# LA-Sign: Looped Transformers with Geometry-aware Alignment for Skeleton-based Sign Language Recognition

arXiv: https://arxiv.org/abs/2603.29057 · HTML: https://mlsanity.com/paper/2603.29057/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 47 of its 67 references that are in the map.
- First posted on arXiv 30 Mar 2026.

## Builds on (most cited first, 30 of 47)

- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [mT5: A massively multilingual pre-trained text-to-text transformer](https://mlsanity.com/paper/2010.11934/) (2020)
- [Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition](https://mlsanity.com/paper/1801.07455/) (2018)
- [TSM: Temporal Shift Module for Efficient Video Understanding](https://mlsanity.com/paper/1811.08383/) (2018)
- [Poincar'e Embeddings for Learning Hierarchical Representations](https://mlsanity.com/paper/1705.08039/) (2017)
- [Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach](https://mlsanity.com/paper/2502.05171/) (2025)
- [Hyperbolic Neural Networks](https://mlsanity.com/paper/1805.09112/) (2018)
- [Reasoning with Latent Thoughts: On the Power of Looped Transformers](https://mlsanity.com/paper/2502.17416/) (2025)
- [Looped Transformers as Programmable Computers](https://mlsanity.com/paper/2301.13196/) (2023)
- [Word-level Deep Sign Language Recognition from Video: A New Large-scale Dataset and Methods Comparison](https://mlsanity.com/paper/1910.11006/) (2019)
- [DeepMDP: Learning Continuous Latent Space Models for Representation Learning](https://mlsanity.com/paper/1906.02736/) (2019)
- [Riemannian Adaptive Optimization Methods](https://mlsanity.com/paper/1810.00760/) (2018)
- [Learning Spatio-Temporal Features with 3D Residual Networks for Action Recognition](https://mlsanity.com/paper/1708.07632/) (2017)
- [RTMPose: Real-Time Multi-Person Pose Estimation based on MMPose](https://mlsanity.com/paper/2303.07399/) (2023)
- [Looped Transformers are Better at Learning Learning Algorithms](https://mlsanity.com/paper/2311.12424/) (2023)
- [Hyperbolic Image-Text Representations](https://mlsanity.com/paper/2304.09172/) (2023)
- [On the Turing Completeness of Modern Neural Network Architectures](https://mlsanity.com/paper/1901.03429/) (2019)
- [MS-ASL: A Large-Scale Data Set and Benchmark for Understanding American Sign Language](https://mlsanity.com/paper/1812.01053/) (2018)
- [Can You Learn an Algorithm? Generalizing from Easy to Hard Problems with Recurrent Networks](https://mlsanity.com/paper/2106.04537/) (2021)
- [Skeleton Aware Multi-modal Sign Language Recognition](https://mlsanity.com/paper/2103.08833/) (2021)
- [Looped Transformers for Length Generalization](https://mlsanity.com/paper/2409.15647/) (2024)
- [YouTube-ASL: A Large-Scale, Open-Domain American Sign Language-English Parallel Corpus](https://mlsanity.com/paper/2306.15162/) (2023)
- [Can Looped Transformers Learn to Implement Multi-step Gradient Descent for In-context Learning?](https://mlsanity.com/paper/2410.08292/) (2024)
- [Transferring Cross-domain Knowledge for Video Sign Language Recognition](https://mlsanity.com/paper/2003.03703/) (2020)
- [SignBERT: Pre-Training of Hand-Model-Aware Representation for Sign Language Recognition](https://mlsanity.com/paper/2110.05382/) (2021)
- [mm-Pose: Real-Time Human Skeletal Posture Estimation using mmWave Radars and CNNs](https://mlsanity.com/paper/1911.09592/) (2019)
- [SignBERT+: Hand-model-aware Self-supervised Pre-training for Sign Language Understanding](https://mlsanity.com/paper/2305.04868/) (2023)
- [MoEUT: Mixture-of-Experts Universal Transformers](https://mlsanity.com/paper/2405.16039/) (2024)
- [Inner Thinking Transformer: Leveraging Dynamic Depth Scaling to Foster Adaptive Internal Thinking](https://mlsanity.com/paper/2502.13842/) (2025)

## Abstract (arXiv)

Skeleton-based isolated sign language recognition (ISLR) demands fine-grained understanding of articulated motion across multiple spatial scales, from subtle finger movements to global body dynamics. Existing approaches typically rely on deep feed-forward architectures, which increase model capacity but lack mechanisms for recurrent refinement and structured representation. We propose LA-Sign, a looped transformer framework with geometry-aware alignment for ISLR. Instead of stacking deeper layers, LA-Sign derives its depth from recurrence, repeatedly revisiting latent representations to progressively refine motion understanding under shared parameters. To further regularise this refinement process, we present a geometry-aware contrastive objective that projects skeletal and textual features into an adaptive hyperbolic space, encouraging multi-scale semantic organisation. We study three looping designs and multiple geometric manifolds, demonstrating that encoder-decoder looping combined with adaptive Poincare alignment yields the strongest performance. Extensive experiments on WLASL and MSASL benchmarks show that LA-Sign achieves state-of-the-art results while using fewer unique layers, highlighting the effectiveness of recurrent latent refinement and geometry-aware representation learning for sign language recognition.