The Emergent Symbolic Structure of Artificial Neural Networks

• URL: http://arxiv.org/abs/2608.29530v1
• Date: Sun, 30 Aug 2026 03:32:13 GMT
• Title: The Emergent Symbolic Structure of Artificial Neural Networks
• Abstract summary: We show that the internal representations of neural networks implicitly realize symbolic structure.<n>This finding holds for both small-scale neural networks trained to manipulate lists as well as large language models.<n>This work provides a potential way to reconcile longstanding symbolic conceptions of intelligence with the vector-based nature of modern AI.
• Score: 18.995004344545226
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Modern systems in artificial intelligence (AI) somehow excel in domains for which they seem poorly suited. Intelligence has traditionally been modeled as operating over structured combinations of symbols, such as logical formulas. However, the strongest modern AI systems are based on neural networks, which instead represent information in continuous vectors. Vectors seem inadequate for capturing the structure of language, logic, and other cognitive domains, yet neural networks achieve impressive performance in these areas. How do they do it? In this work, we propose a potential answer: Despite appearances, perhaps the internal representations of neural networks implicitly realize symbolic structure. In support of this hypothesis, we show that the vector representations of a variety of neural networks can be closely approximated with symbolic structures: we can replace the network's entire representation-generating process with a closed-form equation instantiating a symbolic structure, and the network's behavior remains largely unchanged. This finding holds for both small-scale neural networks trained to manipulate lists as well as large language models (LLMs) operating in four domains that are central in symbolic traditions: arithmetic, logic, computer code, and language. Further, our symbolic approximation allows us to modify an LLM's behavior in targeted ways via precise interventions on its internal representations, showing that the LLM's behavior is reliant on the symbolic structures we have identified. This work provides a potential way to reconcile longstanding symbolic conceptions of intelligence with the vector-based nature of modern AI.

Related papers

• Deriving Equivalent Symbol-Based Decision Models from Feedforward Neural Networks [0.0]
  Despite its rapid adoption, the opacity of AI systems poses significant challenges to trust and acceptance.<n>This work focuses on the derivation of symbolic models, such as decision trees, from feed-forward neural networks (FNNs)
  arXiv  Detail & Related papers  (2025-04-16T19:22:53Z)
• Coding schemes in neural networks learning classification tasks [52.22978725954347]
  We investigate fully-connected, wide neural networks learning classification tasks. We show that the networks acquire strong, data-dependent features. Surprisingly, the nature of the internal representations depends crucially on the neuronal nonlinearity.
  arXiv  Detail & Related papers  (2024-06-24T14:50:05Z)
• Semantic Loss Functions for Neuro-Symbolic Structured Prediction [74.18322585177832]
  We discuss the semantic loss, which injects knowledge about such structure, defined symbolically, into training. It is agnostic to the arrangement of the symbols, and depends only on the semantics expressed thereby. It can be combined with both discriminative and generative neural models.
  arXiv  Detail & Related papers  (2024-05-12T22:18:25Z)
• Symbol Correctness in Deep Neural Networks Containing Symbolic Layers [0.0]
  We formalize a high-level principle that can guide the design and analysis of NS-DNNs. We show that symbol correctness is a necessary property for NS-DNN explainability and transfer learning.
  arXiv  Detail & Related papers  (2024-02-06T03:33:50Z)
• An Algorithm to Train Unrestricted Sequential Discrete Morphological Neural Networks [0.0]
  We propose an algorithm to learn unrestricted sequential DMNN, whose architecture is given by the composition of general W-operators. We illustrate the algorithm in a practical example.
  arXiv  Detail & Related papers  (2023-10-06T20:55:05Z)
• Discrete, compositional, and symbolic representations through attractor dynamics [51.20712945239422]
  We introduce a novel neural systems model that integrates attractor dynamics with symbolic representations to model cognitive processes akin to the probabilistic language of thought (PLoT) Our model segments the continuous representational space into discrete basins, with attractor states corresponding to symbolic sequences, that reflect the semanticity and compositionality characteristic of symbolic systems through unsupervised learning, rather than relying on pre-defined primitives. This approach establishes a unified framework that integrates both symbolic and sub-symbolic processing through neural dynamics, a neuroplausible substrate with proven expressivity in AI, offering a more comprehensive model that mirrors the complex duality of cognitive operations
  arXiv  Detail & Related papers  (2023-10-03T05:40:56Z)
• LOGICSEG: Parsing Visual Semantics with Neural Logic Learning and Reasoning [73.98142349171552]
  LOGICSEG is a holistic visual semantic that integrates neural inductive learning and logic reasoning with both rich data and symbolic knowledge. During fuzzy logic-based continuous relaxation, logical formulae are grounded onto data and neural computational graphs, hence enabling logic-induced network training. These designs together make LOGICSEG a general and compact neural-logic machine that is readily integrated into existing segmentation models.
  arXiv  Detail & Related papers  (2023-09-24T05:43:19Z)
• Emergence of Symbols in Neural Networks for Semantic Understanding and Communication [8.156761369660096]
  We propose a solution to endow neural networks with the ability to create symbols, understand semantics, and achieve communication. SEA-net generates symbols that dynamically configure the network to perform specific tasks. These symbols capture compositional semantic information that allows the system to acquire new functions purely by symbolic manipulation or communication.
  arXiv  Detail & Related papers  (2023-04-13T10:13:00Z)
• Symbolic Synthesis of Neural Networks [0.0]
  I present Graph-basedally Synthesized Neural Networks (GSSNNs) GSSNNs are a form of neural network whose topology and parameters are informed by the output of a symbolic program. I demonstrate that by developing symbolic abstractions at a population level, I can elicit reliable patterns of improved generalization with small quantities of data known to contain local and discrete features.
  arXiv  Detail & Related papers  (2023-03-06T18:13:14Z)
• Emergence of Machine Language: Towards Symbolic Intelligence with Neural Networks [73.94290462239061]
  We propose to combine symbolism and connectionism principles by using neural networks to derive a discrete representation. By designing an interactive environment and task, we demonstrated that machines could generate a spontaneous, flexible, and semantic language.
  arXiv  Detail & Related papers  (2022-01-14T14:54:58Z)
• Dynamic Inference with Neural Interpreters [72.90231306252007]
  We present Neural Interpreters, an architecture that factorizes inference in a self-attention network as a system of modules. inputs to the model are routed through a sequence of functions in a way that is end-to-end learned. We show that Neural Interpreters perform on par with the vision transformer using fewer parameters, while being transferrable to a new task in a sample efficient manner.
  arXiv  Detail & Related papers  (2021-10-12T23:22:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.