# PALADIN: Self-Correcting Language Model Agents to Cure Tool-Failure Cases

arXiv: https://arxiv.org/abs/2509.25238 · HTML: https://mlsanity.com/paper/2509.25238/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- Placed in Tool learning and function calling, one of 16 directions in LLM agents and tool use.
- Builds on 29 of its 56 references that are in the map.
- First posted on arXiv 25 Sep 2025.

## Placed in

- LLM agents and tool use / [Tool learning and function calling](https://mlsanity.com/agents/tool-learning-and-function-calling/)

## Builds on (most cited first, 29 of 29)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [Tree of Thoughts: Deliberate Problem Solving with Large Language Models](https://mlsanity.com/paper/2305.10601/) (2023)
- [ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs](https://mlsanity.com/paper/2307.16789/) (2023)
- [Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference](https://mlsanity.com/paper/1712.05877/) (2017)
- [Gorilla: Large Language Model Connected with Massive APIs](https://mlsanity.com/paper/2305.15334/) (2023)
- [HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face](https://mlsanity.com/paper/2303.17580/) (2023)
- [A White Paper on Neural Network Quantization](https://mlsanity.com/paper/2106.08295/) (2021)
- [Tool Learning with Large Language Models: A Survey](https://mlsanity.com/paper/2405.17935/) (2024)
- [Evaluation and Benchmarking of LLM Agents: A Survey](https://mlsanity.com/paper/2507.21504/) (2025)
- [StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models](https://mlsanity.com/paper/2403.07714/) (2024)
- [Agentic Reasoning and Tool Integration for LLMs via Reinforcement Learning](https://mlsanity.com/paper/2505.01441/) (2025)
- [Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems](https://mlsanity.com/paper/2408.16293/) (2024)
- [AGILE: A Novel Reinforcement Learning Framework of LLM Agents](https://mlsanity.com/paper/2405.14751/) (2024)
- [m&m's: A Benchmark to Evaluate Tool-Use for multi-step multi-modal Tasks](https://mlsanity.com/paper/2403.11085/) (2024)
- [Advancing Tool-Augmented Large Language Models: Integrating Insights from Errors in Inference Trees](https://mlsanity.com/paper/2406.07115/) (2024)
- [ALoRA: Allocating Low-Rank Adaptation for Fine-tuning Large Language Models](https://mlsanity.com/paper/2403.16187/) (2024)
- [ToolScan: A Benchmark for Characterizing Errors in Tool-Use LLMs](https://mlsanity.com/paper/2411.13547/) (2024)
- [AEGIS: An Agent-based Framework for General Bug Reproduction from Issue Descriptions](https://mlsanity.com/paper/2411.18015/) (2024)
- [SHIELDA: Structured Handling of Exceptions in LLM-Driven Agentic Workflows](https://mlsanity.com/paper/2508.07935/) (2025)
- [Multi-Mission Tool Bench: Assessing the Robustness of LLM based Agents through Related and Dynamic Missions](https://mlsanity.com/paper/2504.02623/) (2025)
- [Towards Agentic Runtime Healing](https://mlsanity.com/paper/2408.01055/) (2024)
- [iTool: Reinforced Fine-Tuning with Dynamic Deficiency Calibration for Advanced Tool Use](https://mlsanity.com/paper/2501.09766/) (2025)
- [BugGen: A Self-Correcting Multi-Agent LLM Pipeline for Realistic RTL Bug Synthesis](https://mlsanity.com/paper/2506.10501/) (2025)
- [Enhancing Multi-Agent Coordination through Common Operating Picture Integration](https://mlsanity.com/paper/2311.04740/) (2023)

## Built on by (most cited first, 22 of 22)

- [AgentNoiseBench: Benchmarking Robustness of Tool-Using LLM Agents Under Noisy Condition](https://mlsanity.com/paper/2602.11348/) (2026)
- [OpenSearch-VL: An Open Recipe for Frontier Multimodal Search Agents](https://mlsanity.com/paper/2605.05185/) (2026)
- [When Tools Fail: Benchmarking Dynamic Replanning and Anomaly Recovery in LLM Agents](https://mlsanity.com/paper/2606.05806/) (2026)
- [Agentic Artificial Intelligence (AI): Architectures, Taxonomies, and Evaluation of Large Language Model Agents](https://mlsanity.com/paper/2601.12560/) (2026)
- [Robust Tool Use via Fission-GRPO: Learning to Recover from Execution Errors](https://mlsanity.com/paper/2601.15625/) (2026)
- [DART: Semantic Recoverability for Structured Tool Agents](https://mlsanity.com/paper/2605.23311/) (2026)
- [Can LLMs Clean Up Your Mess? A Survey of Application-Ready Data Preparation with LLMs](https://mlsanity.com/paper/2601.17058/) (2026)
- [Learning to Act under Noise: Enhancing Agent Robustness via Noisy Environments](https://mlsanity.com/paper/2605.27209/) (2026)
- [Safeguarding LLM Agents from Misalignment through Provenance Analysis](https://mlsanity.com/paper/2607.01236/) (2026)
- [Where Does Exactly-Once Live? Model, Harness, and Tool-Contract Effects on Duplicate Side Effects in LLM Agents](https://mlsanity.com/paper/2609.29095/) (2026)
- [RIVA: Leveraging LLM Agents for Reliable Configuration Drift Detection](https://mlsanity.com/paper/2603.02345/) (2026)
- [Retry, Switch, or Abstain? Learning Strategy-Aware Tool-Use Policies via Controlled Error Injection](https://mlsanity.com/paper/2608.11977/) (2026)
- [Outcome Monitors: Recovery Affordances for Silent Tool Failures](https://mlsanity.com/paper/2608.19303/) (2026)
- [The Workload-Router-Pool Architecture for LLM Inference Optimization: A Vision Paper from the vLLM Semantic Router Project](https://mlsanity.com/paper/2603.21354/) (2026)
- [Bounded Autonomy for Enterprise AI: Typed Action Contracts and Consumer-Side Execution](https://mlsanity.com/paper/2604.14723/) (2026)
- [From Cognitive Architectures to Language Agents: A Mechanism-Level Review of Lineage, Convergence, and Migration Gaps](https://mlsanity.com/paper/2607.23942/) (2026)
- [CAST: Critique-Aware Supervision for Training Reliable Long-Horizon Tool-Calling Agents](https://mlsanity.com/paper/2608.30147/) (2026)
- [When the Database Fails: Prompting LLM Dialogue Agents for Safe Recovery in Task-Oriented Dialogue](https://mlsanity.com/paper/2606.31307/) (2026)
- [What Stops a Small Language Model From Driving a Database Agent](https://mlsanity.com/paper/2609.21341/) (2026)
- [When Tools Silently Lie: Evaluating and Mitigating Blind Compliance in Tool-Augmented Data Agents](https://mlsanity.com/paper/2609.37153/) (2026)
- [Understanding and Mitigating Hallucination Escape in Tool-Using LLM Agents](https://mlsanity.com/paper/2610.04409/) (2026)
- [Loud Failures, Quiet Failures: Fault Detection and Recovery in Tool-Using Language Model Agents](https://mlsanity.com/paper/2610.10062/) (2026)

## Abstract (arXiv)

Tool-augmented language agents frequently fail in real-world deployment due to tool malfunctions--timeouts, API exceptions, or inconsistent outputs--triggering cascading reasoning errors and task abandonment. Existing agent training pipelines optimize only for success trajectories, failing to expose models to the tool failures that dominate real-world usage. We propose PALADIN, a generalizable framework for equipping language agents with robust failure recovery capabilities. PALADIN trains on 50,000+ recovery-annotated trajectories constructed via systematic failure injection and expert demonstrations on an enhanced ToolBench dataset. Training uses LoRA-based fine-tuning to retain base capabilities while injecting recovery competence. At inference, PALADIN detects execution-time errors and retrieves the most similar case from a curated bank of 55+ failure exemplars aligned with ToolScan's taxonomy, then executes the corresponding recovery action. This approach generalizes to novel failures beyond the training distribution, retaining 95.2% recovery performance on unseen tool APIs. Evaluation across PaladinEval and ToolReflectEval demonstrates consistent improvements in Recovery Rate (RR), Task Success Rate (TSR), Catastrophic Success Rate (CSR), and Efficiency Score (ES). PALADIN improves RR from 32.76% to 89.68% (+57% relative) over ToolBench and outperforms the strongest baseline CRITIC (76.34%) by +13.3%. Against vanilla agents, PALADIN achieves 89.86% RR (+66% relative improvement from 23.75%). These results establish PALADIN as an effective method for building fault-tolerant agents capable of robust recovery in real-world tool environments.