Pruning Small Pre-Trained Weights Irreversibly and Monotonically Impairs "Difficult" Downstream Tasks in LLMs [ICML 2024]

Official PyTorch implementation of Pruning Small Pre-Trained Weights Irreversibly and Monotonically Impairs "Difficult" Downstream Tasks in LLMs

Lu Yin*, Ajay Jaiswal*, Shiwei Liu,Souvik Kundu, Zhangyang Wang

University of Texas at Austin, University of Surrey, Eindhoven University of Technology, University of Oxford, Intel Labs

The code can be contacted at ajayjaiswal@utexas.edu and l.yin@surrey.ac.uk.

Table of contents

• Installation
• Various Task Diffuclty
• Are Pre-trained Magnitude Values Indeed the True Gem?

---

Abstract

We present $\textit{Junk DNA Hypothesis}$ by adopting a novel $\textit{task-centric}$ angle for the pre-trained weights of large language models (LLMs). It has been believed that weights in LLMs contain significant redundancy, leading to the conception that a considerable chunk of the parameters can be removed by $\textit{pruning}$ without compromising performance. Contrary to this belief, this paper presents a $\textit{counter-argument}$: small-magnitude weights of pre-trained model weights encode vital knowledge essential for tackling difficult downstream tasks - manifested as the $\textbf{monotonic}$ $\textbf{relationship}$ between the performance drop of downstream tasks across the difficulty spectrum, as we prune more pre-trained weights by magnitude. Moreover, we reveal that these seemingly inconsequential weights can result in $\textbf{irreparable loss}$ of knowledge and performance degradation in difficult tasks, even when downstream continual training is allowed. Interestingly, our evaluations show that the other popular compression, namely $\textbf{quantization}$ $\textbf{fail}$ to exhibit similar ``monotonic" effect and does not as convincingly disentangle this task-difficulty information. To study formally, we introduce several quantifiable metrics to $\textit{gauge the downstream task difficulty}$: (a) within the same task category, and (b) across different task categories. Our extensive experiments substantiate the Junk DNA Hypothesis across a diverse range of model sizes, tasks, datasets, and even pruning methods.

Installation

Please check INSTALL.md for installation instructions.

Various Task difficulty

We provide a quick overview of the arguments:

• --model_name_or_path: The identifier for the model on the Hugging Face model hub.
• --TASK_NAME: the name of the fine-tuned tasks.
• --sparsity: Denotes the percentage of weights to be pruned.
• --sparse_init: Specifies the type of sparsity [sparse_nm, sparse_unstuctured] .
• --method: a flag to of the output_dir.

Are Pre-trained Magnitude Values Indeed the True Gem?

Scripts example

-- vary sparse_method with

• --freeze_weights: Sparse Transfer
• --freeze_weights_frompretrain: Dense Transfer with Freezing
• -or leave it emply: Sparse to Dense Transfer

cd ./GLUE_tasks 

for seed in 41 42 43
do
  for TASK_NAME in QNLI 
  do 
    for sparsity in 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
    do
      for validation_split_percentage in 100
      do
      python Glue_prune_oneshot.py \
        --method Glue_noembed_freeze_weights \
        --validation_split_percentage $validation_split_percentage \
        --sparse_method \
        --noembed \
        --sparsity $sparsity \
        --model_name_or_path roberta-large \
        --task_name $TASK_NAME \
        --max_length 512 \
        --per_device_train_batch_size 16 \
        --learning_rate 2e-5 \
        --num_train_epochs 3 \
        --seed $seed \
        --output_dir ./roberta/Glue_noembed_freeze_weights/$TASK_NAME/$sparsity/$validation_split_percentage/$seed/
      done
    done
  done
done

Results

Task Difficulty Setting: Varying the Adequacy of Target Domain Data

---

Scripts Example

TO BE RELEASED SOON

Task Difficulty Setting: Varying the Option Count in Multiple-Choice QA Setting

---

Scripts Example

TO BE RELEASED SOON

Task Difficulty Setting: Varying Context Length for Retrieval-Augmented QA

---

TO BE RELEASED SOON

Task Difficulty Setting: Varying Number of K-Shot Examples for In-Context Learning

TO BE RELEASED SOON

Task Difficulty Setting: Estimating LLM-Facing Task Difficulty by Normalized Human-LLM Performance Gap

TO BE RELEASED SOON

Task Difficulty Setting: Factoid-Based vs. Multiple-Choice QA

TO BE RELEASED SOON

Scripts example

TO BE RELEASED SOON

Citation

if you find this repo is helpful, please cite

@article{yin2024junk,
  title={Pruning Small Pre-Trained Weights Irreversibly and Monotonically Impairs "Difficult" Downstream Tasks in LLMs},
  author={Yin, Lu and Jaiswal, Ajay and Liu, Shiwei  and Kundu, Souvik and Wang, Zhangyang},
  journal={arXiv preprint arXiv:2310.02277v2},
  year={2024}
}