# FailureMem: A Failure-Aware Multimodal Framework for Autonomous Software Repair

arXiv: https://arxiv.org/abs/2603.17826 · HTML: https://mlsanity.com/paper/2603.17826/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 36 references that are in the map.
- First posted on arXiv 18 Mar 2026.

## Builds on (most cited first, 20 of 20)

- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [SWE-bench Multimodal: Do AI Systems Generalize to Visual Software Domains?](https://mlsanity.com/paper/2410.03859/) (2024)
- [SWE-Search: Enhancing Software Agents with Monte Carlo Tree Search and Iterative Refinement](https://mlsanity.com/paper/2410.20285/) (2024)
- [Keep the Conversation Going: Fixing 162 out of 337 bugs for 0.42 each using ChatGPT](https://mlsanity.com/paper/2304.00385/) (2023)
- [MarsCode Agent: AI-native Automated Bug Fixing](https://mlsanity.com/paper/2409.00899/) (2024)
- [Alibaba LingmaAgent: Improving Automated Issue Resolution via Comprehensive Repository Exploration](https://mlsanity.com/paper/2406.01422/) (2024)
- [How Effective Are Neural Networks for Fixing Security Vulnerabilities](https://mlsanity.com/paper/2305.18607/) (2023)
- [RepairLLaMA: Efficient Representations and Fine-Tuned Adapters for Program Repair](https://mlsanity.com/paper/2312.15698/) (2023)
- [RAP-Gen: Retrieval-Augmented Patch Generation with CodeT5 for Automatic Program Repair](https://mlsanity.com/paper/2309.06057/) (2023)
- [ScreenCoder: Advancing Visual-to-Code Generation for Front-End Automation via Modular Multimodal Agents](https://mlsanity.com/paper/2507.22827/) (2025)
- [GAMMA: Revisiting Template-based Automated Program Repair via Mask Prediction](https://mlsanity.com/paper/2309.09308/) (2023)
- [Domain Knowledge Matters: Improving Prompts with Fix Templates for Repairing Python Type Errors](https://mlsanity.com/paper/2306.01394/) (2023)
- [How Far Can We Go with Practical Function-Level Program Repair?](https://mlsanity.com/paper/2404.12833/) (2024)
- [CREF: An LLM-based Conversational Software Repair Framework for Programming Tutors](https://mlsanity.com/paper/2406.13972/) (2024)
- [Exploring Reasoning Reward Model for Agents](https://mlsanity.com/paper/2601.22154/) (2026)
- [MORepair: Teaching LLMs to Repair Code via Multi-Objective Fine-tuning](https://mlsanity.com/paper/2404.12636/) (2024)
- [Aligning the Objective of LLM-based Program Repair](https://mlsanity.com/paper/2404.08877/) (2024)
- [Enhancing Automated Program Repair with Solution Design](https://mlsanity.com/paper/2408.12056/) (2024)
- [PredicateFix: Repairing Static Analysis Alerts with Bridging Predicates](https://mlsanity.com/paper/2503.12205/) (2025)
- [Automated and Context-Aware Repair of Color-Related Accessibility Issues for Android Apps](https://mlsanity.com/paper/2308.09029/) (2023)

## Built on by (most cited first, 2 of 2)

- [MM-IssueLoc: A Controlled Benchmark for Evaluating Visual Evidence in Multimodal Repository-Level Issue Localization](https://mlsanity.com/paper/2607.15205/) (2026)
- [SWE-PolyVision: Benchmarking Cross-Image Abductive Reasoning for Repository-Level Software Engineering](https://mlsanity.com/paper/2609.29754/) (2026)

## Abstract (arXiv)

Multimodal Automated Program Repair (MAPR) extends traditional program repair by requiring models to jointly reason over source code, textual issue descriptions, and visual artifacts such as GUI screenshots. While recent LLM-based repair systems have shown promising results, existing approaches face several limitations: rigid workflow pipelines restrict exploration during debugging, visual reasoning is often performed over full-page screenshots without localized grounding, and failed repair attempts are rarely transformed into reusable knowledge. To address these challenges, we propose FailureMem, a multimodal repair framework that integrates three key mechanisms: a hybrid workflow-agent architecture that balances structured localization with flexible reasoning, active perception tools that enable region-level visual grounding, and a Failure Memory Bank that converts past repair attempts into reusable guidance. Experiments on SWE-bench Multimodal demonstrate FailureMem improves the resolved rate over GUIRepair by 3.7%.