Contrastive Learning

Modify a classroom task with language scaffolds that preserve cognitive demand for EAL learners. Use when adapting existing tasks for students at different English proficiency levels.

Generates structured VGL JSON for Bria FIBO models, giving deterministic control over objects, lighting, camera, composition, and style instead of natural language prompts.

Strategic patterns for solving intractable problems through cascading approximation, self-improvement, and heterogeneous evaluation from DeepMind's AlphaGo system

Perform nonlinear dimensionality reduction, 2D/3D embeddings, clustering preprocessing, supervised or semi-supervised UMAP, DensMAP, AlignedUMAP, and Parametric UMAP workflows using the umap-learn library.

Improved Baselines with Visual Instruction Tuning

Predict 3D binding poses of small molecule ligands to protein targets using diffusion-based molecular docking, supporting single complexes, batch processing, and virtual screening.

Multi-phase dialectical stress-test for HIGH-STAKES decisions only (architecture choices, irreversible product calls, strategic bets). Heavy-cost skill — do NOT use for routine questions, brainstorming, or simple tradeoffs. User must explicitly invoke or describe a decision they call "high-stakes", "irreversible", or "needs stress-testing". Triggers: "stress test this decision", "dialectic on", "challenge this thesis", "should I really".

当寻求个人行动或项目在更宏大系统中的意义，或需要将日常工作与长远价值连接时

融合分类系统与范式锻造能力，将不同领域的分类体系融合成新的分类范式。

Push the LLM to reconsider, refine, and improve its recent output. Use when user asks for deeper critique or mentions a known deeper critique method, e.g. socratic, first principles, pre-mortem, red team.

Pose-conditioned generation on RunComfy via the `runcomfy` CLI. Routes across Kling 2-6 Motion Control Pro / Standard (transfer the motion / blocking of a reference video onto a target character), community Wan 2-2 Animate (audio-driven character animation with pose conditioning), and Z-Image Turbo ControlNet LoRA (pose-conditioned image generation from an OpenPose / DWPose / canny / depth control image). Picks the right route based on video vs still and stylized vs photoreal. Triggers on "controlnet", "control net", "pose control", "openpose", "DWPose", "transfer pose", "motion control", "pose driven", "character pose", "depth control", "canny edge", "use this pose", or any explicit ask to condition generation on a pose / skeleton / motion / depth / canny reference.