7 papers
WikiSkill: Compiling Agent Experience into Persistent Knowledge for Skill Evolution
Liyan Tang, Cyrus Rashtchian, Chun-Sung Ferng +3
Agent skills package specialized knowledge and workflows into reusable resources that extend AI agent capabilities. Recent work automatically discovers such skills from agent exper…
On the Generalization Gap in Self-Evolving Language Model Reasoning
Zhenting Qi, Susanna Maria Baby, Stefanie Anna Baby +5
Recent work suggests that large language models (LLMs) can improve through self-evolution (SE), using supervision signals generated by the model itself. In this work, we ask: under…
The Master Key Hypothesis: Unlocking Cross-Model Capability Transfer via Linear Subspace Alignment
Rishab Balasubramanian, Pin-Jie Lin, Rituraj Sharma +6
We investigate whether post-trained capabilities can be transferred across models without retraining, with a focus on transfer across different model scales. We propose the Master…
: Structure-Originated Reasoning Data Improves Long-Context Reasoning Ability of Large Language Models
Quyet V. Do, Thinh Pham, Nguyen Nguyen +3
We study a QA curation pipeline for improving long-context complex reasoning in large language models (LLMs). Our approach, , constructs high-quality reasoning data through ri…
ATEB: Evaluating and Improving Advanced NLP Tasks for Text Embedding Models
Simeng Han, Frank Palma Gomez, Tu Vu +6
Traditional text embedding benchmarks primarily evaluate embedding models' capabilities to capture semantic similarity. However, more advanced NLP tasks require a deeper understand…
Efficient Model Development through Fine-tuning Transfer
Pin-Jie Lin, Rishab Balasubramanian, Fengyuan Liu +2
Modern LLMs struggle with efficient updates, as each new pretrained model version requires repeating expensive alignment processes. This challenge also applies to domain- or langua…
Few-shot Continual Relation Extraction via Open Information Extraction
Thiem Nguyen, Anh Nguyen, Quyen Tran +4
Typically, Few-shot Continual Relation Extraction (FCRE) models must balance retaining prior knowledge while adapting to new tasks with extremely limited data. However, real-world…