3.7k citations · 12.4k across the 59 of their papers we have counts for
65 papers
Merging by Matching Models in Task Parameter Subspaces
Derek Tam, Mohit Bansal, Colin Raffel
Model merging aims to cheaply combine individual task-specific models into a single multitask model. In this work, we view past merging methods as leveraging different notions of a…
ComPEFT: Compression for Communicating Parameter Efficient Updates via Sparsification and Quantization
Prateek Yadav, Leshem Choshen, Colin Raffel +1
Parameter-efficient fine-tuning (PEFT) techniques make it possible to efficiently adapt a language model to create "expert" models that specialize to new tasks or domains. Recent t…
Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Model
Haikang Deng, Colin Raffel
While large language models have proven effective in a huge range of downstream applications, they often generate text that is problematic or lacks a desired attribute. In this pap…
Uncovering Model Processing Strategies with Non-Negative Per-Example Fisher Factorization
Michael Matena, Colin Raffel
We introduce NPEFF (Non-Negative Per-Example Fisher Factorization), an interpretability method that aims to uncover strategies used by a model to generate its predictions. NPEFF de…
Git-Theta: A Git Extension for Collaborative Development of Machine Learning Models
Nikhil Kandpal, Brian Lester, Mohammed Muqeeth +6
Currently, most machine learning models are trained by centralized teams and are rarely updated. In contrast, open-source software development involves the iterative development of…
Soft Merging of Experts with Adaptive Routing
Mohammed Muqeeth, Haokun Liu, Colin Raffel
Sparsely activated neural networks with conditional computation learn to route their inputs through different "expert" subnetworks, providing a form of modularity that densely acti…
TIES-Merging: Resolving Interference When Merging Models
Prateek Yadav, Derek Tam, Leshem Choshen +2
Transfer learning - i.e., further fine-tuning a pre-trained model on a downstream task - can confer significant advantages, including improved downstream performance, faster conver…
Scaling Data-Constrained Language Models
Niklas Muennighoff, Alexander M. Rush, Boaz Barak +6
The current trend of scaling language models involves increasing both parameter count and training dataset size. Extrapolating this trend suggests that training dataset size may so…
Knowledge is a Region in Weight Space for Fine-tuned Language Models
Almog Gueta, Elad Venezian, Colin Raffel +3
Research on neural networks has focused on understanding a single model trained on a single dataset. However, relatively little is known about the relationships between different m…
Improving Few-Shot Generalization by Exploring and Exploiting Auxiliary Data
Alon Albalak, Colin Raffel, William Yang Wang
Few-shot learning is valuable in many real-world applications, but learning a generalizable model without overfitting to the few labeled datapoints is challenging. In this work, we…
ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning
Shachar Don-Yehiya, Elad Venezian, Colin Raffel +3
We propose a new paradigm to continually evolve pretrained models, denoted ColD Fusion. It provides the benefits of multitask learning but leverages distributed computation with li…
What Language Model to Train if You Have One Million GPU Hours?
Teven Le Scao, Thomas Wang, Daniel Hesslow +16
The crystallization of modeling methods around the Transformer architecture has been a boon for practitioners. Simple, well-motivated architectural variations can transfer across t…
Evaluating the Factual Consistency of Large Language Models Through News Summarization
Derek Tam, Anisha Mascarenhas, Shiyue Zhang +3
While large language models (LLMs) have proven to be effective on a large variety of tasks, they are also known to hallucinate information. To measure whether an LLM prefers factua…
Large Language Models Struggle to Learn Long-Tail Knowledge
Nikhil Kandpal, Haikang Deng, Adam Roberts +2
The Internet contains a wealth of knowledge -- from the birthdays of historical figures to tutorials on how to code -- all of which may be learned by language models. However, whil…
BLOOM: A 176B-Parameter Open-Access Multilingual Language Model
BigScience Workshop, :, Teven Le Scao +391
Large language models (LLMs) have been shown to be able to perform new tasks based on a few demonstrations or natural language instructions. While these capabilities have led to wi…
Crosslingual Generalization through Multitask Finetuning
Niklas Muennighoff, Thomas Wang, Lintang Sutawika +16
Multitask prompted finetuning (MTF) has been shown to help large language models generalize to new tasks in a zero-shot setting, but so far explorations of MTF have focused on Engl…
Compositional Generalization in Unsupervised Compositional Representation Learning: A Study on Disentanglement and Emergent Language
Zhenlin Xu, Marc Niethammer, Colin Raffel
Deep learning models struggle with compositional generalization, i.e. the ability to recognize or generate novel combinations of observed elementary concepts. In hopes of enabling…
A Combinatorial Perspective on the Optimization of Shallow ReLU Networks
Michael Matena, Colin Raffel
The NP-hard problem of optimizing a shallow ReLU network can be characterized as a combinatorial search over each training example's activation pattern followed by a constrained co…
Bidirectional Language Models Are Also Few-shot Learners
Ajay Patel, Bryan Li, Mohammad Sadegh Rasooli +3
Large language models such as GPT-3 (Brown et al., 2020) can perform arbitrary tasks without undergoing fine-tuning after being prompted with only a few labeled examples. An arbitr…
Petals: Collaborative Inference and Fine-tuning of Large Models
Alexander Borzunov, Dmitry Baranchuk, Tim Dettmers +5
Many NLP tasks benefit from using large language models (LLMs) that often have more than 100 billion parameters. With the release of BLOOM-176B and OPT-175B, everyone can download…
Efficient Methods for Natural Language Processing: A Survey
Marcos Treviso, Ji-Ung Lee, Tianchu Ji +19
Recent work in natural language processing (NLP) has yielded appealing results from scaling model parameters and training data; however, using only scale to improve performance mea…
Emergent Abilities of Large Language Models
Jason Wei, Yi Tay, Rishi Bommasani +13
Scaling up language models has been shown to predictably improve performance and sample efficiency on a wide range of downstream tasks. This paper instead discusses an unpredictabl…
Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models
Aarohi Srivastava, Abhinav Rastogi, Abhishek Rao +448
Language models demonstrate both quantitative improvement and new qualitative capabilities with increasing scale. Despite their potentially transformative impact, these new capabil…
Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning
Haokun Liu, Derek Tam, Mohammed Muqeeth +4
Few-shot in-context learning (ICL) enables pre-trained language models to perform a previously-unseen task without any gradient-based training by feeding a small number of training…