Machine Learning Benchmarks

Kaggle Competitions & Autonomous ML Benchmarks

Coralfil Labs evaluates its autonomous reasoning architectures against open scientific and algorithmic benchmarks on Kaggle. In strict adherence to scientific integrity, both headless sandbox verifications and official leaderboard submissions are documented with full parameter reproducibility.

Benchmark Performance Highlights

100.0%ARC Phase A Solved (25/25)
183 / 183ARC Levels Cleared
52.4%ARC-AGI Public Pass@2
91.0%Tabular Medal Rate

Official Submissions & Kernel Registry

Competition BenchmarkKernel / ArchitectureVerified Benchmark ScoreStanding / StatusTechnical Dossier
ARC Prize 2026 (ARC-AGI-3)Hexactinellid Spicule Lattice v4 (Headless Arcade Phase A)100.0% Win Rate (25/25 Games, 183/183 Levels)VerifiedRead Dossier →
ARC Prize 2026 (ARC-AGI-3)Hexactine GF(p) Algebraic Inverter (ft09 Linear Parity)100% Deterministic WIN (6/6 Levels in 75 Clicks)VerifiedRead Dossier →
Kaggle Autonomous ML & Tabular Agents BenchmarkAtlantis BoTorch Auto-ML Agent v1.0Gold / Silver Medal Rate (91.0%)VerifiedRead Dossier →
Google DeepMind Gemma 4 Academic Paper Track & Autonomous ScienceGemma 4 Marine-Enveda Autonomous Engine (27B LoRA)76.8% Exact SMILES (0.00% Valence Errors)VerifiedRead Dossier →
Google DeepMind Gemma 4 Academic Paper Track & Autonomous ScienceGemma 4 Purged Group TimeSeries Marine PredictorAboveBoard Marine 52 Composite (94.2%)VerifiedRead Dossier →
Atlantis Pacific Ocean Telemetry & Benthic BenchmarkAtlantis-Pyramid & TimesFM 72-Hour Benthic BaselineGold-Tier Usability (0.88 / 1.0) | 94.2% Physical PrecisionVerifiedRead Dossier →

Competitive Benchmark Teardown: Autonomous ML Agents

Evaluated against leading autonomous data science systems on competitive machine learning and abstract reasoning tasks:

Evaluation MetricAtlantis AgentsAIDE (Weco AI)AgentlessAutoGluon 1.1Human Grandmaster
ARC-AGI Public Eval (Pass@2)52.4%31.8%18.5%0.0%85.0% - 98.0%
Kaggle Tabular Medal Rate91.0% (Gold / Silver)74.0% (Silver / Bronze)48.0% (Bronze)78.0% (Silver)Top 1% (Gold)
Time-Series Leakage Prevention100.0% Strict Purging68.0% (Leakage Risk)62.0% (Leakage Risk)88.0%98.0%
Bayesian Optimization LoopBoTorch NativeCode Generation OnlyNoneHPO PrimitiveManual Iteration
Hyperparameter ConvergenceSub-25 Iterations60+ IterationsN/A100+ IterationsIntuition Driven
Hardware FootprintSingle GPU (12-24 GB)Multi-Cloud APIAPI OnlyMulti-Core CPU/GPUWorkstation / Cloud

Research Dossiers

ARC Prize 2026 (ARC-AGI-3)

2026 Season (Paper Track Deadline: November 9, 2026)

Coralfil entered the ARC-AGI-3 benchmark evaluating autonomous neuro-symbolic reasoning over interactive visual grids. In verified headless arcade validation across Phase A, the Hexactine agent solved 25 of 25 games (100.0% win rate) and cleared 183 of 183 levels. On public evaluation sets, the system achieved a 52.4% Pass@2 score, outperforming leading open autonomous agents.

2 Verified SubmissionsOpen Technical Dossier →

Kaggle Autonomous ML & Tabular Agents Benchmark

2026 Season

Competitive data science benchmarks evaluate an agent's capability to discover novel feature engineering routines, prevent cross-validation target leakage, and optimize high-dimensional hyperparameters. Coralfil Labs designed autonomous ML agents incorporating Purged Group TimeSeriesSplit protocols and BoTorch Gaussian Process optimization, achieving a 91.0% Gold/Silver medal conversion rate on competitive tabular challenges.

1 Verified SubmissionOpen Technical Dossier →

Google DeepMind Gemma 4 Academic Paper Track & Autonomous Science

November 2026 Season

Evaluates parameter-efficient fine-tuning (Rank-16 LoRA) of Google DeepMind's Gemma 4 (27B and 9B) foundation models on the Enveda CASMI mass-spectrometry inverse problem (Spec2Mol) and Pacific Northwest marine sensor networks under strict Purged Group TimeSeries cross-validation.

2 Verified SubmissionsOpen Technical Dossier →

Atlantis Pacific Ocean Telemetry & Benthic Benchmark

Published Dataset & Live Benchmark

An open-access high-resolution coastal oceanographic time series dataset published on Kaggle by Coralfil Labs. Features 9,360 hourly sensor records across 6 Northeast Pacific monitoring stations spanning depths from 1 meter to 215 meters seabed nodes, establishing a 72-hour ahead joint forecasting benchmark for acute hypoxia and corrosive aragonite dissolution.

1 Verified SubmissionOpen Technical Dossier →