Technology Category
Cognitive Modeling & Cognitive Systems: Conceptual Inference and ReasoningKnowledge Representation and Reasoning: Computational Complexity of ReasoningNatural Language Processing: Sentence-level Semantics, Textual Inference, etc.
The Price of Meaning: Why Every Semantic Memory System Forgets
🤖 AI Summary
This work rigorously demonstrates, from geometric and dimensional perspectives, that semantic generality inherently entails memory fragility, revealing an intrinsic trade-off between the two. Semantic memory systems inevitably suffer from interference, forgetting, and erroneous recall when supporting generalization and analogy. The authors establish four theoretical results characterizing this mechanism through inner-product semantic space modeling, local intrinsic dimensionality analysis, δ-convexity verification, and power-law reachability statistics. Empirical validation across five mainstream architectures shows that purely semantic systems exhibit direct forgetting; incorporating reasoning mitigates this issue but introduces catastrophic failure modes; and entirely avoiding interference necessarily compromises generalization capacity.
Technology Category
Application Category
📝 Abstract
Every major AI memory system in production today organises information by meaning. That organisation enables generalisation, analogy, and conceptual retrieval -- but it comes at a price. We prove that the same geometric structure enabling semantic generalisation makes interference, forgetting, and false recall inescapable. We formalise this tradeoff for \textit{semantically continuous kernel-threshold memories}: systems whose retrieval score is a monotone function of an inner product in a semantic feature space with finite local intrinsic dimension.
Within this class we derive four results: (1) semantically useful representations have finite effective rank; (2) finite local dimension implies positive competitor mass in retrieval neighbourhoods; (3) under growing memory, retention decays to zero, yielding power-law forgetting curves under power-law arrival statistics; (4) for associative lures satisfying a $δ$-convexity condition, false recall cannot be eliminated by threshold tuning.
We test these predictions across five architectures: vector retrieval, graph memory, attention-based context, BM25 filesystem retrieval, and parametric memory. Pure semantic systems express the vulnerability directly as forgetting and false recall. Reasoning-augmented systems partially override these symptoms but convert graceful degradation into catastrophic failure. Systems that escape interference entirely do so by sacrificing semantic generalisation.
The price of meaning is interference, and no architecture we tested avoids paying it.
Problem
Research questions and friction points this paper is trying to address.
semantic memory
interference
forgetting
false recall
generalisation
Innovation
Methods, ideas, or system contributions that make the work stand out.
semantic memory
interference
forgetting
kernel-threshold memory
semantic generalisation
🔎 Similar Papers
No similar papers found.
