---
title: "CARE: Contrastive Representation Learning with Environment-Invariance Regularization for Behavioural Intention Prediction"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.02806v1"
permalink: "https://feed.astralor.com/items/itm_e6c2ce87eb32254571d19f4f"
published: "2026-11-02T09:13:09.000Z"
score: 58
channel: "tech"
tags: ["CARE","Contrastive Representation Learning","Environment-Invariance Regularization","Behavioural Intention Prediction","Trajectory-based"]
---

# CARE: Contrastive Representation Learning with Environment-Invariance Regularization for Behavioural Intention Prediction

## 速览

Trajectory-based behavioural intention prediction faces environmental variation, and representations learned with a pooled contrastive objective may retain environment-specific regularities. We present CARE, a self-supe…

## 关键点

- Trajectory-based behavioural intention prediction faces environmental variation, and representations learned with a pooled contrastive objective may retain environment-specific regularities.
- We present CARE, a self-supervised trajectory-learning framework that couples temporally asymmetric contrastive views with an invariant-risk-minimisation (IRM)-style gradient penalty on environment-wise contrastive risk…
- Each input window yields an anchor, a partially masked positive view, and a temporally perturbed negative view.

## 全文

Trajectory-based behavioural intention prediction faces environmental variation, and representations learned with a pooled contrastive objective may retain environment-specific regularities. We present CARE, a self-supervised trajectory-learning framework that couples temporally asymmetric contrastive views with an invariant-risk-minimisation (IRM)-style gradient penalty on environment-wise contrastive risks. Each input window yields an anchor, a partially masked positive view, and a temporally perturbed negative view. Masked reconstruction and observability-aware evidence consistency complement this objective by promoting temporal information retention under partial input. On an IMPTC-derived four-class intention task, CARE achieves 0.840 accuracy and 0.023 expected calibration error (ECE) under IID fine-tuning, and 0.806 accuracy with a Brier score of 0.308 on the reported held-out environment. Under frozen-encoder linear probing, reconstruction provides the largest individual held-out accuracy gain, whereas the complete objective yields the lowest held-out ECE (0.049) among tested variants. These results support the combined objective as a competitive representation-learning approach under the evaluated shift, while the IRM-style regulariser is not taken as evidence of formal invariance.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_e6c2ce87eb32254571d19f4f