FAMA: Failure-Aware Meta-Agentic Framework for Open-Source LLMs in Interactive Tool Use Environments

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Saeidi, Amir, Mishra, Venkatesh, Mukhopadhyay, Souradeep, Liu, Gaowen, Payani, Ali, Srinivasa, Jayanth, Baral, Chitta
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913067349049344
author Saeidi, Amir
Mishra, Venkatesh
Mukhopadhyay, Souradeep
Liu, Gaowen
Payani, Ali
Srinivasa, Jayanth
Baral, Chitta
author_facet Saeidi, Amir
Mishra, Venkatesh
Mukhopadhyay, Souradeep
Liu, Gaowen
Payani, Ali
Srinivasa, Jayanth
Baral, Chitta
contents Large Language Models are being increasingly deployed as the decision-making core of autonomous agents capable of effecting change in external environments. Yet, in conversational benchmarks, which simulate real-world customer-centric issue resolution scenarios, these agents frequently fail due to the cascading effects of incorrect decision-making. These challenges are particularly pronounced for open-source LLMs with smaller parameter sizes, limited context windows, and constrained inference budgets, which contribute to increased error accumulation in agentic settings. To tackle these challenges, we present the Failure-Aware Meta-Agentic (FAMA) framework. FAMA operates in two stages: first, it analyzes failure trajectories from baseline agents to identify the most prevalent errors; second, it employs an orchestration mechanism that activates a minimal subset of specialized agents tailored to address these failures by injecting a targeted context for the tool-use agent before the decision-making step. Experiments across open-source LLMs demonstrate performance gains up to 27% across evaluation modes over standard baselines. These results highlight that targeted curation of context through specialized agents to address common failures is a valuable design principle for building reliable, multi-turn tool-use LLM agents that simulate real-world conversational scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2604_25135
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle FAMA: Failure-Aware Meta-Agentic Framework for Open-Source LLMs in Interactive Tool Use Environments
Saeidi, Amir
Mishra, Venkatesh
Mukhopadhyay, Souradeep
Liu, Gaowen
Payani, Ali
Srinivasa, Jayanth
Baral, Chitta
Computation and Language
Large Language Models are being increasingly deployed as the decision-making core of autonomous agents capable of effecting change in external environments. Yet, in conversational benchmarks, which simulate real-world customer-centric issue resolution scenarios, these agents frequently fail due to the cascading effects of incorrect decision-making. These challenges are particularly pronounced for open-source LLMs with smaller parameter sizes, limited context windows, and constrained inference budgets, which contribute to increased error accumulation in agentic settings. To tackle these challenges, we present the Failure-Aware Meta-Agentic (FAMA) framework. FAMA operates in two stages: first, it analyzes failure trajectories from baseline agents to identify the most prevalent errors; second, it employs an orchestration mechanism that activates a minimal subset of specialized agents tailored to address these failures by injecting a targeted context for the tool-use agent before the decision-making step. Experiments across open-source LLMs demonstrate performance gains up to 27% across evaluation modes over standard baselines. These results highlight that targeted curation of context through specialized agents to address common failures is a valuable design principle for building reliable, multi-turn tool-use LLM agents that simulate real-world conversational scenarios.
title FAMA: Failure-Aware Meta-Agentic Framework for Open-Source LLMs in Interactive Tool Use Environments
topic Computation and Language
url https://arxiv.org/abs/2604.25135