Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents
Large language model (LLM) agents can induce skills from completed tasks and reuse them later to grow more capable with experience. In practice, induced skills may transfer unreliably and can even harm the agent that retrieves them. When agent-induced skills transfer reliably across tasks remains an open question. We conduct a comprehensive and controlled study of how the way skills are induced shapes their transfer across tasks. Specifically, we compare task-level with subtask-level skill induction and text with code skill formats, the two axes along which existing methods differ. Task-level
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 52%Getting better at coding doesn't make a model better at everything else →
- LinkedLinked via arxiv author · 85%Yiyang Feng →
“Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents”
- LinkedLinked via arxiv author · 85%Biddut Sarker Bijoy →
“Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents”
- LinkedLinked via arxiv author · 85%Niranjan Balasubramanian →
“Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents”
- LinkedLinked via arxiv author · 85%Jiawei Zhou →
“Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents”
- FuzzySimilar title/name (fuzzy) · 59%jnMetaCode/agency-agents-zh →
“Fuzzy title match (0.73): “Break It Down, Pass It On: Cross-Task Skill Transfer in LLM ” ≈ “jnMetaCode/agency-agents-zh””
- FuzzySimilar title/name (fuzzy) · 87%NirDiamant/GenAI_Agents →
“Fuzzy title match (0.94): “Break It Down, Pass It On: Cross-Task Skill Transfer in LLM ” ≈ “NirDiamant/GenAI_Agents””
- FuzzySimilar title/name (fuzzy) · 84%Unity-Technologies/ml-agents →
“Fuzzy title match (0.92): “Break It Down, Pass It On: Cross-Task Skill Transfer in LLM ” ≈ “Unity-Technologies/ml-agents””