10 papers
How Much Can We Prune? Exploring the Performance Limits of Visual Token Pruning
Janghoon Cho, Shubhankar Borse, Reza Pourreza +4
Visual token pruning is increasingly used to reduce the memory and inference costs of vision-language models. Existing work has largely focused on designing method-specific selecto…
Feedback Adaptation for Retrieval-Augmented Generation
Jihwan Bang, Seunghan Yang, Kyuhong Shim +3
Retrieval-Augmented Generation (RAG) systems are typically evaluated under static assumptions, despite being frequently corrected through user or expert feedback in deployment. Exi…
Think Straight, Stop Smart: Structured Reasoning for Efficient Multi-Hop RAG
Jihwan Bang, Juntae Lee, Seunghan Yang +1
Multi-hop retrieval-augmented generation (RAG) is a promising strategy for complex reasoning, yet existing iterative prompting approaches remain inefficient. They often regenerate…
CIFLEX: Contextual Instruction Flow for Sub-task Execution in Multi-Turn Interactions with a Single On-Device LLM
Juntae Lee, Jihwan Bang, Seunghan Yang +1
We present CIFLEX (Contextual Instruction Flow for Sub-task Execution), which is a novel execution system for efficient sub-task handling in multi-turn interactions with a single o…
Learning Contextual Retrieval for Robust Conversational Search
Seunghan Yang, Juntae Lee, Jihwan Bang +3
Effective conversational search demands a deep understanding of user intent across multiple dialogue turns. Users frequently use abbreviations and shift topics in the middle of con…
Chain-of-Rank: Enhancing Large Language Models for Domain-Specific RAG in Edge Device
Juntae Lee, Jihwan Bang, Seunghan Yang +2
Retrieval-augmented generation (RAG) with large language models (LLMs) is especially valuable in specialized domains, where precision is critical. To more specialize the LLMs into…
Unlocking Transfer Learning for Open-World Few-Shot Recognition
Byeonggeun Kim, Juntae Lee, Kyuhong Shim +1
Few-Shot Open-Set Recognition (FSOSR) targets a critical real-world challenge, aiming to categorize inputs into known categories, termed closed-set classes, while identifying open-…
Crayon: Customized On-Device LLM via Instant Adapter Blending and Edge-Server Hybrid Inference
Jihwan Bang, Juntae Lee, Kyuhong Shim +2
The customization of large language models (LLMs) for user-specified tasks gets important. However, maintaining all the customized LLMs on cloud servers incurs substantial memory a…
Scalable Weight Reparametrization for Efficient Transfer Learning
Byeonggeun Kim, Jun-Tae Lee, Seunghan yang +1
This paper proposes a novel, efficient transfer learning method, called Scalable Weight Reparametrization (SWR) that is efficient and effective for multiple downstream tasks. Effic…
Domain Generalization with Relaxed Instance Frequency-wise Normalization for Multi-device Acoustic Scene Classification
Byeonggeun Kim, Seunghan Yang, Jangho Kim +3
While using two-dimensional convolutional neural networks (2D-CNNs) in image processing, it is possible to manipulate domain information using channel statistics, and instance norm…