TabAgent: A Framework for Replacing Agentic Generative Components with Tabular-Textual Classifiers

2026-02-18

AI Summary

This work addresses the high latency and inference costs incurred by existing agent systems that frequently invoke large language models (LLMs) for closed-set decision-making tasks. To overcome this limitation, the authors propose TabAgent, a novel framework that replaces generative decision modules with a structured table-text classifier. TabAgent leverages TabSchema to extract execution trajectory features, employs TabSynth to generate synthetic supervised data, and utilizes a lightweight TabHead classifier for efficient candidate scoring. Evaluated on the AppWorld benchmark, TabAgent maintains comparable task success rates while reducing latency by 95% and cutting inference costs by 85–91%. The approach demonstrates strong generalizability across diverse agent decision-making scenarios.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Adversarial Agents
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Agentic systems, AI architectures that autonomously execute multi-step workflows to achieve complex goals, are often built using repeated large language model (LLM) calls for closed-set decision tasks such as routing, shortlisting, gating, and verification. While convenient, this design makes deployments slow and expensive due to cumulative latency and token usage. We propose TabAgent, a framework for replacing generative decision components in closed-set selection tasks with a compact textual-tabular classifier trained on execution traces. TabAgent (i) extracts structured schema, state, and dependency features from trajectories (TabSchema), (ii) augments coverage with schema-aligned synthetic supervision (TabSynth), and (iii) scores candidates with a lightweight classifier (TabHead). On the long-horizon AppWorld benchmark, TabAgent maintains task-level success while eliminating shortlist-time LLM calls, reducing latency by approximately 95% and inference cost by 85-91%. Beyond tool shortlisting, TabAgent generalizes to other agentic decision heads, establishing a paradigm for learned discriminative replacements of generative bottlenecks in production agent architectures.

Problem

• agentic systems
• closed-set decision tasks
• large language model
• latency
• inference cost

Innovation

• TabAgent
• agentic systems
• tabular-textual classifier
• LLM replacement
• execution traces