Papers › Empowering Source-Free Domain Adaptation with MLLM-driven Curriculum Learning

Empowering Source-Free Domain Adaptation with MLLM-driven Curriculum Learning

28 May 2024arXiv:2405.18376archive 2025-07-28

Dongjie Chen, Kartik Patwari, Zhengfeng Lai, Sen-ching Cheung, Chen-Nee Chuah

Source-Free Domain Adaptation (SFDA) aims to adapt a pre-trained source model to a target domain using only unlabeled target data. Current SFDA methods face challenges in effectively leveraging pre-trained knowledge and exploiting target domain data. Multimodal Large Language Models (MLLMs) offer remarkable capabilities in understanding visual and textual information, but their applicability to SFDA poses challenges such as instruction-following failures, intensive computational demands, and difficulties in performance measurement prior to adaptation. To alleviate these issues, we propose Reliability-based Curriculum Learning (RCL), a novel framework that integrates multiple MLLMs for knowledge exploitation via pseudo-labeling in SFDA. Our framework incorporates proposed Reliable Knowledge Transfer, Self-correcting and MLLM-guided Knowledge Expansion, and Multi-hot Masking Refinement to progressively exploit unlabeled data in the target domain. RCL achieves state-of-the-art (SOTA) performance on multiple SFDA benchmarks, e.g., +9.4 on DomainNet, demonstrating its effectiveness in enhancing adaptability and robustness without requiring access to source data. Code: https://github.com/Dong-Jie-Chen/RCL.

Paper PDF Code

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

Dong-Jie-Chen/RCL officialmentioned in papermentioned on GitHub report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Domain Adaptation Instruction Following Source-Free Domain Adaptation Transfer Learning Unsupervised Domain Adaptation

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Domain Adaptation	Office-Home	RCL	Accuracy	90.0	#2 of 29	Archive leaderboard	report
Domain Adaptation	VisDA2017	RCL	Accuracy	93.2	#2 of 28	Archive leaderboard	report
Source-Free Domain Adaptation	VisDA-2017	RCL	Accuracy	93.2	#1 of 10	Archive leaderboard	report
Unsupervised Domain Adaptation	Office-Home	RCL	Accuracy	90.0	#2 of 20	Archive leaderboard	report
Unsupervised Domain Adaptation	VisDA2017	RCL	Accuracy	93.2	#2 of 13	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections