armin-aptura/skilltrainbench-public
skilltrainbench-public: skilltrainbench training tasks: Harbor dataset on Hugging Face with 733 tasks. The training half of the skilltrainbench benchmark suite: for each of the four datasets, the dev task names of its pinned train/test split, in Harbor task format.
Tasks
- 13F Amendment-Aware Crowding, Overlap, and Turnover
- Alpha Signal Construction with Factor Exposure Analysis
- Arithmetic Asian Options: Exact, Levy, and Curran Approximations
- Binance BTC Participation TCA
- Black-Litterman with Regime-Switching HMM
- Brinson-Fachler Sector Attribution
- Cliquet (Ratchet) Option Pricing
- CME HDD Option Pricing: Burn Analysis, OU Model, and Greeks
- Compound Options Pricing with Geske Formula
- Copula Fitting for Equity Pairs
- Copula Sampling and Rank Correlation Analysis
- Task: Corporate Action Price Adjustment
- Credit Rating Migration Matrix Analysis
- Portfolio Credit VaR with Copula Default Correlation
- Credit Spread Decomposition with HAC Inference and Variance Attribution
- CreditMetrics Portfolio Credit VaR
- Crypto Perpetual Funding Rate Economics & Basis Carry Analysis
- Digital Options & Barrier-Style Binary Pricing
- Task: Betting Against Beta × Momentum Corner Portfolio
- Task: Earnings Surprise and SUE Calculator
- ETF Cross-Asset Lead-Lag Discovery
- ETF overlap-aware redemption pressure attribution
- Task: Earnings Event Study — Abnormal Returns and CAAR
- Tail-Risk VaR/ES with EVT and GARCH-EVT
- EWMA Portfolio Risk Decomposition — Debug and Fix
- Aggregate Loss Distribution via FFT-Based Characteristic Function Inversion
- First Passage Time & Running Extrema of Geometric Brownian Motion
- FOMC Tone Surprise and Treasury Yield Event Study
- Cross-sectional insider sale-pressure reconstruction from Form 4 filings
- FX Carry Trade with Dealer-Style Forward and Option Hedges
- FX Forward Portfolio Valuation and Risk Analysis
- Geometric Mean-Reverting Jump-Diffusion Model
- Hull-White Swaption Pricing: Calibration, Jamshidian Decomposition, and Trinomial Tree
- Objective
- LOB PC Signal: Weighted OFI, PCA, and Rolling Regression
- Local-Vol Surface Cleaning and Barrier Pricing
- Lookback Option Pricing and Monte Carlo Validation
- Monte Carlo Greeks: Pathwise, Likelihood Ratio, and Finite Difference Methods
- USD-Collateralized MTM GBP/USD Cross-Currency Basis Swap
- Hard Multimodal Quantamental Alpha with Committee Disagreement and Process Filtering
- OHLC Realized Volatility Estimators
- Polars API 0.x to 1.x Migration
- Task: Multi-Frequency Realized Volatility with Microstructure-Noise Correction
- Regime-Aware CTA Strategy Analysis
- Eigenvalue Regime Detection with Risk-Parity and CVaR
- Task: Residual Momentum vs. Raw Momentum
- SEC 10-K Report Information and Fundamental Metric Extraction
- 8-K Event Alpha
- Task: Stable Residual Alpha with Beta-Neutral, Turnover-Capped Execution
- Implied Volatility Surface Under a Two-Factor Heston Model
- Structured Note Valuation and Risk
- USD Swap Curve Bootstrap (OIS + 3M LIBOR) — Debug and Fix
- VaR and ES Estimation Methods
- Task: Yield Curve Bootstrapping, Key Rate Durations, and Portfolio Immunization
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
- One hard-variant HealthBench behavioural task scored by the pinned official verifier-side LLM grader.
583 more tasks.