Pith. sign in

REVIEW

MindFlow+: A Self-Evolving Agent for E-Commerce Customer Service

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2507.18884 v1 pith:7F36AYDP submitted 2025-07-25 cs.CL

MindFlow+: A Self-Evolving Agent for E-Commerce Customer Service

classification cs.CL
keywords dialoguelearningmindflowe-commerceagentcombiningcustomerinteractions
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

High-quality dialogue is crucial for e-commerce customer service, yet traditional intent-based systems struggle with dynamic, multi-turn interactions. We present MindFlow+, a self-evolving dialogue agent that learns domain-specific behavior by combining large language models (LLMs) with imitation learning and offline reinforcement learning (RL). MindFlow+ introduces two data-centric mechanisms to guide learning: tool-augmented demonstration construction, which exposes the model to knowledge-enhanced and agentic (ReAct-style) interactions for effective tool use; and reward-conditioned data modeling, which aligns responses with task-specific goals using reward signals. To evaluate the model's role in response generation, we introduce the AI Contribution Ratio, a novel metric quantifying AI involvement in dialogue. Experiments on real-world e-commerce conversations show that MindFlow+ outperforms strong baselines in contextual relevance, flexibility, and task accuracy. These results demonstrate the potential of combining LLMs tool reasoning, and reward-guided learning to build domain-specialized, context-aware dialogue systems.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.