2026-06-12原文2 min readUpdated: 2026-06-12

Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents

Compact language models face challenges beyond isolated function calling when using tools. Evoflux uses evolutionary search at inference time to repair executable tool workflows, raising execution feasibility from 3% to 17-24% on MCP-Bench tasks, outperforming SFT and DPO baselines.

SourcearXiv AIAuthor: Kushal Raj Bhandari, Ling Yue, Ching-Yun Ko, Dhaval Patel, Shaowu Pan, Pin-Yu Chen, Jianxi Gao

[2606.12674] Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents

[Submitted on 10 Jun 2026]

Title:Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents

View a PDF of the paper titled Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents, by Kushal Raj Bhandari and 5 other authors

View PDF

Abstract:Compact language models (LMs) reduce cost, latency, and deployment risk for tool agents. Yet MCP-style tool use requires more than isolated function calling: an agent must discover tools from live catalogs, satisfy schemas, preserve dependencies across intermediate outputs, and ground final responses in executed evidence. Small planners often generate plausible workflow graphs that fail under tool resolution, parameter validation, dependency tracking, or execution. We argue that this failure mode is poorly handled by small-corpus distillation. A few hundred teacher traces can teach workflow format, but rarely cover the recovery behavior needed to repair failed plans over changing tool catalogs. We introduce Evoflux, an inference-time evolutionary search method that treats compact tool use as the repair of executable tool workflows. It evolves typed workflow graphs through structured edits, execution feedback, adaptive intensity, meta-guided redesign, and diversity pruning. On held-out MCP-Bench tasks spanning live MCP servers and 250 tools, Evoflux raises execution feasibility from roughly 3% to 17-24% across small planners. In contrast, SFT and SFT+DPO on the same search-mined data match, underperform, or collapse below zero-shot performance; ReAct reaches higher peaks, but with higher variance and token cost. These results show that execution-grounded search is more reliable under scarce teacher-trace budgets.

Comments: Code is available at this https URL

Subjects:

Artificial Intelligence (cs.AI)

Cite as: arXiv:2606.12674 [cs.AI]

(or arXiv:2606.12674v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2606.12674

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Kushal Raj Bhandari [view email] [v1] Wed, 10 Jun 2026 21:01:06 UTC (2,172 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Evoflux: Inference-Time Evolution of Executable Tool Workflows for Compact Agents, by Kushal Raj Bhandari and 5 other authors

View PDF

TeX Source

view license

Current browse context:

cs.AI

new | recent | 2026-06

Change to browse by:

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)