---
title: "Object-Centric Alignment and Anchor Distillation for Weakly Supervised Referring Expression Comprehension"
authors:
  - "Yi Tian"
  - "Cheng Yang"
  - "Qingbao Huang"
doi: "10.24963/ijcai.2026/190"
url: "https://ijcai.org/proceedings/2026/190"
pdf: "https://ijcai.org/proceedings/2026/0190.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "1704-1712"
track: "Main Track"
keywords:
  - "Machine Learning: Weakly supervised learning"
  - "Computer Vision: Vision, language and reasoning"
---

# Object-Centric Alignment and Anchor Distillation for Weakly Supervised Referring Expression Comprehension

Yi Tian, Cheng Yang, Qingbao Huang

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 1704-1712. DOI: [10.24963/ijcai.2026/190](https://doi.org/10.24963/ijcai.2026/190)

## Abstract

Weakly supervised Referring Expression Comprehension (WREC) aims to localize referred objects
using only image-text pairs without box-level annotations. Existing one-stage methods predominantly
rely on anchor-level alignment, which suffers from
two fundamental limitations: (1) anchors represent
local visual patches rather than holistic objects, and
(2) they lack the capability to model inter-object
relations. To address these issues, we propose
OCAAD, an Object-Centric Alignment and Anchor Distillation framework. Our key insight is that
different self-attention heads in DINOv2 naturally
attend to distinct semantic regions, effectively capturing object-level information. Building on this,
OCAAD introduces two complementary modules:
(1) Anchor-Object Distillation Module (AODM),
which bridges the semantic gap between anchors
and objects by transferring object-level knowledge
from attention heads to anchors via overlap-aware
contrastive learning; and (2) Intra-Modal Relation Consistency (IRC), which explicitly models
inter-object relations by enforcing the relational
structure among linguistic entities to match that of
their visual counterparts. Extensive experiments on
RefCOCO, RefCOCO+, and RefCOCOg demonstrate that OCAAD achieves new state-of-the-art
performance, validating the effectiveness of objectcentric alignment for WREC. Code is available at
https://github.com/VILAN-Lab/OCAAD.