agent/papers/items/2026-2607-18039-evidence-in-the-loop-trace-driven-optimization-for-customer-service-llm-agents.md

Paper: Evidence-in-the-Loop: Trace-Driven Optimization for Customer-Service LLM Agents

---

type: paper title: "Evidence-in-the-Loop: Trace-Driven Optimization for Customer-Service LLM Agents" authors: Chunming Wu, Dafei Qiu, Congde Yuan, Charles Quan, Jun Wu, Suipeng Li, Mo Wu, Gavin Xie, et al. year: 2026 venue: arXiv url: https://arxiv.org/abs/2607.18039 code_url: source: arxiv collected_at: 2026-07-27 published_at: 2026-07-20 updated_at: 2026-07-20 status: queued relevance: high topics:

• agent-evaluation
• agent-safety
• computer-use
• memory
• rag
• tool-use
• workflow-agent methods:

benchmarks:

models:

datasets:

• cs.IR related_concepts:

related_jobs:

related_experiments:

related_projects:

collection_score: 20 collection_queries: agentic-ai, llm-agent, rag-agent

One-line Takeaway

Auto-collected from arXiv because it matched the Agent collection queries. Needs human skim.

Why Collected

• matched queries: agentic-ai, llm-agent, rag-agent
• inferred topics: agent-evaluation, agent-safety, computer-use, memory, rag, tool-use, workflow-agent
• arXiv categories: cs.IR
• collection score: 20

Review Checklist

• Does this paper directly inform Agent architecture, evaluation, memory, tools, safety, coding agents, GUI/browser agents, or multi-agent workflows?
• Does it include a benchmark, dataset, code, or reproducible experimental setup?
• Should it be promoted from queued to skimmed or summarized?

Links

• arXiv: https://arxiv.org/abs/2607.18039