linear probing
PulseAugur coverage of linear probing — every cluster mentioning linear probing across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
New research explores efficient fine-tuning and compression for vision foundation models
Two recent arXiv papers explore methods for making large vision foundation models more efficient. The first paper, by Niclas Hergenröther, investigates data-efficient fine-tuning techniques like Visual Prompting (VP), L…
-
LLM Training: Part-of-Speech Impacts Concept Augmentation Effectiveness
A new research paper explores how the part of speech (POS) of a word influences the effectiveness of concept augmentation during the post-training phase of large language models (LLMs). The study found that while concep…
-
New early stopping rule for neural networks bypasses training
Researchers have developed a new data-dependent early stopping rule for training neural networks that estimates generalization error analytically, bypassing the need for numerical estimation through gradient descent. Th…
-
LLM activations reveal concept content in text, outperforming surface analysis
Researchers have developed a new method to measure concept content within text by analyzing Large Language Model (LLM) activations, rather than just surface-level word usage. This approach, utilizing linear probing and …
-
New framework aligns recommender foundation models with business metrics · 2 sources tracked
Researchers have developed a novel three-phase post-training framework to better align recommender foundation models with business metrics. This progressive approach separates downstream adaptation, using Linear Probing…
-
LLMs struggle to decode mythological knowledge beyond dominant traditions
A new research paper investigates how 18 open-source large language models represent and decode mythological knowledge. The study found that while models can represent knowledge about deities like Zeus, Jupiter, and Tho…
-
New optimal self-distillation method improves generative model training
Researchers have developed a method called optimal self-distillation (SD) for rectified flow (RF) models, aiming to improve generative model training. This technique involves training a student model on a mix of true RF…
-
LLMs encode essay quality representations linearly, study finds
Researchers have investigated how large language models (LLMs) represent essay quality internally, finding that this information is encoded in a linearly accessible form within the models' representations. This informat…