Synthetic Pre-pretraining Survives Scale, but Not as a Grammatical Prior
Paper • 2609.39827 • Published • 12
AI & ML interests
None defined yet.
Recent Activity
View all activity
michal-stefanik
authored a
paper 5 days ago
michal-stefanik
updated a
dataset 20 days ago
michal-stefanik
published a
dataset 20 days ago
michal-stefanik
published a
dataset 2 months ago
Tresiwald
updated a
dataset 5 months ago
Tresiwald
published a
dataset 5 months ago
rbelanec
authored 4
papers 7 months ago
PEFT-Bench: A Parameter-Efficient Fine-Tuning Methods Benchmark
Paper • 2511.21285 • Published • 2
PEFT-Factory: Unified Parameter-Efficient Fine-Tuning of Autoregressive Large Language Models
Paper • 2512.02764 • Published • 1
Revisiting Prompt Sensitivity in Large Language Models for Text Classification: The Role of Prompt Underspecification
Paper • 2602.04297 • Published • 1
Better as Generators Than Classifiers: Leveraging LLMs and Synthetic Data for Low-Resource Multilingual Classification
Paper • 2601.16278 • Published • 1
rbelanec
authored 3
papers 12 months ago
Task Prompt Vectors: Effective Initialization through Multi-Task Soft-Prompt Transfer
Paper • 2408.01119 • Published • 1
Controlling the Output of a Generative Model by Latent Feature Vector Shifting
Paper • 2311.08850 • Published • 1
Fighting Randomness with Randomness: Mitigating Optimisation Instability of Fine-Tuning using Delayed Ensemble and Noisy Interpolation
Paper • 2406.12471 • Published • 1
Tresiwald
authored 2
papers over 2 years ago