Results for “catalyst-classification”

Catalysis Based Testing Skill

Catalysis Design Expert Skill

Detect poisoned training data and backdoored models across the ML pipeline using statistical analysis, activation clustering, and spectral signatures.

Catalysis Based Design Skill

Multi-dimensional lead qualification scoring. Evaluates leads against BANT criteria, firmographic fit, behavioral signals, and intent indicators. Outputs qualified/disqualified verdict with detailed reasoning.

Applied Catalysis Design Skill

Train, evaluate, distill, quantize, export, and run inference for PyTorch-based TAO image classification models with support for multiple backbones.

Defines ideal customer profile filters, scores inbound and outbound leads, and builds a lightweight qualification rubric to sharpen pipeline focus for founder-led sales.

Analyzes Stockbee-style Day 1 Episodic Pivot candidates by scoring catalyst quality alongside price/volume confirmation, gap/range expansion, volume shock, neglect/revaluation context, liquidity, and risk to the EP-day low.

Cardinality and model family jointly determine the encoding.

Rank an existing set of context chunks by relevance, diversity, freshness, and utility. Use when retrieval has already produced candidates that must be scored or reranked; use context-retrieval when the source corpus still needs to be searched.

Hard Negative Mixing for Contrastive Learning

Cross-runtime 2-stage pipeline for Claude Code, Codex/OMX, and Gemini/Antigravity/OMA: trace causal hypotheses, inject evidence into deep-interview style requirements crystallization, then hand off to the right runtime planner/executor.

Grades and classifies evaluation batch results, applying exclusions, diagnosing failure modes, computing pass rates, and generating summary tables for papers.

Analyse a learning task for cognitive load problems and recommend specific design improvements. Use when tasks overwhelm students, instructions feel complex, or materials need simplifying.

Evaluates machine learning classifiers on their ability to distinguish signal from background in particle physics simulations, measuring how well algorithms rank signal events above background ones using the AUC metric.

Specific description with trigger conditions

Train or fine-tune sentence-transformers models for retrieval, similarity, clustering, classification, and reranking, with support for bi-encoders, cross-encoders, and sparse encoders.

Guides technology selection and implementation of AI and ML features in .NET 8+ applications using ML.NET, Microsoft.Extensions.AI, Microsoft Agent Framework, GitHub Copilot SDK, ONNX Runtime, and OllamaSharp.

Migrate React class component lifecycle methods (componentWillMount, componentWillReceiveProps, componentWillUpdate) to React 18.3.1 compliant patterns with decision trees and before/after code examples.

Train, evaluate, export, and run inference for pose classification models using ST-GCN on skeleton keypoint sequences.

Kinetics-400: A Large Video Understanding Dataset

Analyzes prompt injection, retrieval poisoning, memory contamination, planner drift, and tool-boundary abuse in agentic systems, mapping trust boundaries and proving exploit chains.

Post-execution skill crystallization and learning engine updates for WinDAGs. Runs after successful execution to update Thompson sampling parameters, track method quality, detect monster-barring, log near-miss events, and signal Kuhnian crises. Activate on "curator", "learning update", "skill crystallization", "Thompson sampling", "monster-barring", "near-miss", "Kuhnian crisis", "post-execution learning". NOT for pre-execution risk scanning (use windags-premortem), retrospective analysis (use windags-looking-back), or DAG construction (use windags-architect).

融合分类系统与逻辑抽象能力，构建基于逻辑的分类体系，支持逻辑驱动的分类需求和逻辑分析。

Systematically evaluates evidence to determine which threat actor is responsible for a cyber operation using the Diamond Model and Analysis of Competing Hypotheses.

Generate and rank theory-first or proof-oriented research ideas. Use when the user wants non-experimental research ideas, theoretical methods, proof programs, theorem candidates, impossibility results, convergence/sample-complexity analyses, or "analysis" variants of idea creation.

Demystifying CLIP Data

Predict biomolecular complexes with Chai-1, an open AlphaFold3-style model that folds multi-entity assemblies (proteins, ligands, nucleic acids) from a single typed FASTA — strong on antibody–antigen and protein–ligand complexes, with optional MSA and restraint inputs. Use when predicting an antibody–antigen complex, folding a mixed protein/ligand/nucleic-acid assembly described in one FASTA, or generating a complex with experimental restraints. For binding-affinity prediction or a ligand-focused co-fold prefer alterlab-boltz; for protein-only or protein–protein folding prefer alterlab-alphafold; to dock into a fixed receptor prefer alterlab-diffdock. Part of the AlterLab Academic Skills suite.

Quantize LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss using bitsandbytes. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers.

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.

Analyzes type design for encapsulation, invariant expression, usefulness, and enforcement, providing scores and improvement suggestions.

Applies medicinal-chemistry filters with the medchem library — drug-likeness rules (Lipinski, Veber), PAINS filters, structural alerts, and molecular complexity metrics for compound prioritization and library cleanup. Use when filtering or triaging a compound library, flagging PAINS or reactive groups, or assessing drug-likeness of candidate molecules. Part of the AlterLab Academic Skills suite.

Multi-tiered memory management (HOT/WARM/COLD) for context compaction. Invoke ONLY for explicit compaction events: post-`/compact` cleanup, MEMORY.md tier promotion, archive batch, or "trim my context". NOT for general recall (use deep-recall) or routine memory writes (use storage-router). Triggers: "compact memory", "promote to durable", "archive old context", "tier this".

SNLI-VE: Visual Entailment Dataset