This study examines how the size and alignment of pretraining data affect the performance of large language models on downstream tasks.
problem Understanding how the size and alignment of pretraining data impact the performance of large language models on downstream tasks.
method Investigated the scaling behavior of large language models in a transfer learning setting, focusing on machine translation tasks.
result The size of the finetuning dataset and the distribution alignment between pretraining and downstream data significantly influence the scaling behavior of downstream performance.
Analysis of pretrained models' effectiveness in downstream tasks.
problem Understanding why pretrained models perform well in NLP tasks.
method Analyzed head and prompt tuning approaches using latent variable models.
result Prompt tuning provides stronger guarantees than head tuning.
Paper defines and solves a problem in representation learning to ensure fairness with high confidence.
problem Learning fair representations with high confidence guarantees for all downstream tasks.
method Formally defines the problem, introduces FRG framework, proves high probability fairness, and demonstrates effectiveness empirically.
result FRG framework provides high-confidence guarantees for limiting unfairness across all downstream models and tasks.
Study shows scaling up models doesn't always improve downstream tasks.
problem Understanding why scaling up models doesn't always improve downstream performance.
method Systematic study of 4800 experiments on various models, analyzing performance on 20 downstream tasks.
result Performance on downstream tasks saturates as model size increases, revealing a nonlinear relationship.
Theoretical analysis shows pretext-based self-supervised learning can be boosted by downstream data under certain conditions.
problem Theoretical analysis of pretext-based self-supervised learning and downstream data refinement.
method Theoretical analysis and experiments on synthetic and real-world datasets.
result Theoretical lower bounds and experiments show that downstream data refinement can boost or hurt performance depending on conditions.
New framework explains how larger pre-trained models reduce downstream learning sample complexity.
problem Understanding why larger pre-trained models reduce sample complexity in downstream tasks.
method Introducing a novel framework called Caulking inspired by PEFT methods.
result Improved pre-trained models provably decrease downstream task sample complexity.
Study shows how pretraining robustness transfers to downstream tasks.
problem Understanding how robustness is transferred from pretraining to downstream tasks.
method Theoretical analysis and practical validation of robustness constraints.
result Robustness of a linear predictor on downstream tasks can be constrained by the robustness of its underlying representation.
This paper characterizes VAE training pathologies and their effects on tasks.
problem Characterizing VAE training pathologies and their impact on downstream tasks.
method Concretely characterizing conditions for VAE training pathologies and their connection to specific downstream tasks.
result Connects VAE training pathologies to specific downstream tasks like learning compressed and disentangled representations, adversarial robustness, and semi-supervised learning.
End-to-end approach for weak supervision improves downstream model performance.
problem Data-labeling bottleneck in machine learning applications.
method Directly learning the downstream model by maximizing its agreement with probabilistic labels generated from weak supervision sources.
result Improved performance over prior work in terms of downstream model performance and robustness.
UBM transfers bias mitigation from upstream to downstream tasks efficiently.
problem Bias in fine-tuned language models across various tasks.
method Apply bias mitigation to an upstream model, then fine-tune a downstream model on this mitigated model.
result UBM effects transfer to new downstream tasks, creating less biased models.
Optimal feature transfer identified through bias-variance analysis.
problem Optimizing feature transfer in transfer learning.
method Simple linear model with fine-grained bias-variance decomposition.
result Optimal pretrained feature transform is naturally sparse.
The paper explores how word embeddings affect the stability of downstream NLP models.
problem Small changes in training data can cause significant changes in model predictions.
method Empirical and theoretical analysis of embedding instability, including the introduction of eigenspace instability measure.
result Increasing embedding memory can reduce the disagreement in predictions by 5% to 37%.
This work investigates how neural collapse improves transfer learning for large-scale models.
problem Improving transfer learning for large-scale models with limited labeled data.
method Investigates neural collapse and develops a fine-tuning method using skip-connections.
result Feature collapse on downstream data correlates with higher transfer accuracy.
FWC creates fair synthetic samples for machine learning tasks.
problem Addressing biases in machine learning models for fair decision-making.
method FWC uses an efficient majority minimization algorithm to minimize Wasserstein distance while enforcing demographic parity.
result FWC achieves a competitive fairness-utility tradeoff and reduces biases in predictions from large language models.
This work proposes a new pre-processing method for supervised learning to improve fairness without sacrificing utility.
problem Improving fairness in supervised learning without compromising model performance.
method Task-tailored pre-processing approach that balances fairness and utility.
result The proposed method preserves consistent trade-offs among multiple downstream models and improves fairness in computer vision tasks.
New framework minimizes model complexity for improved few-shot learning.
problem Empirical benefits of pre-training scale with data size but lack theoretical explanation.
method Complexity Minimization framework for meta-representation learning.
result Theoretical analysis shows error rate improves with more meta-training data.
Framework fine-tunes foundation models with semi-supervised learning for downstream tasks and latent spaces.
problem Training foundation models with limited labelled data.
method Mutual information decomposition for downstream and latent spaces, semi-supervised fine-tuning.
result Significant improvements in classification tasks under low-labelled conditions.
Language models help text classification tasks by predicting next words.
problem Lack of theoretical understanding of why language models perform well on downstream tasks.
method Mathematical study of the connection between next word prediction and text classification, formalizing it and quantifying the benefit.
result Language models that are ε-optimal in cross-entropy learn features that can solve classification tasks with linear approximation.
New method predicts wind farm power and wakes using weather patterns.
problem Inefficient and computationally intensive wind energy resource assessment.
method Unsupervised clustering of ERA5 data on wind velocity, WRF simulations at cluster centers, and post-processing.
result Accurate long-term predictions of power and wakes with reduced computational time.
Proposes a new metric to quantify the difference between neural network representations based on downstream task performance.
problem The lack of a consistent metric to measure the difference between neural network representations.
method Introduced the Transferred Discrepancy (TD) metric, which evaluates the difference between representations based on their performance on downstream tasks.
result TD provides fine-grained information for various downstream tasks and can evaluate the effectiveness of different training strategies.
Enhances performance on downstream tasks using multi-domain data.
problem Improving performance on tasks with limited downstream data.
method Deep transfer learning framework leveraging shared and domain-specific features.
result Significantly improves convergence rate for learning Lipschitz functions.
Self-supervised metric learning boosts downstream tasks in multi-view data.
problem Improving distance-based downstream tasks without labeled data.
method Developed a statistical framework to study self-supervised metric learning in multi-view data.
result Self-supervised metric learning improves target distances for various downstream tasks.
Object-centric learning improves generalization and robustness in multi-object scenes.
problem Improving generalization and robustness in neural networks for scenes with multiple objects.
method Training state-of-the-art unsupervised models on multi-object datasets and evaluating segmentation metrics and downstream tasks.
result Object-centric representations are useful for downstream tasks and generally robust to most distribution shifts affecting objects, but less so for less structured shifts.
Optimal Word2Vec hyper-parameters improve NLP tasks.
problem Finding the best Word2Vec hyper-parameters for NLP tasks.
method Empirical evaluation of various hyper-parameter combinations on NLP tasks.
result The best hyper-parameters vary by task, and high analogy scores don't always correlate with performance.
Theoretical study shows adversarial training improves robustness in deep learning models.
problem Ensuring robustness in pre-trained deep learning models.
method Theoretical analysis of adversarial training and feature purification in two-layer neural networks.
result Adversarial training leads to feature purification, making models more robust to attacks.
TaskMet learns a metric to improve model performance on unseen tasks.
problem Deep models trained on one task may struggle on another task due to conflicting objectives.
method TaskMet learns a metric in the prediction space to balance task and prediction losses.
result TaskMet achieves better performance on downstream tasks without altering the prediction model.
The paper finds a surprising positive correlation between upstreamness and downstreamness in global value chains.
problem The puzzling positive correlation between upstreamness and downstreamness in industries and countries.
method Analysis of a simple model of random Input/Output tables and experiments on empirical data.
result Upstreamness and downstreamness of the same industrial sector/country are positively correlated with a slope close to +1.
Action-BED: Task-Driven Bayesian Experimental Design
problem Bayesian experimental design with doubly intractable objectives
method Formulating BED in terms of expected future loss (EFL) and optimising it with stochastic gradients
result Simplified and task-driven framework for BED
DKPS provides guarantees for synthetic data from Transformer models, improving downstream tasks.
problem Lack of labeled data for building performant AI models.
method Data Kernel Perspective Space (DKPS) for mathematical analysis of synthetic data quality.
result Concrete statistical guarantees for the quality of transformer model outputs.
Optimizes ad pruning in sponsored search systems using reinforcement learning.
problem How to efficiently select top K ads from N candidates to maximize revenue.
method Model-free reinforcement learning approach considering downstream as a black-box environment.
result Remarkable improvements in revenue achieved through reinforcement learning.
Proposes FairRR to improve fairness in machine learning models through randomized response.
problem Achieving group fairness in machine learning models.
method Formulates group fairness as optimizing a design matrix in Randomized Response, proposing FairRR.
result Demonstrates FairRR yields excellent model utility and fairness.
Study shows uncertainty of deep learning models can be measured from their embeddings.
problem Uncertainty in contrastive learning models for critical applications.
method Estimating the distribution of training data in embedding space and accounting for local consistency.
result Uncertainty of an embedding vector correlates strongly with downstream accuracy.
New insights into contrastive learning reveal how projectors affect downstream performance.
problem Understanding how projectors in contrastive learning impact downstream linear classification accuracy.
method Identified and modeled two effects: expansion and shrinkage induced by contrastive loss.
result Linear projectors operating in the shrinkage regime hinder downstream classification accuracy.
PEARL combines multiple representation learning methods to enhance model performance.
problem Different representation learning methods extract distinct data aspects, potentially missing important insights.
method Combines multiple representation learning approaches using surrogate loss functions for efficient weight estimation.
result Asymptotically achieves optimal performance in downstream tasks, assigning nonzero weights to correctly specified models.
A comprehensive benchmark of 15 scRNA-seq imputation methods across various datasets and analyses.
problem Imputation of single-cell RNA sequencing data to recover latent transcriptional signals.
method Evaluation of 15 imputation methods across 30 datasets and 6 downstream analyses.
result Traditional methods generally outperform DL-based methods in scRNA-seq data analysis.
This work studies the impact of intra-/inter-class diversity on pre-training datasets and finds a balance for optimal performance.
problem The impact of intra-/inter-class diversity on supervised pre-training datasets and their effect on downstream tasks.
method Empirical study and theoretical analysis of the relationship between diversity types and downstream performance.
result The optimal class-to-sample ratio is invariant to the size of the pre-training dataset and can be predicted.
AI bias arises from human-defined goals, not algorithmic flaws.
problem AI bias due to human-defined goals in LLMs.
method Purpose-conditioned cognition and revealing downstream use of LLM outputs.
result AI bias can be reduced by purpose-aware prompting but not fully by regularization.
Improved VAE representations lead to better image classification.
problem VAE representations are inferior to non-latent models for image classification.
method Used a decoder that prefers local features, improving global feature capture in latent variables.
result Significant improvement in downstream semantic classification tasks.
Pre-training improves model coverage, crucial for downstream performance.
problem Understanding why pre-training enhances model performance.
method Coverage principle, focusing on next-token prediction and model quality.
result Coverage generalizes faster than cross-entropy, improving downstream performance.
GPT-GNN pre-trains GNNs on unlabeled graphs to improve downstream performance.
problem Training GNNs requires labeled data, which is expensive.
method Generative pre-training of GNNs on unlabeled data with self-supervision.
result GPT-GNN significantly outperforms state-of-the-art GNNs without pre-training.
Fine-tunes GNNs by preserving generative patterns to improve transferability.
problem Vanilla fine-tuning fails due to structural divergence between pre-training and downstream graphs.
method G-Tuning, which reconstructs the generative patterns of the downstream graph using graphon bases.
result G-Tuning achieves an average improvement of 0.5% and 2.6% on in-domain and out-of-domain transfer learning experiments.
New algorithm balances spatial data approximation and prediction accuracy.
problem Lack of methods considering spatial correlation and downstream modeling in dimension reduction.
method Formalizes approximation and modeling utility as metrics, proposes a balanced algorithm.
result Optimal trade-off between approximation accuracy and downstream modeling utility.
Paper examines adversarial attacks on weather forecasting models, focusing on TC trajectory prediction.
problem Adversarial attacks can mislead downstream TC trajectory predictions in DLWF models.
method Proposes Cyc-Attack, a method using a surrogate model and skewness-aware loss function to generate adversarial TC paths.
result Cyc-Attack achieves higher true positive rates and lower false alarm rates compared to conventional methods.
PRAGMA models financial event sequences for various banking tasks.
problem Handling diverse financial data for multiple applications.
method Pre-training a Transformer model on a large banking event corpus with a self-supervised objective.
result PRAGMA achieves superior performance across multiple financial domains from raw event sequences.
Mask-reconstruction pretraining helps in downstream tasks by capturing more semantic features.
problem How mask-reconstruction pretraining helps in downstream tasks and why it surpasses supervised learning.
method Theoretical analysis and experimental validation of mask-reconstruction pretraining (MRP) on auto-encoders.
result MRP provably captures more semantic features than supervised learning, leading to better performance in downstream tasks.
Survey on uncertainty in GNNs for improved model reliability.
problem Inherent randomness and model training errors lead to unstable GNN predictions.
method Comprehensive overview of GNN uncertainty theory and methods.
result Integration of uncertainty in GNNs enhances model performance and reliability.
New method uses imperfect LLM annotations for valid statistical inference in social science.
problem Inaccurate large language model annotations in social science research.
method Design-based supervised learning (DSL) combining imperfect LLM surrogates with gold-standard labels.
result DSL provides valid statistical inference with comparable predictive accuracy to existing methods.
New measure assesses time series pre-training data quality without labels.
problem Challenges in collecting diverse pre-training datasets for time series classification.
method Contrastive-learning-based foundation model and contrastive accuracy measure.
result Contrastive accuracy correlates with model performance on downstream tasks.