TacMamba: A Tactile History Compression Adapter Bridging Fast Reflexes and Slow VLA Reasoning

• URL: http://arxiv.org/abs/2603.01700v1
• Date: Mon, 02 Mar 2026 10:26:54 GMT
• Title: TacMamba: A Tactile History Compression Adapter Bridging Fast Reflexes and Slow VLA Reasoning
• Abstract summary: We introduce TacMamba, a high-frequency tactile interface for flexible integration with low-frequency visual planning.<n>We show that TacMamba achieves 100% success rates, significantly outperforming the visual-only pi_0.5 baseline, while strictly satisfying hard real-time constraints.<n> Experiments on discrete counting and implicit state switching demonstrate that TacMamba achieves 100% success rates, significantly outperforming the visual-only pi_0.5 baseline, while strictly satisfying hard real-time constraints.
• Score: 16.21461083401249
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In visually ambiguous manipulation such as detecting button click tactile feedback is often the sole source of ground truth. However, fusing tactile data poses a significant challenge due to a spatiotemporal mismatch: tactile perception requires high-frequency processing with long-horizon memory (System 1), whereas visual policies operate at low control frequencies (System 2). Existing architectures struggle to bridge this gap: Transformers are computationally prohibitive for high-frequency loops (>100Hz), while LSTMs suffer from forgetting over extended interaction histories. In this paper, we introduce TacMamba, a hierarchical architecture that aligns high-bandwidth tactile reflexes with low-frequency visual planning. Our approach comprises three core contributions: (1) a custom high-frequency tactile interface designed for flexible integration; (2) a Mamba-based Tactile History Compressor that encodes continuous force history into a compact state with O(1) inference latency (0.45 ms), enabling plug-and-play fusion with VLA models without joint pre-training and (3) a Tactile-Guided Dual-Stage Training strategy that leverages temporal discrimination for self-supervised representation learning and phase-uniform sampling to mitigate data sparsity. Experiments on discrete counting and implicit state switching demonstrate that TacMamba achieves 100% success rates, significantly outperforming the visual-only pi_0.5 baseline, while strictly satisfying hard real-time constraints.

Related papers

• A Hierarchical Approach to Imitation Learning for Manipulation Tasks Requiring Time Varying Forces [14.372073538832803]
  We introduce the Diffusion Policy Augmented by Fast Trajectory Generation (DPA-FTG)<n>Compared to recent visual-tactile approaches that focus on positional correction, DPA-FTG decouples low-frequency planning from high-frequency force regulation.<n>We validate our approach on a bimanual battery disassembly task involving the separation of a compliant sheet.
  arXiv  Detail & Related papers  (2026-08-04T04:21:00Z)
• WaveZip: Wavelet-Driven Space-Time Decoupling for Video Token Condensation [69.37452950816122]
  WaveZip is a joint signal-frequency-domain framework for efficient video inference.<n> experiments on long video understanding benchmarks demonstrate that WaveZip retains 99.6% of the full performance.
  arXiv  Detail & Related papers  (2026-07-25T16:05:22Z)
• UniFS: Unified Fast-to-Slow Hierarchical Architecture for Vision-Language-Action Models [15.95393143175101]
  We introduce UniFS, a unified fast-to-slow architecture for vision-language models.<n>We show that UniFS achieves state-of-the-art performance (98.3% average success rate, a 2.5% gain over VLA-Adapter baseline)<n> Experiments on LIBERO show that UniFS achieves state-of-the-art performance (98.3% average success rate, a 2.5% gain over VLA-Adapter baseline)
  arXiv  Detail & Related papers  (2026-06-22T03:10:19Z)
• T-Rex: Tactile-Reactive Dexterous Manipulation [139.71263755530654]
  We introduce a variable-rate Mixture-of-Transformers (MoT) architecture equipped with a novel temporal tactile VQ-VAE encoder.<n>We demonstrate the effectiveness of tactile-reactive policies on 12 manipulation tasks requiring delicate force control and deformable object manipulation.
  arXiv  Detail & Related papers  (2026-06-15T17:59:55Z)
• DexWorldModel: Causal Latent World Modeling towards Automated Learning of Embodied Tasks [54.32016216994156]
  We introduce the Causal Latent World Model (CLWM), which employs DINOv3 features as generative targets to disentangle interaction semantics from visual noise.<n>To overcome memory scaling, CLWM features a Dual-State Test-Time Training (TTT) Memory that guarantees a strict $mathcalO(1)$ footprint for long-horizon tasks.<n>We present EmbodiChain, an online framework that establishes the Efficiency Law by injecting an infinite flow of physics-grounded trajectories during training.
  arXiv  Detail & Related papers  (2026-04-13T03:19:36Z)
• SkeFi: Cross-Modal Knowledge Transfer for Wireless Skeleton-Based Action Recognition [20.020503149009787]
  Existing solutions use RGB cameras to annotate skeletal keypoints, but their performance declines in dark environments and raises privacy concerns.<n>This paper explores non-invasive wireless sensors, i.e., LiDAR and mmWave, to mitigate these challenges.<n> Experiments demonstrate that SkeFi realizes state-of-the-art performances on mmWave and LiDAR.
  arXiv  Detail & Related papers  (2026-01-18T14:39:02Z)
• Lightweight Test-Time Adaptation for EMG-Based Gesture Recognition [2.414036142474149]
  We propose a lightweight framework for Test-Time Adaptation (TTA) using a Temporal Convolutional Network (TCN) backbone.<n>We introduce three deployment-ready strategies: causal adaptive batch normalization for real-time statistical alignment; (ii) a Gaussian Mixture Model (GMM) alignment with experience replay to prevent forgetting; and (iii) meta-learning for rapid, few-shot calibration.<n>Our results show that experience-replay updates yield superior stability under limited data, while meta-learning achieves competitive performance in one- and two-shot regimes.
  arXiv  Detail & Related papers  (2026-01-07T18:48:31Z)
• Real-Time LiDAR Super-Resolution via Frequency-Aware Multi-Scale Fusion [0.4078247440919472]
  FLASH (Frequency-aware LiDAR Adaptive Super-resolution with Hierarchical fusion) is a novel framework that overcomes limitations through dual-domain processing.<n> FLASH integrates two key innovations: (i) Frequency-Aware Window Attention that combines local spatial attention with global frequency-domain analysis via FFT, capturing both fine-grained geometry and periodic scanning patterns at log-linear complexity, and (ii) Adaptive Multi-Scale Fusion that replaces conventional skip connections with learned position-specific feature aggregation, enhanced by CBAM attention for dynamic feature selection.
  arXiv  Detail & Related papers  (2025-11-10T18:38:15Z)
• Learning Frequency and Memory-Aware Prompts for Multi-Modal Object Tracking [74.15663758681849]
  We present Learning Frequency and Memory-Aware Prompts, a dual-adapter framework that injects lightweight prompts into a frozen RGB tracker.<n>A frequency-guided visual adapter adaptively transfers complementary cues across modalities.<n>A multilevel memory adapter with short, long, and permanent memory stores, updates, and retrieves reliable temporal context.
  arXiv  Detail & Related papers  (2025-06-30T15:38:26Z)
• FUSE: Label-Free Image-Event Joint Monocular Depth Estimation via Frequency-Decoupled Alignment and Degradation-Robust Fusion [92.4205087439928]
  Image-event joint depth estimation methods leverage complementary modalities for robust perception, yet face challenges in generalizability.<n>We propose the Self-supervised Transfer (PST) and the FrequencyDe-coupled Fusion module (FreDF)<n>PST establishes cross-modal knowledge transfer through latent space alignment with image foundation models, effectively mitigating data scarcity.<n>FreDF explicitly decouples high-frequency edge features from low-frequency structural components, resolving modality-specific frequency mismatches.<n>This combined approach enables FUSE to construct a universal image-event that only requires lightweight decoder adaptation for target datasets.
  arXiv  Detail & Related papers  (2025-03-25T15:04:53Z)
• Multi-granular body modeling with Redundancy-Free Spatiotemporal Fusion for Text-Driven Motion Generation [10.843503146808839]
  We introduce HiSTF Mamba, a framework with three parts: Dual-tial Mamba, Bi-Temporal Mamba and a Spatiotemporal Fusion Module (DSFM)<n>Experiments on the HumanML3D benchmark show that HiSTF Mamba performs well across several metrics, achieving high fidelity and tight semantic alignment between text and motion.
  arXiv  Detail & Related papers  (2025-03-10T04:01:48Z)
• Reactive Diffusion Policy: Slow-Fast Visual-Tactile Policy Learning for Contact-Rich Manipulation [58.95799126311524]
  Humans can accomplish contact-rich tasks using vision and touch, with highly reactive capabilities such as fast response to external changes and adaptive control of contact forces.<n>Existing visual imitation learning approaches rely on action chunking to model complex behaviors.<n>We introduce TactAR, a low-cost teleoperation system that provides real-time tactile feedback through Augmented Reality.
  arXiv  Detail & Related papers  (2025-03-04T18:58:21Z)

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.