lab NVIDIA Research · just now · research

LLMs match or beat Bayesian optimization for hyperparameter optimization

from English

A new paper explores using foundational large language models (LLMs) to automate hyperparameter optimization (HPO), a process that typically relies on manual approaches in limited-budget settings. The authors developed a methodology where LLMs suggest hyperparameter configurations based on dataset and model descriptions, which are then iteratively refined according to model performance.

Empirical evaluations on standard benchmarks show that within constrained search budgets, LLMs can match or outperform traditional HPO methods like Bayesian optimization across different models.
The approach treats the code specifying the model as a hyperparameter, allowing the LLM to output it directly for greater flexibility than existing HPO approaches.

This work suggests that LLMs can serve as effective alternatives to traditional optimization techniques when computational resources are limited.

Importance 1/3 Trust 3/3 NVIDIA Research Research paper Training methods

Read original

lab NVIDIA Research · 1h ago · 27 views Live

Source introduces approximate unrolled differentiation for training data attribution

Researchers introduce Source, a new training data attribution (TDA) method that combines the computational efficiency of influence functions with the accuracy of unrolling-based approaches. By using an influence-function-like formula to approximate unrolled differentiation, Source addresses limitations in implicit differentiation methods, such as their failure to account for optimization bias or multi-stage pipelines.

lab NVIDIA Research · 1h ago Live

JacNet learns functions by directly modeling structured Jacobians

The paper introduces JacNet, a neural network architecture that directly learns the Jacobian of an input-output function rather than the mapping itself. This approach allows for precise control over derivative properties, enabling the enforcement of structural priors such as invertibility and k-Lipschitz continuity.

arxiv arXiv cs.CL · 8h ago

MiLMMT-46-v1.0 improves multilingual translation via reference-free post-training

Researchers have released MiLMMT-46-v1.0, an open large language model for multilingual machine translation that utilizes reference-free post-training. Starting from the supervised-finetuned MiLMMT-46-v0.1, the team applied Group Relative Policy Optimization (GRPO) with a reward based on two reference-free quality estimation models gated by language identification.

lab NVIDIA Research · 1d ago · 2 views

Masters distills vision-language models via masking teacher and reinforcing student

Researchers propose Masters, a mask-progressive reinforcement learning framework for distilling large-scale vision-language models into compact versions suitable for edge deployment. The method addresses the instability caused by the size gap between teacher and student models.

lab NVIDIA Research · 1d ago · 2 views

ZPPO uses prompts instead of gradients to improve small vision-language model training

Researchers introduce Zone of Proximal Policy Optimization (ZPPO), a method that injects teacher knowledge into prompts rather than policy gradients to address brittleness in knowledge distillation and drift in reinforcement learning. ZPPO constructs Binary Candidate-included Questions (BCQ) and Negative Candidate-included Questions (NCQ) for hard questions, recirculating them via a replay buffer until the student's accuracy improves or they are evicted.