# SToLa: Self-Adaptive Touch-Language Framework with Tactile Commonsense Reasoning in Open-Ended Scenarios

arXiv: https://arxiv.org/abs/2505.04201 · HTML: https://mlsanity.com/paper/2505.04201/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 27 of its 41 references that are in the map.
- First posted on arXiv 7 May 2025.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)
- Mixture-of-experts / [Multimodal MoE models](https://mlsanity.com/moe/multimodal-moe-models/)

## Builds on (most cited first, 27 of 27)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [CIDEr: Consensus-based Image Description Evaluation](https://mlsanity.com/paper/1411.5726/) (2014)
- [ShareGPT4V: Improving Large Multi-Modal Models with Better Captions](https://mlsanity.com/paper/2311.12793/) (2023)
- [GLaM: Efficient Scaling of Language Models with Mixture-of-Experts](https://mlsanity.com/paper/2112.06905/) (2021)
- [LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention](https://mlsanity.com/paper/2303.16199/) (2023)
- [Scaling Vision with Sparse Mixture of Experts](https://mlsanity.com/paper/2106.05974/) (2021)
- [NExT-GPT: Any-to-Any Multimodal LLM](https://mlsanity.com/paper/2309.05519/) (2023)
- [LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model](https://mlsanity.com/paper/2304.15010/) (2023)
- [MoE-LLaVA: Mixture of Experts for Large Vision-Language Models](https://mlsanity.com/paper/2401.15947/) (2024)
- [Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints](https://mlsanity.com/paper/2212.05055/) (2022)
- [Fine-tuned CLIP Models are Efficient Video Learners](https://mlsanity.com/paper/2212.03640/) (2022)
- [Uni-MoE: Scaling Unified Multimodal LLMs with Mixture of Experts](https://mlsanity.com/paper/2405.11273/) (2024)
- [Binding Touch to Everything: Learning Unified Multimodal Tactile Representations](https://mlsanity.com/paper/2401.18084/) (2024)
- [Touch and Go: Learning from Human-Collected Vision and Touch](https://mlsanity.com/paper/2211.12498/) (2022)
- [Sparsh: Self-supervised touch representations for vision-based tactile sensing](https://mlsanity.com/paper/2410.24090/) (2024)
- [Octopi: Object Property Reasoning with Large Tactile-Language Models](https://mlsanity.com/paper/2405.02794/) (2024)
- [AnyTouch: Learning Unified Static-Dynamic Representation across Multiple Visuo-tactile Sensors](https://mlsanity.com/paper/2502.12191/) (2025)
- [LLaVA-MoD: Making LLaVA Tiny via MoE Knowledge Distillation](https://mlsanity.com/paper/2408.15881/) (2024)
- [The ObjectFolder Benchmark: Multisensory Learning with Neural and Real Objects](https://mlsanity.com/paper/2306.00956/) (2023)
- [Multimodal Visual-Tactile Representation Learning through Self-Supervised Contrastive Pre-Training](https://mlsanity.com/paper/2401.12024/) (2024)
- [Touch100k: A Large-Scale Touch-Language-Vision Dataset for Touch-Centric Multimodal Representation](https://mlsanity.com/paper/2406.03813/) (2024)
- [EVE: Efficient Vision-Language Pre-training with Masked Prediction and Modality-Aware MoE](https://mlsanity.com/paper/2308.11971/) (2023)
- [MoE Jetpack: From Dense Checkpoints to Adaptive Mixture of Experts for Vision Tasks](https://mlsanity.com/paper/2406.04801/) (2024)
- [Analysing the Interplay of Vision and Touch for Dexterous Insertion Tasks](https://mlsanity.com/paper/2410.23860/) (2024)
- [Visuo-Tactile Zero-Shot Object Recognition with Vision-Language Model](https://mlsanity.com/paper/2409.09276/) (2024)

## Built on by (most cited first, 3 of 3)

- [UniTac: A Unified Multimodal Model for Cross-Sensor Tactile Understanding and Generation](https://mlsanity.com/paper/2606.31451/) (2026)
- [Tactile-based Multimodal Fusion in Embodied Intelligence: A Survey of Vision, Language, and Contact-Driven Paradigms](https://mlsanity.com/paper/2605.17336/) (2026)
- [Touch-R1: Reinforcing Touch Reasoning in MLLMs](https://mlsanity.com/paper/2605.27154/) (2026)

## Abstract (arXiv)

This paper explores the challenges of integrating tactile sensing into intelligent systems for multimodal reasoning, particularly in enabling commonsense reasoning about the open-ended physical world. We identify two key challenges: modality discrepancy, where existing large touch-language models often treat touch as a mere sub-modality of language, and open-ended tactile data scarcity, where current datasets lack the diversity, open-endness and complexity needed for reasoning. To overcome these challenges, we introduce SToLa, a Self-Adaptive Touch-Language framework. SToLa utilizes Mixture of Experts (MoE) to dynamically process, unify, and manage tactile and language modalities, capturing their unique characteristics. Crucially, we also present a comprehensive tactile commonsense reasoning dataset and benchmark featuring free-form questions and responses, 8 physical properties, 4 interactive characteristics, and diverse commonsense knowledge. Experiments show SToLa exhibits competitive performance compared to existing models on the PhysiCLeAR benchmark and self-constructed datasets, proving the effectiveness of the Mixture of Experts architecture in multimodal management and the performance advantages for open-scenario tactile commonsense reasoning tasks.