LazyAgent is presented, a unified execution framework for agent-authored programs organized around a live, goal-derived demanded set that consistently outperforms the strongest goal-stopping eager baseline by refusing unrelated work before it starts and motivates benchmarks built from standing programs and sequences.
Abstract
Current agent runtimes that plan before acting generally execute a step once it becomes ready. We present LazyAgent, a unified execution framework for agent-authored programs organized around a live, goal-derived demanded set. LazyAgent refreshes a backward closure from requested outputs as execution state changes and materializes a ready node only when the active goal requires it. This replaces repeated local judgments with one linear-time graph analysis followed by constant-time membership tests, allowing programs to remain broad while execution stays request-specific. On programs that describe more than the current request needs, LazyAgent consistently outperforms the strongest goal-stopping eager baseline by refusing unrelated work before it starts. Adding one unrelated product raises the eager bill by 22.5% and LazyAgent's by 0.0%. LazyAgent saves 42.0% of measured CPU on production scientific workflows and 51.7% of container time on a live release gate spanning four repositories. We also prove and verify exact equivalence when the request reaches the whole graph, leaving no unrelated work to avoid. Beyond permission, goal-relative output projection saves up to approximately 90% of a shared step on two third-party test suites while the identical eager control saves 0.0%; the advantage disappears when the omitted output has no other consumer or the request needs it. Ordering, reuse, and pruning can also save cost, but do not replace permission. Finally, we show that current public benchmarks are eager-shaped and contain almost no unrequested work. A pre-registered planning intervention did not broaden them. These findings motivate benchmarks built from standing programs and sequences.
DynBranch is proposed, which makes an unresolved branch addressable before it resolves, and its stable coordinate lets candidate subgraphs run during resolution and completed subgraph results be reused across later requests.
Jun-Yi Shen, Noppanat Wadlom, Zheng-Yuan Su et al.· 0 citations
This work presents AgentEvolver, a system for developing capabilities during task execution while keeping the foundation model fixed, and evaluates task outcomes on SWE-bench Pro Public and examines capability changes in six application cases.
Wen-Tao Zhang, Fu-Chao Yang, Yi-Lei Zhao et al.· 0 citations
The results suggest that scientific-agent capability depends not only on the base model but on whether an architecture can be formed to match the task's cognitive structure, as well as establishing results on regret, planning invalidation, amortization, subtree interfaces, serializability, and verification.
TomasuLLM is presented, a runtime that executes agent tool calls out of trajectory order while preserving task-execution correctness, and improves the reported benchmark means and scales with tool latency.
Jiang-Nan Yu, Ce-Yu Xu, Meng-Ming Li et al.· arXiv.org· 0 citations
Long-horizon agentic workflows require models to sustain repeated state-dependent actions all while the context grows, sub-task complexity changes, and new data arrives. Each situation represents an independent axis along which an agent may fail. An agent reconciling a long ledger, for example, must repeatedly read its...
Jeffrey Willette, Krishna C. Puvvada, Boris Ginsburg· arXiv.org· 0 citations
The Adaptive Consistency Graph (ACG) is introduced, which incrementally organizes execution evidence and its provenance in a persistent graph, then constructs a temporary requirement-centered view for each decision under a bounded context budget.