AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Collaborative and Efficient Fine-tuning: Leveraging Task Similarity

Gagik Magakyan, Amirhossein Reisizadeh, Chanwoo Park, Pablo A. Parrilo, Asuman E. Ozdaglar

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train collaboratively with CoLoRA instead of local LoRAQwen2.5-1.5B-Instruct · task group, similarity 0.265RougeL: 0.910 -> 0.970Improved
train collaboratively with CoLoRA instead of local LoRAQwen2.5-1.5B-Instruct · task group, similarity 0.193RougeL: 0.937 -> 0.948Improved
train collaboratively with CoLoRA instead of local LoRAQwen2.5-1.5B-Instruct · task group, similarity 0.114RougeL: 0.405 -> 0.378Hurt
train collaboratively with CoLoRA instead of local LoRAQwen2.5-1.5B-Instruct · task group, similarity 0.234RougeL: 0.635 -> 0.727Improved
train collaboratively with CoLoRA instead of local LoRAQwen2.5-1.5B-Instruct · task group, similarity 0.222RougeL: 0.527 -> 0.633Improved
train collaboratively with CoLoRA instead of local LoRAQwen2.5-1.5B-Instruct · task group, similarity 0.211RougeL: 0.742 -> 0.733Hurt
use CoLoRA instead of RAVAN at rank 16Qwen2.5-1.5B-Instruct · similarity 0.193RougeL: 0.813 -> 0.979Improved
use CoLoRA instead of ALoRA at rank 16Qwen2.5-1.5B-Instruct · similarity 0.265RougeL: 0.970 -> 0.960Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy