Search results

Prompt caching is unused: the system prompt is re-sent at full price on every LLM call

Status: Open.

#163 In HolobiomicsLab/Mimosa-AI;

· lfnothias opened on Jun 29, 2026

Agent context grows unbounded on long runs (no compaction)

Status: Open.

#161 In HolobiomicsLab/Mimosa-AI;

· lfnothias opened on Jun 29, 2026

Add workflow admission cards before similarity-based template reuse

Status: Open.

#154 In HolobiomicsLab/Mimosa-AI;

· sunghunkwag opened on Jun 12, 2026

Evaluation with ARC-Bench ?

Status: Open.

#148 In HolobiomicsLab/Mimosa-AI;

· Fosowl opened on May 28, 2026

Evaluation pipeline for Metabolomics using AgenticScienceBuilder

Status: Open.

#147 In HolobiomicsLab/Mimosa-AI;

· Fosowl opened on May 12, 2026

Idea: Prompt "gradient" based on workflow past patterns

Status: Open.

#131 In HolobiomicsLab/Mimosa-AI;

· Fosowl opened on Mar 23, 2026

Investigating evaluation methodologies that yield informative judge feedback signals

Status: Open.

#120 In HolobiomicsLab/Mimosa-AI;

· Fosowl opened on Feb 27, 2026

baby Mimosa on GSM8K

Status: Open.

#118 In HolobiomicsLab/Mimosa-AI;

· Fosowl opened on Feb 27, 2026

Crossover / Mutation operator

Status: Open.

#117 In HolobiomicsLab/Mimosa-AI;

· Fosowl opened on Feb 27, 2026

Idea: shared variables across SmolAgents

Status: Open.

#114 In HolobiomicsLab/Mimosa-AI;

· Fosowl opened on Feb 13, 2026

Workflow mutation using more workflow as inspiration, lower similarity score to generalize pattern ?

Status: Open.

#112 In HolobiomicsLab/Mimosa-AI;

· Fosowl opened on Jan 28, 2026

orchestrator generating scenario file "on the flight" for workflow evaluation

Status: Open.

#111 In HolobiomicsLab/Mimosa-AI;

· Fosowl opened on Jan 28, 2026