Think-Verify-Revise: Neuro-Symbolic Visual Reasoning with Vision-Language Models and Dynamic Logic Tensor Networks

• URL: http://arxiv.org/abs/2609.05388v1
• Date: Fri, 04 Sep 2026 17:39:47 GMT
• Title: Think-Verify-Revise: Neuro-Symbolic Visual Reasoning with Vision-Language Models and Dynamic Logic Tensor Networks
• Abstract summary: Visual reasoning tasks require a system to jointly perceive visual content and apply formal relational constraints.<n>This paper proposes a Neuro-Symbolic (NeSy) framework that closes this gap by tightly coupling a Vision-Language Model (VLM)<n>The proposed method achieves scores matching or outperforming previous methods (PSL, LTN)
• Score: 41.54450067095476
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Visual reasoning tasks require a system to jointly perceive visual content and apply formal relational constraints---a combination that neither pure neural nor purely symbolic approaches handle well in isolation. This paper proposes a Neuro-Symbolic (NeSy) framework that closes this gap by tightly coupling a Vision-Language Model (VLM) for automatic First-Order Logic (FOL) rule induction with a Dynamic Logic Tensor Network (D-LTN) for differentiable rule verification, in a closed iterative feedback loop. The VLM receives a small set of labelled visual examples and proposes candidate FOL rules conforming to a strict grammar (Think); the D-LTN is automatically assembled from these rules at runtime and evaluates them grounding on CNN-produced visual embeddings (Verify); and verification failures are fed back to guide the VLM's next hypothesis (Revise). Evaluated on the ViSudo-PC benchmark across four visual domains (MNIST, EMNIST, KMNIST, FMNIST), the system induces valid Sudoku constraint rules using only three training examples as visual context. The proposed method achieves AUC scores matching or outperforming previous methods (NeuPSL, LTN), showing the potential for automatic rule discovery through VLM. Code is available at https://github.com/homayoun-afshari/nesy.

Related papers

• A Logical-Rule Autoencoder for Interpretable Recommendations [5.768273842489658]
  We propose a Logical-rule Interpretable Autoencoder (LIA) for collaborative filtering that is interpretable by design.<n>LIA introduces a learnable logical rule layer in which each rule neuron is equipped with a gate parameter.<n>LIA allows users to directly trace the decision process behind each recommendation.
  arXiv  Detail & Related papers  (2026-04-05T21:21:37Z)
• Learning Representations Through Contrastive Neural Model Checking [5.774786149181392]
  Contrastive Neural Model Checking (CNML) is a novel method that leverages the model checking task as a guiding signal for learning aligned representations.<n> CNML considerably outperforms both algorithmic and neural baselines in cross-modal and intra-modal settings.<n>These findings demonstrate that model checking can serve as an objective for learning representations for formal languages.
  arXiv  Detail & Related papers  (2025-10-02T09:52:47Z)
• A Constructive Framework for Nondeterministic Automata via Time-Shared, Depth-Unrolled Feedforward Networks [0.0]
  We present a formal and constructive simulation framework for nondeterministic finite automata (NFAs) using time-shared, depth-unrolled feedforward networks (TS-FFNs)<n>Our formulation symbolically encodes automaton states as binary vectors, transitions as sparse matrix transformations, and nondeterministic branching-including $varepsilon$-closures-as compositions of shared thresholded updates.
  arXiv  Detail & Related papers  (2025-05-30T01:18:35Z)
• Symbolic Rule Extraction from Attention-Guided Sparse Representations in Vision Transformers [1.3812010983144802]
  Recent neuro-symbolic approaches have successfully extracted symbolic rule-sets from CNN-based models to enhance interpretability.<n>We propose a framework for symbolic rule extraction from Vision Transformers (ViTs) by introducing a sparse concept layer inspired by Sparse Autoencoders (SAEs)<n>Our method achieves a 5.14% better classification accuracy than the standard ViT while enabling symbolic reasoning.
  arXiv  Detail & Related papers  (2025-05-10T19:45:15Z)
• Verification Learning: Make Unsupervised Neuro-Symbolic System Feasible [49.084647802326366]
  This paper introduces a novel learning paradigm, Verification Learning (VL), which transforms the label-based reasoning process in Nesy into a label-free verification process.<n>VL achieves excellent learning results solely by relying on unlabeled data and a function that verifies whether the current predictions conform to the rules.<n>Our theoretical analysis points out which tasks in Nesy systems can be completed without labels and explains why rules can replace infinite labels for some tasks, while for others the rules have no effect.
  arXiv  Detail & Related papers  (2025-03-17T08:28:58Z)
• Lyapunov-stable Neural Control for State and Output Feedback: A Novel Formulation [67.63756749551924]
  Learning-based neural network (NN) control policies have shown impressive empirical performance in a wide range of tasks in robotics and control. Lyapunov stability guarantees over the region-of-attraction (ROA) for NN controllers with nonlinear dynamical systems are challenging to obtain. We demonstrate a new framework for learning NN controllers together with Lyapunov certificates using fast empirical falsification and strategic regularizations.
  arXiv  Detail & Related papers  (2024-04-11T17:49:15Z)
• State space models can express n-gram languages [51.823427608117626]
  We build state space language models that can solve the next-word prediction task for languages generated from n-gram rules.<n>Our proof shows how SSMs can encode n-gram rules using new theoretical results on their capacity.<n>We conduct experiments with a small dataset generated from n-gram rules to show how our framework can be applied to SSMs and RNNs obtained through gradient-based optimization.
  arXiv  Detail & Related papers  (2023-06-20T10:41:23Z)
• NeSyFOLD: Neurosymbolic Framework for Interpretable Image Classification [1.3812010983144802]
  We present a novel framework called NeSyFOLD to create a neurosymbolic (NeSy) model for image classification tasks. A rule-based machine learning algorithm called FOLD-SE-M is used to derive the stratified answer set program. A justification for the predictions made by the NeSy model can be obtained using an ASP interpreter.
  arXiv  Detail & Related papers  (2023-01-30T05:08:05Z)
• PROTOtypical Logic Tensor Networks (PROTO-LTN) for Zero Shot Learning [2.236663830879273]
  Logic Networks (LTNs) are neuro-symbolic systems based on a differentiable, first-order logic grounded into a deep neural network. We focus here on the subsumption or textttisOfClass predicate, which is fundamental to encode most semantic image interpretation tasks. We propose a common textttisOfClass predicate, whose level of truth is a function of the distance between an object embedding and the corresponding class prototype.
  arXiv  Detail & Related papers  (2022-06-26T18:34:07Z)
• A Correspondence Variational Autoencoder for Unsupervised Acoustic Word Embeddings [50.524054820564395]
  We propose a new unsupervised model for mapping a variable-duration speech segment to a fixed-dimensional representation. The resulting acoustic word embeddings can form the basis of search, discovery, and indexing systems for low- and zero-resource languages.
  arXiv  Detail & Related papers  (2020-12-03T19:24:42Z)
• Logical Natural Language Generation from Open-Domain Tables [107.04385677577862]
  We propose a new task where a model is tasked with generating natural language statements that can be emphlogically entailed by the facts. To facilitate the study of the proposed logical NLG problem, we use the existing TabFact dataset citechen 2019tabfact featured with a wide range of logical/symbolic inferences. The new task poses challenges to the existing monotonic generation frameworks due to the mismatch between sequence order and logical order.
  arXiv  Detail & Related papers  (2020-04-22T06:03:10Z)

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.