2 citations · 4 across the 4 of their papers we have counts for
6 papers
When PTQ Calibration Helps or Hurts: Dose-Driven Collapse and Validation-Gated Abstention in Quantized Text Embedders
Hyojung Han, Ryangkyung Kang, Youngeun Kim +1
Learned post-training quantization (PTQ) calibration fits quantizer parameters on a small calibration set, and for a text embedder the natural supervision is retrieval relevance. W…
When Should Enterprise Policy Decisions Be Learned, and When Should They Be Reasoned?
Hyojung Han, Ryangkyung Kang, Youngeun Kim +1
An enterprise agent that answers under a written policy must choose an action before it says anything: answer, ask for a missing fact, ask for confirmation, call one or several too…
SkillRet: A Large-Scale Benchmark for Skill Retrieval in LLM Agents
Ryangkyung Kang, Hongcheol Cho, Youngeun Kim
As LLM agents are increasingly deployed with large libraries of reusable skills, selecting the right skill for a user request has become a critical systems challenge. In small libr…
BanditPAM++: Faster -medoids Clustering
Mo Tiwari, Ryan Kang, Donghyun Lee +4
Clustering is a fundamental task in data science with wide-ranging applications. In -medoids clustering, cluster centers must be actual datapoints and arbitrary distance metrics…
MABSplit: Faster Forest Training Using Multi-Armed Bandits
Mo Tiwari, Ryan Kang, Je-Yong Lee +4
Random forests are some of the most widely used machine learning models today, especially in domains that necessitate interpretability. We present an algorithm that accelerates the…
Faster Maximum Inner Product Search in High Dimensions
Mo Tiwari, Ryan Kang, Je-Yong Lee +5
Maximum Inner Product Search (MIPS) is a ubiquitous task in machine learning applications such as recommendation systems. Given a query vector and atom vectors in -dimension…