---
title: "Aligning MLLMs with the Latent Structure of Human Cognition via Behavior-Derived Semantic Dimensions"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2610.41951v1"
permalink: "https://feed.astralor.com/items/itm_de0b46ba3032cb5689d3aa11"
published: "2026-10-30T16:06:08.000Z"
score: 56
channel: "tech"
tags: ["LLM","安全","多模态","Aligning MLLMs","Latent Structure","Human Cognition","Behavior-Derived Semantic Dimensions","Multimodal Large Language"]
---

# Aligning MLLMs with the Latent Structure of Human Cognition via Behavior-Derived Semantic Dimensions

## 速览

Multimodal Large Language Models (MLLMs) have achieved impressive performance across a wide range of reasoning tasks, yet it remains unclear whether their internal representations and decision criteria align with the la…

## 关键点

- Multimodal Large Language Models (MLLMs) have achieved impressive performance across a wide range of reasoning tasks, yet it remains unclear whether their internal representations and decision criteria align with the la…
- In this work, we propose a novel framework to align MLLMs with the multidimensional mental representations underlying human similarity judgements.
- We build on a previously established behavior-derived semantic embedding space, in which objects are represented along 66 interpretable dimensions derived from 4.70 million human Odd-One-Out (O1O) judgements.

## 全文

Multimodal Large Language Models (MLLMs) have achieved impressive performance across a wide range of reasoning tasks, yet it remains unclear whether their internal representations and decision criteria align with the latent structure of human cognition. In this work, we propose a novel framework to align MLLMs with the multidimensional mental representations underlying human similarity judgements. We build on a previously established behavior-derived semantic embedding space, in which objects are represented along 66 interpretable dimensions derived from 4.70 million human Odd-One-Out (O1O) judgements. For each triplet, we infer the most salient decision dimension and convert it into a dimension-grounded linguistic rationale. This enables MLLMs to learn both the final choice and a justification consistent with the inferred behavioral criterion. Experimental results demonstrate that our approach improves model consistency with human similarity judgements while retaining competitive performance on standard multimodal benchmarks. Furthermore, using searchlight Representational Similarity Analysis (RSA) on an independent fMRI Natural Scenes Dataset (NSD), we observe increased representational alignment between model and human. Ultimately, our findings point to a data-driven route for incorporating human cognitive structure into MLLMs alignment. The code is available \href{https://github.com/ening-open/cogalign}{here}.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_de0b46ba3032cb5689d3aa11