Search

Results for "Span Reproduction"

When Copying Is Hard: Copy-Constrained Decoding for Exact Span Reproduction

arXiv · Oct 20

Researchers at MBZUAI proposed CopyGen, a lightweight, copy-constrained plugin designed to improve the exact reproduction of text spans by large language models (LLMs). CopyGen addresses the challenge of LLMs failing to precisely copy information from context by decomposing the process into distinct decisions for when to copy, how far, and when to stop, and adaptively adjusting logits. Evaluated across three LLM backbones and four benchmark tasks, CopyGen achieved a 35% relative improvement in average exact match without fine-tuning and further accelerated generation. Why it matters: This research from a leading regional AI institution provides a practical solution to enhance LLM reliability for critical applications requiring precise text reproduction, such as code generation or contract analysis.

Two million eggs and growing: Inside KAUST’s coral spawning program

KAUST · May 6

KAUST scientists have successfully expanded coral spawning through controlled systems, producing approximately two million eggs from six coral species this season to support reef research. Researchers at KAUST's Coastal and Marine Resources (CMR) Core Lab developed facilities that replicate environmental signals to trigger spawning beyond natural seasonal cycles, enabling year-round access to gametes and larvae. This program represents the first successful implementation of shifted phase coral spawning in Saudi Arabia. Why it matters: This breakthrough provides unprecedented opportunities for advancing Red Sea reef restoration and marine science, allowing for continuous study of coral early-life resilience.