Papers by Wonjong Rhee
Soft Head Selection for Injecting ICL-Derived Task Embeddings (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) are commonly adapted to downstream tasks using parameter-efficient fine-tuning (PEFT) or in-context learning (ICL). |
| Approach: | They propose a gradient-based method that derives task-specific embeddings from activations using few-shot prompts and injects them during inference. |
| Outcome: | The proposed method outperforms existing methods on open-ended generation, reasoning, and natural language understanding tasks while using fewer trainable parameters. |
Progressive Multimodal Search and Reasoning for Knowledge-Intensive Visual Question Answering (2026.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to knowledge-intensive visual question answering lack mechanisms to revise misdirected reasoning. |
| Approach: | They propose a framework that progressively constructs a structured reasoning trajectory . they use dual-scope queries to retrieve diverse knowledge from heterogeneous knowledge bases . |
| Outcome: | The proposed framework improves retrieval recall and end-to-end answer accuracy. |