Loft
PulseAugur coverage of Loft — every cluster mentioning Loft across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
New LOFT method enables efficient machine unlearning in low-dimensional feature subspaces
Researchers have introduced LOFT, a novel machine unlearning method that operates within low-dimensional feature subspaces. This approach aims to remove specific data's influence from a pretrained model while retaining …
-
Stratified Sparse Attention boosts LLM long-context inference speed and accuracy
Researchers have developed Stratified Sparse Attention (SSA), a new framework designed to improve the efficiency and accuracy of large language model inference over long contexts. SSA combines deterministic retrieval of…
-
Mastodon user reflects on Machida shopping trip and DeNA's CS victory · 4 sources tracked
On October 17, 2026, an individual documented their day on Mastodon, starting with health and medication records. They visited Machida, stopping at McDonald's for a break before exploring new stores like Surugaya and Lo…
-
LOFT framework enhances parameter-efficient fine-tuning with task-aware support selection
Researchers have introduced LOFT, a novel framework for low-rank orthogonal parameter-efficient fine-tuning (PEFT). This method explicitly separates the adaptation subspace from the transformation applied within it, off…