BERT fine-tuning is unstable due to optimization issues, not forgetting or dataset size.
problem Stability of fine-tuning BERT-based models across different random seeds.
method Analysis of BERT, RoBERTa, and ALBERT fine-tuned on GLUE datasets, identifying optimization difficulties as the cause of instability.
result Fine-tuning instability is due to optimization difficulties leading to vanishing gradients, not forgetting or dataset size.
The paper develops a theory for iterative self-improvement of models, proving conditions for better performance with easy-to-hard curricula.
problem Lack of theoretical foundation for iterative self-improvement in practical settings.
method Modeling self-improvement as maximum-likelihood fine-tuning on reward-filtered distributions and proving finite-sample guarantees.
result Explicit feedback loop and conditions for better performance with easy-to-hard curricula.
Learning shrinks hard tail, improving inference performance.
problem Improving inference performance in neural networks.
method Latent Instance Difficulty (LID) model analyzing fine-tuning of neural networks.
result Training-dependent inference scaling, with βexteff growing with sample size before saturating. Paper defines a new distance metric for comparing learning tasks.
problem Comparing difficulty of learning tasks between source and target.
method Information geometry, optimal transport, coupled transfer distance.
result Coupled transfer distance correlates with fine-tuning difficulty.
Novel approach trains LLMs for inductive reasoning using probabilistic programs.
problem Training LLMs for inductive reasoning with sparse, ambiguous data.
method Program-based Posterior Training (PPT) using probabilistic inference.
result Significant improvement in estimation accuracy and alignment with human judgments.
New method improves language model fine-tuning without forgetting.
problem Fine-tuning language models to match specific distributions without forgetting.
method Combines Distribution Matching and Reinforcement Learning techniques.
result Adding a baseline improves constraint satisfaction, stability, and efficiency.
Curriculum learning improves model training efficiency and performance.
problem Expensive training of reasoning models using human or synthetic data.
method Autocurriculum, where models use their own performance to decide training problems.
result Autocurriculum reduces training costs and improves model performance.
This study introduces balanced DRPS and OrderedLogitNN for better QDE of discrete-level questions.
problem Lack of ordinal regression methods and fair evaluation metrics for discrete-level QDE.
method Introduces balanced DRPS and OrderedLogitNN, fine-tunes BERT on RACE++ and ARC datasets.
result OrderedLogitNN outperforms other models on complex QDE tasks.
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.
This paper tackles few-shot classification by improving GAN-based data augmentation.
problem Improving few-shot classification performance using GANs with limited data.
method Fine-tuning GANs for few-shot classification, addressing training and evaluation challenges.
result Semi-supervised fine-tuning is a more effective approach for few-shot classification with limited data.
Transformer models improve arithmetic accuracy with number decomposition.
problem Transformer models struggle with arithmetic operations without decomposition.
method Fine-tuning models with a pipeline that decomposes numbers into units, tens, etc.
result Accuracy increased by 63% in five-digit addition tasks.
Improved Markov models learn from their mistakes and adapt to problem complexity.
problem Limitations of standard masked discrete diffusion models in reasoning tasks.
method Learning a Markov transition kernel trained on its own outputs, allowing remasking and adaptation.
result Significant improvement in solving reasoning problems, especially Sudoku-Extreme and Countdown-4.
Medical ultrasound technology is widely used in routine clinical applications such as disease diagnosis and treatment as well as other applications like real-time monitoring of human tongue shapes and motions as visual feedback in second language training. Due to the low-contrast characteristic and noisy nature of ultr…
New measure FTC quantifies how much a ReLU network can fine-tune.
problem Analyzing memorization capacity in fine-tuned neural networks.
method Defined Fine-Tuning Capacity (FTC) for additive fine-tuning of ReLU networks.
result Upper and lower bounds on FTC for 2 and 3-layer ReLU networks.
This paper re-evaluates hyperparameters for fine-tuning pre-trained models.
problem Current hyperparameter settings for fine-tuning are often ad-hoc and fixed.
method Empirical evaluation of learning rate, batch size, and momentum for fine-tuning.
result Optimal hyperparameters are not only dataset-dependent but also sensitive to domain similarity.
New method quantifies uncertainty in fine-tuned LLMs using LoRA ensembles.
problem Uncertainty in fine-tuned LLMs and how to trust their predictions.
method Posterior approximations using low-rank adaptation ensembles.
result Unexpected retention of acquired knowledge during fine-tuning in overfitting regime.
ConCuR generates high-quality CUDA kernels with concise reasoning traces.
problem Scarce high-quality data for kernel generation.
method Developed a pipeline to generate and curate high-quality CUDA kernels with reasoning traces.
result Our model achieves significant improvements in KernelBench setup.
Improved code translation by preserving structure with composed fine-tuning.
problem Improving code translation accuracy with unlabeled code outputs.
method Pre-trained denoiser to capture output structure, composed fine-tuning to fine-tune predictor.
result Composed fine-tuning significantly improves generalization over standard fine-tuning.
A new algorithm is proposed which accelerates the mini-batch k-means algorithm of Sculley (2010) by using the distance bounding approach of Elkan (2003). We argue that, when incorporating distance bounds into a mini-batch algorithm, already used data should preferentially be reused. To this end we propose using nested …
The paper introduces a Hessian-based method to improve generalization in fine-tuned deep neural networks.
problem Improving generalization in fine-tuned deep neural networks, especially in noisy conditions.
method PAC-Bayesian analysis to identify a Hessian-based distance measure, proving generalization bounds, and developing an algorithm with a generalization error guarantee.
result Hessian-based distance measure correlates well with observed generalization gaps and can match the scale of these gaps in practice.
A simple DQN-based multi-agent RL system for binary actions.
problem Complexity and training issues in multi-agent reinforcement learning.
method Shared state and rewards, agent-specific actions, experience replay pool.
result Better performance and faster convergence compared to conventional methods.
Fine-tuning LLMs improves capability but harms safety, study finds.
problem Balancing capability and safety in LLM fine-tuning.
method Theoretical framework and numerical experiments for two safety-aware fine-tuning strategies.
result Characterization of fundamental limits of safety-capability trade-off in LLM fine-tuning.
Optimizes sparse fine-tuning for privacy in neural networks.
problem Performance gap between DP-SGD and non-private fine-tuning.
method Optimization-based approach using private gradient information for selecting trainable weights.
result Our selection method leads to better prediction accuracy compared to existing approaches.
Fine-tuning harms in-context learning, but restricting updates to the value matrix improves zero-shot performance.
problem Fine-tuning harms in-context learning, reducing zero-shot performance on unseen tasks.
method Theoretical analysis of linear attention models, identifying conditions for degraded few-shot performance.
result Restricting updates to the value matrix improves zero-shot performance while preserving in-context learning.
Measures consistency of tabular LLM predictions under fine-tuning multiplicity.
problem Conflicting predictions from fine-tuned tabular LLMs.
method Local stability measure in embedding space.
result Probabilistic guarantees on prediction consistency under multiplicity.
Self-play fine-tuning improves diffusion models for text-to-image generation.
problem Plateauing performance of diffusion models after data saturation.
method Self-play fine-tuning (SPIN-Diffusion) using competition among model versions.
result Significantly improved model performance and human preference alignment.
The paper develops a theory linking pretraining and fine-tuning in neural networks.
problem Understanding how initialization choices impact feature learning and generalization in neural networks.
method Analytical theory of diagonal linear networks, deriving generalization error as a function of initialization parameters and task statistics.
result Different initialization choices place networks into four fine-tuning regimes with varying abilities to support feature learning and generalization.
New method reduces fine-tuning cost for reused models.
problem Repeating fine-tuning costs with outdated foundation models.
method Portable Reward Tuning (PRT) trains a reward model to maximize the same loss function as fine-tuning.
result PRT achieves comparable accuracy to inference-time tuning with less inference cost.
Compact models match or exceed GPT's performance in financial news sentiment analysis.
problem Improving financial sentiment analysis models without large computational costs.
method Fine-tuning non-generative, small-sized models (FinBERT, FinDRoBERTa) on a novel market score database.
result Fine-tuned models outperform GPT-3.5 and GPT-4 in zero-shot learning for financial news sentiment analysis.
Unsupervised pre-training improves model generalization, but lacks theoretical understanding.
problem Lack of theoretical understanding of unsupervised pre-training's impact on model generalization.
method Introduces a novel theoretical framework to analyze and enhance generalization.
result Enhances understanding of unsupervised pre-training and fine-tuning, proposing a new regularization method.
LP-FT improves personalized model training in FL by balancing generalization and personalization.
problem Federated Learning struggles with balancing global generalization and local personalization due to non-identical data distributions.
method Adapting Linear Probing followed by full Fine-Tuning (LP-FT) to the FL setting.
result LP-FT outperforms standard fine-tuning in balancing personalization and generalization across various datasets and PFT variants.
Deep neural networks are data hungry models and thus face difficulties when attempting to train on small text datasets. Transfer learning is a potential solution but their effectiveness in the text domain is not as explored as in areas such as image analysis. In this paper, we study the problem of transfer learning for…
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.
Improved fine-tuning with regularization and robustness for noisy labels.
problem Fine-tuning pre-trained models on small datasets can lead to overfitting and memorization.
method PAC-Bayes generalization bound analysis, layer-wise regularization, self-label-correction, label-reweighting.
result Improves performance by 1.76% on average for image classification tasks and 0.75% for few-shot classification.
Simplified trust region method reduces representation change during fine-tuning.
problem Stability and representational collapse in fine-tuning pre-trained models.
method Replaces adversarial objectives with parametric noise in trust region theory.
result Matches or exceeds previous trust region methods in performance and speed.
Adjoint Matching improves flow and diffusion models with reward fine-tuning.
problem Improving generative models with reward fine-tuning.
method Casting reward fine-tuning as stochastic optimal control (SOC) and enforcing a specific noise schedule.
result Adjoint Matching outperforms existing SOC algorithms.
Fine-tunes LLMs to correct bias in predictions.
problem LLMs exhibit bias in predictions from data.
method Supervised fine-tuning with Low-Rank Adaptation (LoRA).
result Fine-tuning corrects bias in both controlled and real-world settings.
Transfer learning, which allows a source task to affect the inductive bias of the target task, is widely used in computer vision. The typical way of conducting transfer learning with deep neural networks is to fine-tune a model pre-trained on the source task using data from the target task. In this paper, we propose an…
Fine-tuning large language models requires minimal data, making them efficient.
problem Achieving state-of-the-art performance with large language models.
method Using BERT as an example, fine-tuning only the most critical layers of the pre-trained model.
result Fine-tuned models are close in parameter space to the pre-trained model, with many good solutions found in sparsified versions.
Fine-tuning improves meta-learning by leveraging shared representations.
problem Meta-learning's challenge in rapidly learning new tasks.
method Theoretical framework and risk bounds on gradient descent fine-tuning.
result Fine-tuning-based methods can provably leverage shared structure.
New method enhances model fine-tuning with minimal data.
problem Improving model performance on new tasks with limited data.
method Introducing α-LoRA, a reparameterization method for fine-tuning. result Enhanced generalization ability of fine-tuned models.
Transformers fine-tuned on synthetic data boost tabular data classification performance.
problem Improving tabular data classification accuracy.
method Fine-tuning ICL-transformers on synthetic datasets with complex decision boundaries.
result Fine-tuned ICL-transformers outperform regular neural networks on real-world datasets.
Paper uses LLMs for financial forecasting, overcoming sequence reasoning and multi-modal challenges.
problem Challenges in financial time series forecasting, especially cross-sequence reasoning and multi-modal signals.
method Combines LLMs with financial data and news, using zero-shot/few-shot inference and instruction-based fine-tuning.
result LLMs can offer explainable financial forecasts, leveraging cross-sequence reasoning and multi-modal information.
A new method constrains deep networks during fine-tuning to improve generalization.
problem Improving generalization of fine-tuned deep networks.
method A neural network generalisation bound based on distance from initial weights constrains the hypothesis class to a small sphere.
result Empirical evaluation shows superior generalization performance compared to existing methods.
Fine-tuning with pre-training data improves performance.
problem Limited training data for tasks.
method Theoretical analysis of excess risk bound and selection of pre-training data subset.
result Improvement in generalization performance with pre-training data.
New method fine-tunes discrete diffusion models for RLHF tasks.
problem Fine-tuning discrete diffusion models with policy gradient methods is challenging.
method Proposed SEPO algorithm for efficient fine-tuning over non-differentiable rewards.
result Numerical experiments show scalability and efficiency of SEPO.
Transfer learning boosts chemically accurate neural network potentials for organic molecules.
problem Developing accurate interatomic potentials from ab-initio data.
method Discriminative fine-tuning of pre-trained neural networks.
result Fine-tuning with energy labels alone can achieve accurate atomic forces.
CoLoRA leverages task similarity to boost fine-tuning efficiency.
problem Efficiently fine-tuning large foundation models with scarce labeled data.
method CoLoRA trains a shared adapter for task similarity and personalized adapters for user-specific tasks.
result CoLoRA significantly boosts fine-tuning performance when tasks are similar.