Agent-Native Primitives improve real-world task resolution for coding agents
HEELO framework using eight primitive families raises Terminal-Bench 4.0 resolution from 37.3% to 55.5% and reduces task time on heelo.dev
3 stories, newest first · one card per story
HEELO framework using eight primitive families raises Terminal-Bench 4.0 resolution from 37.3% to 55.5% and reduces task time on heelo.dev
A new benchmark and self-distillation method diagnose and repair weak mathematical reasoning in LLMs.
A primitive-based diagnosis finds Discovery is the main bottleneck in LLM mathematical reasoning and shows it can be repaired.
Today 485 readers · 546 views · 176 reads · all time 187K views