EvoX：将 Agent Harness 重新思考为运行时策略

View PDF HTML (experimental)

Abstract:Large language model (LLM) agents increasingly interact with tools and environments, yet plausible model decisions do not always translate into successful execution due to tool-call mismatches, interrupted workflows, and unsupported completion claims. We present EvoX, a Rust-based, multi-model agent runtime guided by The Second Policy, which views the harness as an Operational Policy complementing the foundation model's Semantic Policy. EvoX combines model-facing guidance with deterministic runtime control through three complementary mechanisms: Action Realization, Execution Continuity, and Outcome Grounding. To evaluate harnesses without suppressing their native behaviors, we introduce Contrast, a controlled and auditable framework integrating explicit fairness constraints, provider-bound request verification, and evidence-based adjudication. Across five benchmarks, six foundation models, and six agent harnesses, EvoX achieves the highest numerical success rate of 56.2% over 691 common task--model pairs, exceeding the average of the other five harnesses by approximately 3.0 percentage points. Its larger advantages on GLM-5.3 (+10.7 points) and Kimi K3 (+7.9 points) highlight the model-dependent effectiveness of harness design and the potential for cross-provider adaptability. Achieved without memory, evolution, or extensions, these results establish EvoX as a competitive multi-model agent runtime and underscore the importance of harness-level operational control in realizing foundation-model capabilities.

Comments:	15 pages, 5 figures. Technical report
Subjects:	Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
Cite as:	arXiv:2611.15006 [cs.AI]
	(or arXiv:2611.15006v1 [cs.AI] for this version)
	https://doi.org/10.48550/arXiv.2611.15006 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Zhang Xiao [view email]
[v1] Wed, 11 Nov 2026 04:05:54 UTC (2,485 KB)