mlsanityData to 13 Oct 2026

How does fine-tuning improve sensorimotor representations in large language models?

arXiv 2603.03313 ↗First posted 9 Feb 2026cs.CL cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Large Language Models (LLMs) exhibit a significant "embodiment gap", where their text-based representations fail to align with human sensorimotor experiences. This study systematically investigates whether and how task-specific fine-tuning can bridge this gap. Utilizing Representational Similarity Analysis (RSA) and dimension-specific correlation metrics, we demonstrate that the internal representations of LLMs can be steered toward more embodied, grounded patterns through fine-tuning. Furthermore, the results show that while sensorimotor improvements generalize robustly across languages and related sensory-motor dimensions, they are highly sensitive to the learning objective, failing to transfer across two disparate task formats.

Key facts. No citing papers in the map yet (data through 13 Oct 2026). Builds on 2 of its 11 references that are in the map. First posted on arXiv 9 Feb 2026.

Builds on

2 of 11 references.

References · 2

  1. 2025Psycholinguistic Word Features: a New Approach for the Evaluation of LLMs Alignment with Humanscited by 5
  2. 2025Does Visual Grounding Enhance the Understanding of Embodied Knowledge in Large Language Models?cited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Alignment among Language, Vision and Action Representations
  2. 2023Instruction-tuning Aligns LLMs to the Human Brain
  3. 2025Uncovering the Computational Ingredients of Human-Like Representations in LLMs
  4. 2025Task-conditioned probing of instruction-tuned multimodal LLMs: Region-specific brain alignment patterns under naturalistic stimuli
  5. 2026Decoupling Internal Representational Changes and Causal Importance in Fine-Tuned Large Language Models
  6. 2023Dissociating language and thought in large language models
  7. 2025Testing the Limits of Fine-Tuning for Improving Visual Cognition in Vision Language Models
  8. 2023Language Models Meet World Models: Embodied Experiences Enhance Language Models