Research map / Vision-language and multimodal / Cross-modal learning
Collaborative Representation Learning for Alignment of Tactile, Language, and Vision Modalities
Yiyun Zhou, Mingjing Xu, Jingwei Shi, Quanjiang Li, Jingyuan Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Sensor-Aware Modulator (ablation) | TLV-CoRe · TAG subsets (trained on TacQuad) | accuracy | Improved |
| add tactile-irrelevant decoupled loss L_DL (ablation) | TLV-CoRe · TAG subsets (trained on TacQuad) | accuracy | Improved |
| add Unified Bridging Adapter (ablation) | TLV-CoRe · TAG subsets (trained on TacQuad) | accuracy | Improved |
| increase batch size to 128 for single-sensor baselines (ablation) | TLV-Link and VIT-LENS-2 · TAG subsets (trained on TacQuad) | accuracy | No clear effect |
| increase batch size (ablation) | TLV-CoRe · TAG subsets (trained on TacQuad) | accuracy | Improved |
| use UBA with shared bridging across modalities (TLV-CoRe vs VIT-LENS-2) | CLIP tactile encoder · CIFAR-10 via tactile encoder (trained on TAG) | accuracy (%): 44.38 -> 68.15 | Improved |
| train TLV-CoRe with 0.30% trainable parameters instead of larger adapters | TLV-CoRe vs VIT-LENS-2 · TAG Material (trained on TAG) | accuracy (%): 65.99 -> 65.44 | Mixed |
| train on Octopi, which lacks vision modality and has limited data | all methods · TAG subtests | accuracy | Hurt |
| add more training data under cross- and multi-sensor settings | all methods · cross-sensor and multi-sensor protocols | accuracy | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- AnyTouch — AnyTouch: Learning Unified Static-Dynamic Representation across Multiple Visuo-tactile Sensors (ICLR 2025) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Universal Visuo-Tactile Video Understanding for Embodied Interaction (NeurIPS 2025 · similarity 0.91)
- A Touch, Vision, and Language Dataset for Multimodal Alignment (ICML 2024 · similarity 0.90)
- AnyTouch: Learning Unified Static-Dynamic Representation across Multiple Visuo-tactile Sensors (ICLR 2025 · similarity 0.90)
- Binding Touch to Everything: Learning Unified Multimodal Tactile Representations (CVPR 2024 · similarity 0.89)
- Sensor-Invariant Tactile Representation (ICLR 2025 · similarity 0.89)
- Touch-R1: Reinforcing Touch Reasoning in MLLMs (NeurIPS 2026 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
