4 papers
Beyond RLHF: A Unified Theoretical Framework of Alignment
Jihun Yun, Juno Kim, Jongho Park +4
Alignment via reinforcement learning from human feedback (RLHF) has become the dominant paradigm for controlling the quality of outputs from large language models (LLMs). However,…
Not All Bits Are Equal: Scale-Dependent Memory Optimization Strategies for Reasoning Models
Junhyuck Kim, Ethan Ewer, Taehong Moon +2
While 4-bit quantization has emerged as a memory-optimal choice for non-reasoning models and zero-shot tasks across scales, we show that this universal prescription fails for reaso…
Task Diversity Shortens the ICL Plateau
Jaeyeon Kim, Sehyun Kwon, Joo Young Choi +4
In-context learning (ICL) describes a language model's ability to generate outputs based on a set of input demonstrations and a subsequent query. To understand this remarkable capa…
Lexico: Extreme KV Cache Compression via Sparse Coding over Universal Dictionaries
Junhyuck Kim, Jongho Park, Jaewoong Cho +1
We introduce Lexico, a novel KV cache compression method that leverages sparse coding with a universal dictionary. Our key finding is that key-value cache in modern LLMs can be acc…