This work forms SLM-LLM collaboration as an information acquisition problem, under an API budget constraint, and develops a three-stage RLVR framework that learns whether to call the advisor, how to formulate useful queries, and how to integrate the collaboration into the reasoning process by jointly refining advisor i...
Drawing-Recode is a framework that generates Parametric CAD sequences as CAD code from raster 2D CAD drawings using a Large Language Model to generate CAD code in the Structured Parametric CAD Code (SPCC) format and is expected to contribute to digitizing raster 2D CAD drawings in industrial settings and to part reprod...
Mingi Kim, Yongjun Kim, Hyungki Kim· arXiv.org· 0 citations
A lightweight gate is trained that predicts high-NLL tokens from early-layer hidden states, enabling token selection during mid-prefill via early exit, motivated by the hypothesis that less predictable tokens are less recoverable from surrounding context and therefore benefit more from late-position repetition.
Andikawati P Widjaja, Yongjun Kim, Hyounghun Kim et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.