Leeroo open-sources Kapso for enterprise data experiments
Authors claim first place on RelBench and a top open-source rank on OpenAI’s MLE-bench at submission.
7 stories, newest first · one card per story
Authors claim first place on RelBench and a top open-source rank on OpenAI’s MLE-bench at submission.
Finetuned from Qwen3.5-4B, it matched held-out terminal output on 77.9% of programs versus 41.1% for larger prompted models.
35B MLExpert lifts MLE-Bench Lite medal rate from 37.9% to 72.7%, near larger frontier models.
SwarmRSI reports state-of-the-art recursive self-improvement for scientific agent swarms on three research benchmarks.
Humanize MLE Agents alternates two agents and hits 78.2% any-medal on 75 MLE-bench Kaggle tasks.
TabFM-Auto uses an agent to evolve data pipelines and leads TabArena and MLE-Bench tabular results.
AREX-2 trains a Qwen3.8-27B agent on long-horizon reflective tasks, lifting research and coding benchmarks.
Today 763 readers · 835 views · 81 reads · all time 215K views