Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.10182 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917400469831680 |
|---|---|
| author | Zhou, Lingfeng Shi, Junhao Gao, Jin Wang, Dequan |
| author_facet | Zhou, Lingfeng Shi, Junhao Gao, Jin Wang, Dequan |
| contents | Current evaluations of autonomous coding agents assume an unrealistic, infinite-resource environment. However, real-world software engineering is a resource-bound competition. As we scale toward large agent swarms, ignoring compute and time costs risks catastrophic budget exhaustion. To shift the focus from isolated accuracy to cost-aware problem-solving, we introduce USACOArena, an interactive ACM-ICPC-style arena driven by a strict "credit" economy. Every generated token, local test, and elapsed second depletes a fixed budget, forcing agents to make strategic trade-offs. Our comprehensive profiling reveals that frontier single agents and swarms currently fail to optimally balance accuracy with these constraints, exhibiting divergent, path-dependent behaviors. Ultimately, USACOArena provides an essential dynamic training ground for developing highly efficient, resource-aware agent architectures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_10182 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Credit-Budgeted ICPC-Style Coding: When Agents Must Pay for Every Decision Zhou, Lingfeng Shi, Junhao Gao, Jin Wang, Dequan Artificial Intelligence Current evaluations of autonomous coding agents assume an unrealistic, infinite-resource environment. However, real-world software engineering is a resource-bound competition. As we scale toward large agent swarms, ignoring compute and time costs risks catastrophic budget exhaustion. To shift the focus from isolated accuracy to cost-aware problem-solving, we introduce USACOArena, an interactive ACM-ICPC-style arena driven by a strict "credit" economy. Every generated token, local test, and elapsed second depletes a fixed budget, forcing agents to make strategic trade-offs. Our comprehensive profiling reveals that frontier single agents and swarms currently fail to optimally balance accuracy with these constraints, exhibiting divergent, path-dependent behaviors. Ultimately, USACOArena provides an essential dynamic training ground for developing highly efficient, resource-aware agent architectures. |
| title | Credit-Budgeted ICPC-Style Coding: When Agents Must Pay for Every Decision |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2604.10182 |