Martino M. L. Pulici
PulseAugur coverage of Martino M. L. Pulici — every cluster mentioning Martino M. L. Pulici across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
LoRA Adapters Primarily Measure Reading, Not Knowledge Retention, Study Finds
A new arXiv paper titled "Mostly Reading: What Open-Book Scores Measure in Document-Level LoRA Adapters" investigates the effectiveness of LoRA adapters in document-level tasks. Researchers found that open-book scoring,…
-
New research explores knowledge graph integration with LLMs, tackling retrieval and fusion challenges
Researchers are exploring novel methods for integrating knowledge graphs with large language models, addressing limitations in current approaches. One study proposes compiling knowledge graphs into LoRA adapters for par…
-
New MADA-RL framework boosts compact model reasoning with parameter-efficient debate learning
Researchers have developed MADA-RL, a novel post-training framework designed to enhance the reasoning capabilities of compact language models (under 4 billion parameters) using parameter-efficient methods. This framewor…