paper-with-me

Papers

Improving Multitask Retrieval by Promoting Task Specialization

2023-07-01 · Wenzheng Zhang, Chenyan Xiong, Karl Stratos, Arnold Overwijk

In multitask retrieval, a single retriever is trained to retrieve relevant contexts for multiple tasks. Despite its practical appeal, naive multitask retrieval lags behind task-specific retrieval in which a separate retriever is trained for each task. We show that it is possible to train a multitask retriever that outperforms task-specific retrievers by promoting task specialization. The main ingredients are: (1) a better choice of pretrained model (one that is explicitly optimized for multitasking) along with compatible prompting, and (2) a novel adaptive learning method that encourages each parameter to specialize in a particular task. The resulting multitask retriever is highly performant on the KILT benchmark. Upon analysis, we find that the model indeed learns parameters that are more task-specialized compared to naive multitasking without prompting or adaptive learning.

📄 PDF Abstract BibTeX arXiv:2307.00342

Code (1)

wenzhengzhang/taco 공식 구현 pytorch

Tasks

Retrieval

Similar Papers 제목 키워드 기반

ERMoE: Eigen-Reparameterized Mixture-of-Experts for Stable Routing and Interpretable Specialization

2025-11-14 · Anzhe Cheng, Shukai Duan, Shixuan Li, Chenzhong Yin 외 arxiv

Mixture-of-Experts (MoE) architectures expand model capacity by sparsely activating experts but face two core challenges: misalignment between router logits and each expert's internal structure leads to unstable routing …

Text Retrieval

Contest in Multitasking: An Evidence from Chinese County Officials' Promotion Assessment

2023-02-17 · Yuanhao Zhang

Real-world observed contests often take the form of multi-task contests rather than single-task contests, and existing theories are insufficient to explain the incentive for extending the task dimension. This paper propo…

Brain Tumor Image Retrieval via Multitask Learning

2018-10-22 · Maxim Pisov, Gleb Makarchuk, Valery Kostjuchenko, Alexandra Dalechina 외

Classification-based image retrieval systems are built by training convolutional neural networks (CNNs) on a relevant classification problem and using the distance in the resulting feature space as a similarity metric. H…

ClassificationGeneral ClassificationImage RetrievalRetrieval

MemeLens: Multilingual Multitask VLMs for Memes

2026-01-18 · Ali Ezzat Shahroor, Mohamed Bayan Kmainasi, Abul Hasnat, Dimitar Dimitrov 외 arxiv

Memes are a dominant medium for online communication and manipulation because meaning emerges from interactions between embedded text, imagery, and cultural context. Existing meme research is distributed across tasks (ha…

Domain Generalization

RAVEN: Multitask Retrieval Augmented Vision-Language Learning

2024-06-27 · Varun Nagaraj Rao, Siddharth Choudhary, Aditya Deshpande, Ravi Kumar Satzoda 외

The scaling of large language models to encode all the world's knowledge in model parameters is unsustainable and has exacerbated resource barriers. Retrieval-Augmented Generation (RAG) presents a potential solution, yet…

Image CaptioningRAGRetrievalRetrieval-augmented Generation+1