Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

35810 · Jun 202019922001200920172026
48 results for multitask finetuning

New study finds best language model architecture and pretraining objective for zero-shot tasks.

problem Evaluating which language model architectures and pretraining objectives best enable zero-shot generalization.
method Compared three model architectures and two pretraining objectives across 170 billion tokens, with and without finetuning.
result Causal decoder-only models trained on autoregressive language modeling exhibit strongest zero-shot generalization.

Multitask learning has shown promising performance in many applications and many multitask models have been proposed. In order to identify an effective multitask model for a given multitask problem, we propose a learning framework called learning to multitask (L2MT). To achieve the goal, L2MT exploits historical multit…

2018-05-19abs ↗pdf ↗

Establishes connection between MTDNN and multitask GP, revealing weight correlation as key to task sharing.

problem Limited theoretical understanding of information sharing in MTDNN.
method Derives multitask GP kernels for MTDNN and MTBNN, showing shared hyper-parameters and last layer weights.
result Information sharing in MTDNN is due to weight correlation, not intermediate layer weights.

Deep learning methods such as multitask neural networks have recently been applied to ligand-based virtual screening and other drug discovery applications. Using a set of industrial ADMET datasets, we compare neural networks to standard baseline models and analyze multitask learning effects with both random cross-valid…

2016-06-28abs ↗pdf ↗

This paper presents a new multitask learning framework that learns a shared representation among the tasks, incorporating both task and feature clusters. The jointly-induced clusters yield a shared latent subspace where task relationships are learned more effectively and more generally than in state-of-the-art multitas…

2017-03-03abs ↗pdf ↗

Massively multitask neural architectures provide a learning framework for drug discovery that synthesizes information from many distinct biological sources. To train these architectures at scale, we gather large amounts of data from public sources to create a dataset of nearly 40 million measurements across more than 2…

2015-02-06abs ↗pdf ↗

Unified finetuning of all quantization degrees of freedom achieves state-of-the-art 4-bit quantization.

problem Achieving high accuracy in quantized neural networks while maintaining speed and resource constraints.
method Quantization-aware finetuning (QFT) that jointly optimizes all quantization degrees of freedom.
result 4-bit weight quantization results on-par with state-of-the-art (SoTA) within PTQ constraints.

Proposes a method to balance tasks in multitask learning with a single gradient step update.

problem Balancing tasks in multitask learning to avoid imbalance.
method Gradient-based meta-learning to balance tasks at the gradient level, training shared and task-specific layers separately.
result Achieves state-of-the-art performance on various multitask computer vision problems.

Paper proposes a method to identify negative transfers in multitask learning using surrogate models.

problem Identifying subsets of source tasks that improve target task performance in multitask learning.
method Surrogate modeling to precompute multitask learning performances and approximate them with a linear regression model.
result The approach predicts negative transfers from multiple source tasks to target tasks more accurately than existing methods.

Study shows pretraining and finetuning can effectively tackle covariate shift in linear regression.

problem Linear regression under covariate shift where source and target distributions differ but conditional distribution remains similar.
method Pretraining on source data and finetuning on target data using online SGD.
result Transfer learning with O(N2)O(N^2) source data is as effective as supervised learning with NN target data.

LoRAs enable efficient adaptation of large models; this paper explores processing LoRA weights with machine learning.

problem Efficient processing of low-rank weight decompositions in large finetuned models.
method Developed symmetry-aware invariant and equivariant LoL models to process LoRA weights.
result LoL models can predict CLIP scores, finetuning data attributes, and accuracy on downstream tasks.

This paper introduces self-paced task selection to multitask learning, where instances from more closely related tasks are selected in a progression of easier-to-harder tasks, to emulate an effective human education strategy, but applied to multitask machine learning. We develop the mathematical foundation for the appr…

2017-03-02abs ↗pdf ↗

We discuss a general method to learn data representations from multiple tasks. We provide a justification for this method in both settings of multitask learning and learning-to-learn. The method is illustrated in detail in the special case of linear feature learning. Conditions on the theoretical advantage offered by m…

2015-05-23abs ↗pdf ↗

Multitask Gaussian process regression reduces data generation costs for molecular property prediction.

problem Data bottleneck in training surrogate models for molecular properties.
method Multitask Gaussian process regression over heterogeneous data sources (CC and DFT).
result Predicts at CC-level accuracy with over an order of magnitude reduction in data generation cost.

This paper bridges offline and online RL by studying policy finetuning with a reference policy.

problem Sample-efficient reinforcement learning in online and offline settings.
method Design of policy finetuning algorithms and analysis of sample complexity.
result Theoretical analysis shows that the optimal policy finetuning algorithm is either offline reduction or purely online RL.

Research explores the trade-off between multi-task learning and multitasking in deep neural networks.

problem Trade-off between multi-task learning and multitasking in deep neural networks.
method Meta-learning algorithm to manage the trade-off between shared and separated representations.
result Agent successfully optimizes training strategy based on environment.

We learn multiple hypotheses for related tasks under a latent hierarchical relationship between tasks. We exploit the intuition that for domain adaptation, we wish to share classifier structure, but for multitask learning, we wish to share covariance structure. Our hierarchical model is seen to subsume several previous…

2014-08-09abs ↗pdf ↗

We present Multitask Soft Option Learning(MSOL), a hierarchical multitask framework based on Planning as Inference. MSOL extends the concept of options, using separate variational posteriors for each task, regularized by a shared prior. This ''soft'' version of options avoids several instabilities during training in a …

2019-04-01abs ↗pdf ↗

Vanishing gradients hinder reinforcement finetuning of language models.

problem Vanishing gradients impede the optimization of language models using reinforcement finetuning.
method The study identifies vanishing gradients as a fundamental optimization obstacle in reinforcement finetuning and proposes an initial supervised finetuning phase to mitigate this issue.
result An initial supervised finetuning phase is crucial for successful reinforcement finetuning of language models, as it helps prevent vanishing gradients and maximizes rewards.

New framework reduces LLM complexity by directly finetuning in Boolean domain.

problem Reducing the complexity of large language models (LLMs) while maintaining performance.
method Proposes a novel framework using multi-kernel Boolean parameters for direct finetuning in the Boolean domain.
result Significantly reduces complexity during both finetuning and inference, outperforming recent techniques.

Meta-learning techniques like MAML and Reptile fail to generalize well to out-of-distribution tasks compared to simple finetuning.

problem Generalization of meta-learning techniques in low-data scenarios.
method Investigation of MAML, Reptile, and finetuning on various tasks.
result MAML and Reptile specialize for fast adaptation but fail to generalize to out-of-distribution tasks.

The study calculates the risk of semi-supervised multitask learning on Gaussian mixtures.

problem Understanding the risk in semi-supervised multitask learning on Gaussian mixtures.
method Statistical physics methods applied to Gaussian mixture models.
result The study evaluates the performance gain of learning tasks together versus separately.

New algorithm improves multitask learning across diverse agents.

problem Performance degradation in decentralized learning with heterogeneous objectives.
method Developed an exact subspace diffusion algorithm for multitask learning over networks.
result The algorithm outperforms alternatives in noisy gradient approximations.

Paper introduces multitask neural networks for efficient stochastic control problems.

problem Infeasibility of simulating state variables in some stochastic control problems.
method Multitask neural networks with dynamic task balancing.
result Multitask neural networks outperform state-of-the-art approaches in derivatives pricing problems.

Scalable GP model handles functional covariates and multitasks.

problem Uncertainty quantification in complex mechanical systems with time-dependent inputs.
method Introduced a fully separable kernel structure for functional covariates and multitask problems, leveraging Kronecker structure for scalability.
result The model significantly improves over single task GPs, requiring fewer samples for accurate predictions.

Optimal multitask learning method for sparse heterogeneous datasets.

problem Efficiently learning from multiple related datasets with sparse task-specific differences.
method MOLAR estimator, combining weighted median and shrinkage.
result Improves estimation error dependence on data dimension compared to task-wise least squares.

Unified multitask learning framework for mixed-type outcomes.

problem Difficulty in formulating a unified objective for tasks with different outcomes.
method Multitask transformation framework with shared sparsity, using deep neural networks and rank-based optimization.
result Improved prediction and variable selection across continuous, binary, and mixed outcomes.

A new approach simplifies multitask Gaussian processes without rank approximations.

problem Handling multioutput regression problems with conditionally dependent tasks.
method Introduces a novel approach to reduce multitask learning to univariate GPs, eliminating the need for rank approximations.
result Accurately recovers multitask covariance and noise matrices with fewer parameters, improving performance and reducing overfitting risk.

ConMeZO speeds up zeroth-order optimization for large language models.

problem Slow convergence in high-dimensional parameter spaces of large language models.
method Adaptive directional sampling in a cone centered around a momentum estimate.
result Achieves the same convergence rate as MeZO but up to 2X faster.

New algorithm REFUEL shows multitask representation learning is more sample-efficient in RL.

problem Understanding the benefit of representation learning in reinforcement learning.
method Developed REFUEL algorithm for multitask low-rank RL, analyzing both upstream and downstream tasks.
result Multitask representation learning is provably more sample-efficient than individual task learning.

Vision transformers benefit from non-smooth components in adaptation.

problem Understanding the role of non-smoothness in vision transformer adaptation.
method Theoretical analysis and extensive experiments on large-scale vision transformers.
result High plasticity of attention modules and feedforward layers leads to better finetuning performance.

Automated protein function prediction is a challenging problem with distinctive features, such as the hierarchical organization of protein functions and the scarcity of annotated proteins for most biological functions. We propose a multitask learning algorithm addressing both issues. Unlike standard multitask algorithm…

2016-11-03abs ↗pdf ↗

Paper introduces RiskEmbed, a finetuned model for financial risk management.

problem Improving retrieval accuracy in financial question-answering systems.
method Curated dataset and finetuned BERT model for financial domain.
result RiskEmbed significantly outperforms general-purpose and financial embedding models.