English
Related papers

Related papers: Merging Text Transformer Models from Different Ini…

200 papers

How does language model pretraining help transfer learning? We consider a simple ablation technique for determining the impact of each pretrained layer on transfer task performance. This method, partial reinitialization, involves replacing…

Computation and Language · Computer Science 2020-11-11 Alex Tamkin , Trisha Singh , Davide Giovanardi , Noah Goodman

Transformer-based Large Language Models (LLMs) traditionally rely on final-layer loss for training and final-layer representations for predictions, potentially overlooking the predictive power embedded in intermediate layers. Surprisingly,…

Computation and Language · Computer Science 2024-10-18 Haoyan Luo , Lucia Specia

Sequence modelling requires determining which past tokens are causally relevant from the context and their importance: a process inherent to the attention layers in transformers, yet whose underlying learned mechanisms remain poorly…

Machine Learning · Computer Science 2026-04-14 Francesco D'Angelo , Nicolas Flammarion

Parameter-level model merging is an emerging paradigm in multi-task learning with significant promise. Previous research has explored its connections with prediction-level model ensembling-commonly viewed as the upper bound for merging-to…

Machine Learning · Computer Science 2025-03-04 Qi Li , Runpeng Yu , Xinchao Wang

Transformer-based pretrained language models (T-PTLMs) have achieved great success in almost every NLP task. The evolution of these models started with GPT and BERT. These models are built on the top of transformers, self-supervised…

Computation and Language · Computer Science 2021-08-31 Katikapalli Subramanyam Kalyan , Ajit Rajasekharan , Sivanesan Sangeetha

The wide applicability of pretrained transformer models (PTMs) for natural language tasks is well demonstrated, but their ability to comprehend short phrases of text is less explored. To this end, we evaluate different PTMs from the lens of…

Computation and Language · Computer Science 2021-12-16 Sai Muralidhar Jayanthi , Varsha Embar , Karthik Raghunathan

The impressive achievements of transformers force NLP researchers to delve into how these models represent the underlying structure of natural language. In this paper, we propose a novel standpoint to investigate the above issue: using…

Computation and Language · Computer Science 2024-09-11 Elena Sofia Ruzzetti , Federico Ranaldi , Felicia Logozzo , Michele Mastromattei , Leonardo Ranaldi , Fabio Massimo Zanzotto

In-context learning, a capability that enables a model to learn from input examples on the fly without necessitating weight updates, is a defining characteristic of large language models. In this work, we follow the setting proposed in…

Machine Learning · Computer Science 2023-05-29 Kartik Ahuja , David Lopez-Paz

The success of pretrain-finetune paradigm brings about the release of numerous model weights. In this case, merging models finetuned on different tasks to enable a single model with multi-task capabilities is gaining increasing attention…

Machine Learning · Computer Science 2024-09-30 Chenyu Huang , Peng Ye , Tao Chen , Tong He , Xiangyu Yue , Wanli Ouyang

The Transformer model has achieved state-of-the-art performance in many sequence modeling tasks. However, how to leverage model capacity with large or variable depths is still an open challenge. We present a probabilistic framework to…

Computation and Language · Computer Science 2020-10-19 Xian Li , Asa Cooper Stickland , Yuqing Tang , Xiang Kong

Model merging has emerged as a crucial technique in Deep Learning, enabling the integration of multiple models into a unified system while preserving perfor-mance and scalability. In this respect, the compositional properties of low-rank…

Machine Learning · Computer Science 2025-03-11 Riccardo Salami , Pietro Buzzega , Matteo Mosconi , Jacopo Bonato , Luigi Sabetta , Simone Calderara

As scaled language models (LMs) approach human-level reasoning capabilities, self-improvement emerges as a solution to synthesizing high-quality data corpus. While previous research has identified model collapse as a risk in…

Computation and Language · Computer Science 2025-10-28 Xiangchi Yuan , Chunhui Zhang , Zheyuan Liu , Dachuan Shi , Leyan Pan , Soroush Vosoughi , Wenke Lee

This paper proposes a transformer over transformer framework, called Transformer$^2$, to perform neural text segmentation. It consists of two components: bottom-level sentence encoders using pre-trained transformers, and an upper-level…

Computation and Language · Computer Science 2021-10-15 Kelvin Lo , Yuan Jin , Weicong Tan , Ming Liu , Lan Du , Wray Buntine

We seek to understand how the representations of individual tokens and the structure of the learned feature space evolve between layers in deep neural networks under different learning objectives. We focus on the Transformers for our…

Computation and Language · Computer Science 2019-09-05 Elena Voita , Rico Sennrich , Ivan Titov

Transformer models, notably large language models (LLMs), have the remarkable ability to perform in-context learning (ICL) -- to perform new tasks when prompted with unseen input-output examples without any explicit model training. In this…

Machine Learning · Computer Science 2023-11-03 Steve Yadlowsky , Lyric Doshi , Nilesh Tripuraneni

Model merging aims to combine multiple expert models into a more capable single model, offering benefits such as reduced storage and serving costs, improved generalization, and support for decentralized model development. Despite its…

Machine Learning · Computer Science 2024-10-07 Prateek Yadav , Tu Vu , Jonathan Lai , Alexandra Chronopoulou , Manaal Faruqui , Mohit Bansal , Tsendsuren Munkhdalai

Recently, the development of pre-trained language models has brought natural language processing (NLP) tasks to the new state-of-the-art. In this paper we explore the efficiency of various pre-trained language models. We pre-train a list of…

Computation and Language · Computer Science 2023-07-27 Tong Guo

This paper proposes a technique for adding a new source or target language to an existing multilingual NMT model without re-training it on the initial set of languages. It consists in replacing the shared vocabulary with a small…

Computation and Language · Computer Science 2021-10-22 Alexandre Berard

The Transformer architecture has become prominent in developing large causal language models. However, mechanisms to explain its capabilities are not well understood. Focused on the training process, here we establish a meta-learning view…

Machine Learning · Computer Science 2024-03-26 Xinbo Wu , Lav R. Varshney

Advancements in Natural Language Processing have enabled specialized language models, but integrating domain-specific knowledge into general-purpose models in multilingual settings remains challenging, particularly for technical vocabulary.…

Computation and Language · Computer Science 2025-03-13 Thibault Rousset , Taisei Kakibuchi , Yusuke Sasaki , Yoshihide Nomura