aiqbench analysis
SkillDAG addresses the structural complexity of large skill libraries by modeling inter-skill dependencies as a typed directed graph, enabling agents to dynamically retrieve and evolve their knowledge base during execution. This approach outperforms fixed retrieval pipelines and embedding similarity methods by maintaining robust candidate ranking even as the skill pool scales tenfold. The method demonstrates significant gains in success rates and reward metrics across benchmarks, validating its utility for managing complex agent workflows without degrading performance.
SkillDAG: Self-Evolving Typed Skill Graphs for LLM Skill Selection at Scale
arXiv:2606.03056v1 Announce Type: new Abstract: As LLM agents adopt large skill libraries, selecting the right subset becomes a structural problem rather than a similarity-matching one: skills depend on, conflict with, specialize, or duplicate one another, a structure invisible to both full enumeration and embedding similarity. We present SkillDAG, which models inter-skill relationships as a typed directed graph and exposes it to an LLM agent as an inference-time, agent-callable structural retrieval interface, queried and evolved during execution rather than baked into a fixed retrieval pipeline: each search returns vector matches, typed-edge neighbors, and conflict signals, and a propose-then-commit protocol lets the agent register execution-backed edges so the graph accumulates structure across episodes. On ALFWorld and SkillsBench with MiniMax-M2.7, SkillDAG reaches 67.1% success and 27.3% reward, exceeding the strongest reported Graph-of-Skills baseline by +12.8 and +8.6 points; the advantage ports to gpt-5.2-codex, and intrinsic SkillsBench Ret@K rises from 65.5 to 78.2 under matched queries. These gains trace to isolable mechanisms: candidate ranking that stays robust as the pool grows 10x where a fixed seeding-diffusion pipeline degrades, and set-monotone online edits that enlarge ground-truth recall without evicting prior hits.
Abstract:As LLM agents adopt large skill libraries, selecting the right subset becomes a structural problem rather than a similarity-matching one: skills depend on, conflict with, specialize, or duplicate one another, a structure invisible to both full enumeration and embedding similarity. We present SkillDAG, which models inter-skill relationships as a typed directed graph and exposes it to an LLM agent as an inference-time, agent-callable structural retrieval interface, queried and evolved during execution rather than baked into a fixed retrieval pipeline: each search returns vector matches, typed-edge neighbors, and conflict signals, and a propose-then-commit protocol lets the agent register execution-backed edges so the graph accumulates structure across episodes. On ALFWorld and SkillsBench with MiniMax-M2.7, SkillDAG reaches 67.1% success and 27.3% reward, exceeding the strongest reported Graph-of-Skills baseline by +12.8 and +8.6 points; the advantage ports to gpt-5.2-codex, and intrinsic SkillsBench Ret@K rises from 65.5 to 78.2 under matched queries. These gains trace to isolable mechanisms: candidate ranking that stays robust as the pool grows 10x where a fixed seeding-diffusion pipeline degrades, and set-monotone online edits that enlarge ground-truth recall without evicting prior hits.
Artificial Intelligence (cs.AI)
arXiv:2606.03056 [cs.AI]
(or arXiv:2606.03056v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2606.03056 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 275 words
Related coverage
SkillDAG: Self-Evolving Typed Skill Graphs for LLM Skill Selection at Scale
arXiv · Oct 17, 2026
SE-GoS: Self-Evolving Graph-of-Skills for Skill Library at Scale
arXiv · Sep 9, 2026
EmbodiedRSI: Active Continual Robot Learning Through Hypothesis-Guided Co-Evolution
arXiv · Oct 8, 2026
AXRP Episode 50 - Eli Lifland on AI 2027
LessWrong · Aug 3, 2026