{"ID":2828138,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.15829","arxiv_id":"2512.15829","title":"Physics-driven human-like working memory outperforms digital networks in dynamic vision","abstract":"While the unsustainable energy cost of artificial intelligence necessitates physics-driven computing, its performance superiority over full-precision GPUs remains a challenge. We bridge this gap by repurposing the Joule-heating relaxation dynamics of magnetic tunnel junctions, conventionally suppressed as noise, into neuronal intrinsic plasticity, realizing working memory with human-like features. Traditional AI utilizes energy-intensive digital memory that accumulates historical noise in dynamic environments. Conversely, our Intrinsic Plasticity Network (IPNet) leverages thermodynamic dissipation as a temporal filter. We provide direct system-level evidence that this physics-driven memory yields an 18x error reduction compared to spatiotemporal convolutional models in dynamic vision tasks, reducing memory-energy overhead by \u003e90,000x. In autonomous driving, IPNet reduces prediction errors by 12.4% versus recurrent networks. This establishes a neuromorphic paradigm that shatters efficiency limits and surpasses conventional algorithmic performance.","short_abstract":"While the unsustainable energy cost of artificial intelligence necessitates physics-driven computing, its performance superiority over full-precision GPUs remains a challenge. We bridge this gap by repurposing the Joule-heating relaxation dynamics of magnetic tunnel junctions, conventionally suppressed as noise, into n...","url_abs":"https://arxiv.org/abs/2512.15829","url_pdf":"https://arxiv.org/pdf/2512.15829v3","authors":"[\"Jingli Liu\",\"Huannan Zheng\",\"Bohao Zou\",\"Kezhou Yang\"]","published":"2025-12-17T17:24:37Z","proceeding":"cs.ET","tasks":"[\"cs.ET\",\"cs.AI\",\"cs.CV\",\"cs.NE\"]","methods":"[]","has_code":false}