AI Wiki — Knowledge Base

Can Small Language Models Handle Context-Summarized Multi-Turn Customer-Service QA? A Synthetic Data-Driven Comparative Evaluation

arXiv:2602.00665v3 Announce Type: replace Abstract: Customer-service question answering (QA) systems increasingly rely on conversational language understanding. While Large Language Models (LLMs) achi

safetyarxiv-cs-cl

4 May 2026

Search: “customer-service”

Can Small Language Models Handle Context-Summarized Multi-Turn Customer-Service QA? A Synthetic Data-Driven Comparative Evaluation

When Should Service Agents Reconsider? Difficulty-Routed Control in Customer-Service Operations

Staged Linguistic Seeding: Grounded Query Expansion for Verified-Unit QA in AI Contact Centers

PolicyGuide: From Guarding One Action to Guiding the Whole Workflow for Policy-Compliant LLM Agents

Declarative Skills for AI Agents in Knowledge-Grounded Tool-Use Workflows

Back in Style: A Sociolinguistic Approach to Authoring and Measuring Persona Fidelity in User Simulation

Clarify, Then Focus: Statement Normalization for Conversation Analytics at Scale

Constitutional Value Potentials: reading and steering internal priority margins in language models

Greed Is Learned: Visible Incentives as Reward-Hacking Triggers

Asking for What Was Never Requested: Horizontal and Vertical Proactivity in Agents

Frozen Judges, Moving Agents: Version-Dependent LLM-Judge Error and the Limits of Judge-Assisted Agent Evaluation

Beyond Surface Style: Aligning Multi-Turn User Simulators with Behavioral Consistency

au^au-Bench: An Environment for End-To-End, Realistic Agent Construction

Knowledge-Verified Emergent Deception in LLM Agents Under Conflicting Incentives

Robustness Analysis of Agentic AI to Inconsistent and Incomplete Tool Responses

Business Arena: Benchmarking LLM Agents in a Realistic Marketplace

TRACE: Capability-Targeted Agentic Training

Partial Identification under Missing Data Using Weak Shadow Variables from Pretrained Models

Insuring Every Action: An Authority Frontier Framework for Runtime Actuarial Control of Autonomous AI Agents

RouteNLP: Closed-Loop LLM Routing with Conformal Cascading and Distillation Co-Optimization