mlsanityData to 14 Oct 2026

Self-Evolving Multimedia Verification through Memory Consolidation of Contestation Experiences

arXiv 2609.27175 ↗First posted 23 Sep 2026cs.MM cs.AI cs.MAcited by 0

Not on the map. Why?

Abstract (arXiv)

Multimedia verification requires not only accurate decisions but also traceable evidence, reliable human correction, and safe reuse of prior experience. Existing systems often lack explicit mechanisms for revising intermediate reasoning or preventing harmful knowledge transfer. We present SEMV (Self-Evolving Multimedia Verification), a self-evolving multi-agent framework that treats provenance-bearing arguments as the interface between evidence, reasoning, human contestation, and memory. SEMV combines arena-based quantitative bipolar argumentation (A-QBAF), causal and scoped revision, and verification-gated memory consolidation with explicit conflict retention. On COSMOS benchmark, SEMV achieves 91.88% accuracy versus 89.10% for the strongest comparable baseline. Verified memory reduces negative transfer from 5.7% to 0.2%. On CTR benchmark, constructed from reviewer contestations, scoped causal revision corrects 96.7% of initial errors while saving 52.8% compute. MV2026 Grand Challenge dataset further supports evidence-grounded, temporally consistent reporting. These results show that SEMV can evolve through verified experience while keeping accumulated knowledge and subsequent decisions traceable, revisable, and contestable.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 32 of its 62 references that are in the map. First posted on arXiv 23 Sep 2026.

Builds on

32 of 62 references.

References · 32

  1. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
  2. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,510
  3. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,063
  4. 2023ExpeL: LLM Agents Are Experiential Learnerscited by 876
  5. 2024LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memorycited by 707
  6. 2026Gemma 4 Technical Reportcited by 259
  7. 2022End-to-End Multimodal Fact-Checking and Explanation Generation: A Challenging Dataset and Modelscited by 107
  8. 2024Symbolic Learning Enables Self-Evolving Agentscited by 106
  9. 2025How Memory Management Impacts LLM Agents: An Empirical Study of Experience-Following Behaviorcited by 101
  10. 2024Agent-Pro: Learning to Evolve via Policy-Level Reflection and Optimizationcited by 94
  11. 2021Open-Domain, Content-based, Multi-modal Fact-checking of Out-of-Context Images via Online Resourcescited by 71
  12. 2020SBERT-WK: A Sentence Embedding Method by Dissecting BERT-based Word Modelscited by 56
20 more
  1. 2024VLM Agents Generate Their Own Memories: Distilling Experience into Embodied Programs of Thoughtcited by 55
  2. 2024Self-evolving Agents with reflective and memory-augmented abilitiescited by 52
  3. 2025Contextual Experience Replay for Self-Improvement of Language Agentscited by 48
  4. 2024DEFAME: Dynamic Evidence-based FAct-checking with Multimodal Expertscited by 43
  5. 2024Argumentative Large Language Models for Explainable and Contestable Claim Verificationcited by 36
  6. 2019EmbraceNet: A robust deep learning architecture for multimodal classificationcited by 30
  7. 2026Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligencecited by 30
  8. 2023Interactive Explanations by Conflict Resolution via Argumentative Exchangescited by 24
  9. 2026Useful Memories Become Faulty When Continuously Updated by LLMscited by 20
  10. 2024MetaReflection: Learning Instructions for Language Agents using Past Reflectionscited by 19
  11. 2023RED-DOT: Multimodal Fact-checking via Relevant Evidence Detectioncited by 18
  12. 2023Cheap-fake Detection with LLM using Prompt Engineeringcited by 15
  13. 2024Contestable AI needs Computational Argumentationcited by 13
  14. 2025ArgRAG: Explainable Retrieval Augmented Generation using Quantitative Bipolar Argumentationcited by 12
  15. 2026RecMem: Recurrence-based Memory Consolidation for Efficient and Effective Long-Running LLM Agentscited by 8
  16. 2025Multimedia Verification Through Multi-Agent Deep Research Multimodal Large Language Modelscited by 4
  17. 2026Argumentation for Explainable and Globally Contestable Decision Support with LLMscited by 3
  18. 2025Fact-Checking at Scale: Multimodal AI for Authenticity and Context Verification in Online Mediacited by 2
  19. 2026Strength Change Explanations in Quantitative Argumentationcited by 2
  20. 2026Contestable Multi-Agent Debate with Arena-based Argumentative Computation for Multimedia Verificationcited by 2

Built on by

Not cited yet.