SIGMA: Structure-Invariant Generative Molecular Alignment for Chemical Language Models via Autoregressive Contrastive Learning

• URL: http://arxiv.org/abs/2603.25062v1
• Date: Thu, 26 Mar 2026 05:55:17 GMT
• Title: SIGMA: Structure-Invariant Generative Molecular Alignment for Chemical Language Models via Autoregressive Contrastive Learning
• Abstract summary: We propose Structure-Invariant Generative Molecular Alignment ( SIGMA)<n>Rather than altering the linear representation, SIGMA enables the model to strictly recognize geometric symmetries.<n> Empirical evaluations on standard benchmarks demonstrate that SIGMA bridges the gap between sequence scalability and graph fidelity.
• Score: 8.06177122578894
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Linearized string representations serve as the foundation of scalable autoregressive molecular generation; however, they introduce a fundamental modality mismatch where a single molecular graph maps to multiple distinct sequences. This ambiguity leads to \textit{trajectory divergence}, where the latent representations of structurally equivalent partial graphs drift apart due to differences in linearization history. To resolve this without abandoning the efficient string formulation, we propose Structure-Invariant Generative Molecular Alignment (SIGMA). Rather than altering the linear representation, SIGMA enables the model to strictly recognize geometric symmetries via a token-level contrastive objective, which explicitly aligns the latent states of prefixes that share identical suffixes. Furthermore, we introduce Isomorphic Beam Search (IsoBeam) to eliminate isomorphic redundancy during inference by dynamically pruning equivalent paths. Empirical evaluations on standard benchmarks demonstrate that SIGMA bridges the gap between sequence scalability and graph fidelity, yielding superior sample efficiency and structural diversity in multi-parameter optimization compared to strong baselines.

Related papers

• Variational Learning for Insertion-based Generation [50.15012125719375]
  We introduce a probabilistic framework for learning insertion order in variable-length insertion models.<n>We propose the Insertion Process (IP), a generative model that jointly learns where to insert, what to insert, and when to terminate.
  arXiv  Detail & Related papers  (2026-06-01T11:59:46Z)
• PathAR: Structure-First Autoregressive Synthesis of Multimodal Pathology Images [51.428093790826814]
  We propose pathology Autorgressive modeling (PathAR) to factorize structure and appearance for modality-label-conditioned pathology generation.<n>PathAR stabilizes morphology under heterogeneous modality-specific appearances and enables spatially aligned image-mask pair generation.
  arXiv  Detail & Related papers  (2026-06-01T01:43:48Z)
• A Unified Framework for Structure-Aware Clustering and Heterogeneous Causal Graph Learning [2.8350191976016768]
  In complex systems, interactions variables are defined by dependency structures, often encoded as acyclic graphs.<n> dependency structures can vary across subjects, ignoring this structural bias and obscures subpopulation.<n>We propose a unified framework built upon Structural Equation Modeling (SEM) that learns cluster-specific dependency structures.<n>Our method recovers cluster-specific causal dependency structures with a high true positive rate and a low false discovery rate.
  arXiv  Detail & Related papers  (2026-05-19T03:48:54Z)
• Persistent Nonnegative Matrix Factorization via Multi-Scale Graph Regularization [5.524295668575852]
  We propose persistent nonnegative matrix factorization (pNMF) that produces a sequence of persistence-aligned embeddings rather than a single one.<n>We analyze the structural properties of the embeddings along the scale parameter and establish bounds on their increments between consecutive scales.<n>The resulting model defines a nontrivial solution path across scales, rather than a single factorization, which poses new computational challenges.
  arXiv  Detail & Related papers  (2026-02-26T02:23:50Z)
• Rethinking Diffusion Models with Symmetries through Canonicalization with Applications to Molecular Graph Generation [56.361076943802594]
  CanonFlow achieves state-of-the-art performance on the challenging GEOM-DRUG dataset, and the advantage remains large in few-step generation.
  arXiv  Detail & Related papers  (2026-02-16T18:58:55Z)
• A Study of Adaptive Modeling Towards Robust Generalization [14.00955228748485]
  We present a unified all-atom framework that grounds language reasoning in geometric information while adaptively scaling structural tokens.<n>Across diverse all-atom benchmarks, the proposed approach yields consistent gains in heterogeneous structure-grounded reasoning.
  arXiv  Detail & Related papers  (2026-02-02T20:35:44Z)
• Modality Alignment across Trees on Heterogeneous Hyperbolic Manifolds [49.95082206008502]
  Alignment across Trees is a method that constructs and aligns tree-like hierarchical features for both image and text modalities.<n>We introduce a semantic-aware visual feature extraction framework that applies a cross-attention mechanism to visual class tokens from intermediate Transformer layers.
  arXiv  Detail & Related papers  (2025-10-31T11:32:15Z)
• Histopathology Whole Slide Image Analysis with Heterogeneous Graph Representation Learning [78.49090351193269]
  We propose a novel graph-based framework to leverage the inter-relationships among different types of nuclei for WSI analysis. Specifically, we formulate the WSI as a heterogeneous graph with "nucleus-type" attribute to each node and a semantic attribute similarity to each edge. Our framework outperforms the state-of-the-art methods with considerable margins on various tasks.
  arXiv  Detail & Related papers  (2023-07-09T14:43:40Z)
• Staged trees and asymmetry-labeled DAGs [2.66269503676104]
  We introduce a minimal Bayesian network representation of the staged tree, which can be used to read conditional independences in an intuitive way. We also define a new labeled graph, termed asymmetry-labeled directed acyclic graph, whose edges are labeled to denote the type of dependence existing between any two random variables.
  arXiv  Detail & Related papers  (2021-08-04T12:20:47Z)
• Joint Network Topology Inference via Structured Fusion Regularization [70.30364652829164]
  Joint network topology inference represents a canonical problem of learning multiple graph Laplacian matrices from heterogeneous graph signals. We propose a general graph estimator based on a novel structured fusion regularization. We show that the proposed graph estimator enjoys both high computational efficiency and rigorous theoretical guarantee.
  arXiv  Detail & Related papers  (2021-03-05T04:42:32Z)
• Articulated Shape Matching Using Laplacian Eigenfunctions and Unsupervised Point Registration [38.16866987817019]
  Spectral graph theory can be used to map these graphs onto lower dimensional spaces and match shapes by aligning their embeddings. We derive a new formulation that finds the best alignment between two congruent $K$-dimensional sets of points by selecting the best subset of eigenfunctions of the Laplacian matrix.
  arXiv  Detail & Related papers  (2020-12-14T08:49:25Z)
• A Differentiable Relaxation of Graph Segmentation and Alignment for AMR Parsing [75.36126971685034]
  We treat alignment and segmentation as latent variables in our model and induce them as part of end-to-end training. Our method also approaches that of a model that relies on citetLyu2018AMRPA's segmentation rules, which were hand-crafted to handle individual AMR constructions.
  arXiv  Detail & Related papers  (2020-10-23T21:22:50Z)
• Isometric Transformation Invariant and Equivariant Graph Convolutional Networks [5.249805590164902]
  We propose a set of transformation invariant and equivariant models based on graph convolutional networks, called IsoGCNs. We demonstrate that the proposed model has a competitive performance compared to state-of-the-art methods on tasks related to geometrical and physical simulation data.
  arXiv  Detail & Related papers  (2020-05-13T13:44:44Z)

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.