technique
error-aware specialization objective
techniqueactiveprovisional
error-aware-specialization-objective-0a4f89f3·1 events·first seen 20d agoAliases: error-aware specialization objective
Co-occurring entities
More like this (12)
format-aware task prototypesPreference-Aware Rubric Learningweakly supervised anomaly detectionlanguage-aware adapter headsLayer-Adaptive Expert PruningSkillOptEvolveNav: Proactive Preflection and Self-Evolving Memory for Zero-Shot Object Goal Navigationtask-agnostic generationLanguage Model Safety MonitorReference-Augmented TrainingUncertainty-aware Multi-Granularity RAGpredictor-corrector methods
Recent events (1)
LearnWeak: Automated Domain Specialization for Small Computer-Use Agents via Weakness-Targeted Synthesis
LearnWeak is an annotation-free framework for specializing small computer-use agents (CUAs) in specific software domains without deploying large expert models. It uses a stronger reference agent to identify weaknesses in a smaller student agent, synthesizes targeted tasks, and applies an error-aware training objective that disentangles planning from execution errors. On OSWorld, LearnWeak achieves gains of ~11 percentage points over 7B-8B baseline CUAs across eight domains. The work demonstrates that student-aware data synthesis substantially outperforms naive large-scale data generation for domain specialization.