mlsanityData to 7 Oct 2026

Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory

arXiv 2604.01007 ↗First posted 1 Apr 2026cs.AIcited by 13

Not on the map. Why?

Abstract (arXiv)

AI agents increasingly operate over extended time horizons, yet their ability to retain, organize, and recall multimodal experiences remains a critical bottleneck. Building effective lifelong memory requires navigating a vast design space spanning architecture, retrieval strategies, prompt engineering, and data pipelines; this space is too large and interconnected for manual exploration or traditional AutoML to explore effectively. We deploy an autonomous research pipeline to discover Omni-SimpleMem, a unified multimodal memory framework for lifelong AI agents. Starting from a na"ive baseline (F1=0.117 on LoCoMo), the pipeline autonomously executes ~50 experiments across two benchmarks, diagnosing failure modes, proposing architectural modifications, and repairing data pipeline bugs, all without human intervention in the inner loop. The resulting system achieves state-of-the-art on both benchmarks, improving F1 by +411% on LoCoMo (0.117→0.598) and +214% on Mem-Gallery (0.254→0.797) relative to the initial configurations. Critically, the most impactful discoveries are not hyperparameter adjustments: bug fixes (+175%), architectural changes (+44%), and prompt engineering (+188% on specific categories) each individually exceed the cumulative contribution of all hyperparameter tuning, demonstrating capabilities fundamentally beyond the reach of traditional AutoML. We provide a taxonomy of six discovery types and identify four properties that make multimodal memory particularly suited for autoresearch, offering guidance for applying autonomous research pipelines to other AI system domains. Code is available at this https://github.com/aiming-lab/SimpleMem.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 24 of its 36 references that are in the map. First posted on arXiv 1 Apr 2026.

Builds on

24 of 36 references.

References · 24

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  3. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,500
  4. 2017Billion-scale similarity search with GPUscited by 2,477
  5. 2018DARTS: Differentiable Architecture Searchcited by 1,782
  6. 2016Neural Architecture Search with Reinforcement Learningcited by 1,489
  7. 2021Improving language models by retrieving from trillions of tokenscited by 1,278
  8. 2024The AI Scientist: Towards Fully Automated Open-Ended Scientific Discoverycited by 1,002
  9. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,000
  10. 2024Evaluating Very Long-Term Conversational Memory of LLM Agentscited by 800
  11. 2025Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memorycited by 722
  12. 2024A Survey on the Memory Mechanism of Large Language Model based Agentscited by 622
12 more
  1. 2018BOHB: Robust and Efficient Hyperparameter Optimization at Scalecited by 466
  2. 2025The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Searchcited by 375
  3. 2023MLAgentBench: Evaluating Language Agents on Machine Learning Experimentationcited by 339
  4. 2024VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documentscited by 237
  5. 2026SimpleMem: Efficient Lifelong Memory for LLM Agentscited by 140
  6. 2025AI-Researcher: Autonomous Scientific Innovationcited by 88
  7. 2026Mem-Gallery: Benchmarking Multimodal Long-Term Conversational Memory for MLLM Agentscited by 49
  8. 2025Position: Multimodal Large Language Models Can Significantly Advance Scientific Reasoningcited by 33
  9. 2025MemVerse: Multimodal Memory for Lifelong Learning Agentscited by 31
  10. 2026Claudini: Autoresearch Discovers State-of-the-Art Adversarial Attack Algorithms for LLMscited by 13
  11. 2025A Survey of AI Scientistscited by 9
  12. 2026Bilevel Autoresearch: Meta-Autoresearching Itselfcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 13.

13 papers.

Sort:

By area:

  1. 2026From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanismscited by 19
  2. 2026EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agentscited by 11
  3. 2026ClawArena: Benchmarking AI Agents in Evolving Information Environmentscited by 10
  4. 2026MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memorycited by 10
  5. 2026SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memorycited by 5
  6. 2026Mem-W: Latent Memory-Native GUI Agentscited by 3
  7. 2026WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interactioncited by 3
  8. 2026Sandboxed Coding Agents are Competitive Omni-modal Task Solverscited by 2
  9. 2026Workflow Closure Is Not Scientific Closure in Auto-Research Systemscited by 1
  10. 2026Beyond Retrieval: Analytic Memory for Multimodal Agentscited by 1
  11. 2026MemTools: A Unified Research Framework for Interoperable Agent Memorycited by 0
  12. 2026EM^2Mem: Event-Centric Multimodal Memory for Large Language Modelscited by 0
1 more
  1. 2026MemoWM: How World Models Change What Agents Need to Remembercited by 0