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

TL;DR

TabAgent introduces a framework that replaces slow, costly generative decision components in agentic systems with efficient, trained tabular-textual classifiers, significantly reducing latency and inference costs while maintaining performance.

Contribution

The paper presents a novel framework, TabAgent, for replacing generative decision modules with trained classifiers, improving efficiency in agentic systems.

Findings

Reduces latency by approximately 95%.

Cuts inference cost by 85-91%.

Maintains task success on AppWorld benchmark.

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.