arXiv cs.AI· Leann Alhashishi, Christos Margadji, Sebastian William Pattinson·· 13 小时前AI 评分31
inr2action：将机器人片段编码为神经隐式表示以支持回忆与动作
inr2action: Encoding Robot Episodes as Neural Implicit Representations for Recall and Action
AI 导读
inr2action 提出参数空间记忆框架，把每个机器人片段拟合为神经隐式表示，其权重增量即为记忆，大小与片段时长无关。在 RoboMME 上，语言模型在所有题型上超过随机水平，解码器平均 R² 达 0.51；将权重增量替换为等量噪声后性能下降 11.4 分，随机初始权重则无法解码。
正文 · 原文
Abstract:Long-horizon robotic operation requires memory that scales with essential information rather than execution time. We present a parameter-space memory framework where each episode is stored as a separable object. A neural implicit representation is fitted to episode video from a shared, meta-learned initialization; the resulting weight delta constitutes the memory. Crucially, this delta remains a fixed size regardless of whether the episode spans a hundred or a hundred thousand frames. A learned projector maps the weight delta into token embeddings, conditioning a language model for question answering and a decoder for joint command regression, without frame retrieval at query time. On RoboMME, the language model exceeds chance across all question types, and the decoder achieves a mean $R^2$ of 0.51. Replacing the delta with matched noise drops performance by 11.4 points, while random-origin weights are non-decodable, proving the meta-learned initialization is essential for embedding episode content. By representing individual experiences as fixed-size weight deltas in a common parameter space, this approach decouples memory cost from time horizon and provides a basis for scalable streaming memory.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.28882 [cs.RO] |
| (or arXiv:2610.28882v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.28882 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Christos Margadji [view email]
[v1]
Tue, 20 Oct 2026 20:00:50 UTC (1,840 KB)
来源：arXiv cs.AI · arxiv.org