Agent Skills: attention-variants-from-papers
Implement paper-defined attention variants by extracting mechanism invariants, mapping tensor shapes, preserving module interfaces, validating numerical behavior, and integrating the custom attention into an existing transformer stack.
ID: benchflow-ai/skillsbench/attention-variants-from-papers
1,819369
Install this agent skill to your local
Skill Files
The files that make up attention-variants-from-papers.
More agent skills
- browser-testingVERIFY your changes work. Measure CLS, detect theme flicker, test visual stability, check performance. Use BEFORE and AFTER making changes to confirm fixes. Includes ready-to-run scripts: measure-cls.ts, detect-flicker.ts
- conditioningData conditioning techniques for gravitational wave detector data. Use when preprocessing raw detector strain data before matched filtering, including high-pass filtering, resampling, removing filter wraparound artifacts, and estimating power spectral density (PSD). Works with PyCBC TimeSeries data.
- fuzzy-matchA toolkit for fuzzy string matching and data reconciliation. Useful for matching entity names (companies, people) across different datasets where spelling variations, typos, or formatting differences exist.
- manufacturing-failure-reason-codebook-normalizationThis skill should be considered when you need to normalize testing engineers' written defect reasons following the provided product codebooks. This skill will correct the typos, misused abbreviations, ambiguous descriptions, mixed Chinese-English text or misleading text and provide explanations. This skill will do segmentation, semantic matching, confidence calibration and station validation.
- gttsGoogle Text-to-Speech (gTTS) for converting text to audio. Use when creating audiobooks, podcasts, or speech synthesis from text. Handles long text by chunking at sentence boundaries and concatenating audio segments with pydub.
- csv-processingUse this skill when reading sensor data from CSV files, writing simulation results to CSV, processing time-series data with pandas, or handling missing values in datasets.