---
title: "Leaving the Cave by Breaking the Symmetry: Identifiable Structure Can Enable Cross-Lingual Generalization"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2610.38509v1"
permalink: "https://feed.astralor.com/items/itm_e7e9d9bc63bffa36d175407d"
published: "2026-10-28T09:25:32.000Z"
score: 55
channel: "ai"
tags: ["安全","Leaving","Cave","Breaking","Symmetry","Identifiable Structure Can","Enable Cross-Lingual Generalization","The Platonic Representation"]
---

# Leaving the Cave by Breaking the Symmetry: Identifiable Structure Can Enable Cross-Lingual Generalization

## 速览

Why do multilingual language models generalize across languages even without shared vocabulary? The Platonic Representation Hypothesis posits that models trained on different data, objectives, and modalities converge to…

## 关键点

- Why do multilingual language models generalize across languages even without shared vocabulary?
- The Platonic Representation Hypothesis posits that models trained on different data, objectives, and modalities converge toward a shared statistical model of reality.
- Extending this intuition to representations inside a single multilingual model, we propose the Relational Identifiability Hypothesis: relations in a shared world can promote entity alignment by ruling out alternative cr…

## 全文

Why do multilingual language models generalize across languages even without shared vocabulary? The Platonic Representation Hypothesis posits that models trained on different data, objectives, and modalities converge toward a shared statistical model of reality. Extending this intuition to representations inside a single multilingual model, we propose the Relational Identifiability Hypothesis: relations in a shared world can promote entity alignment by ruling out alternative cross-lingual correspondences, making corresponding entities more likely to become directly readable without an additional transformation. We test this hypothesis in a controlled bilingual Transformer whose two token-disjoint languages describe the same modular-arithmetic world. We find that the languages can develop a shared answer subspace without aligning corresponding numbers. Cross-lingual alignment emerges selectively when the available relations constrain number correspondences: subtraction makes zero task-level identifiable, inducing selective alignment, while an order-sensitive operation produces broad alignment across numbers. These patterns predict success in zero-shot cross-lingual transfer. Together, our results show that relational structure in a shared world can promote entity alignment and enable transfer.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_e7e9d9bc63bffa36d175407d