9 papers
HyperThink: Text-to-Parameter Hypernetworks for Efficient Reasoning
Donggyun Kim, Jack Lu, Chanwoo Kim +2
Long-form thinking traces can substantially improve the multi-step reasoning performance of large language models (LLMs), but they introduce high inference-time overhead, with late…
PatchKV: Weight-Space Compensation of KV Cache
Chanryeol Lee, Chanhyuk Lee, Yeonwoo Choi +2
Long-context inference with Large Language Models (LLMs) is bottlenecked by the linearly growing memory of the key-value (KV) cache. Existing compression methods reduce the cache t…
Infinite Mask Diffusion for Few-Step Distillation
Jaehoon Yoo, Wonjung Kim, Chanhyuk Lee +1
Masked Diffusion Models (MDMs) have emerged as a promising alternative to autoregressive models in language modeling, offering the advantages of parallel decoding and bidirectional…
Universal Few-Shot Spatial Control for Diffusion Models
Kiet T. Nguyen, Chanhyuk Lee, Donggyun Kim +2
Spatial conditioning in pretrained text-to-image diffusion models has significantly improved fine-grained control over the structure of generated images. However, existing control…
HyperFlow: Gradient-Free Emulation of Few-Shot Fine-Tuning
Donggyun Kim, Chanwoo Kim, Seunghoon Hong
While test-time fine-tuning is beneficial in few-shot learning, the need for multiple backpropagation steps can be prohibitively expensive in real-time or low-resource scenarios. T…
AdaRank: Adaptive Rank Pruning for Enhanced Model Merging
Chanhyuk Lee, Jiho Choi, Chanryeol Lee +2
Model merging has emerged as a promising approach for unifying independently fine-tuned models into an integrated framework, significantly enhancing computational efficiency in mul…
Learning to Merge Tokens via Decoupled Embedding for Efficient Vision Transformers
Dong Hoon Lee, Seunghoon Hong
Recent token reduction methods for Vision Transformers (ViTs) incorporate token merging, which measures the similarities between token embeddings and combines the most similar pair…
Meta-Controller: Few-Shot Imitation of Unseen Embodiments and Tasks in Continuous Control
Seongwoong Cho, Donggyun Kim, Jinwoo Lee +1
Generalizing across robot embodiments and tasks is crucial for adaptive robotic systems. Modular policy learning approaches adapt to new embodiments but are limited to specific tas…
Revisiting Weight Averaging for Model Merging
Jiho Choi, Donggyun Kim, Chanhyuk Lee +1
Model merging aims to build a multi-task learner by combining the parameters of individually fine-tuned models without additional training. While a straightforward approach is to a…