From the 1 of 7 linked papers with an AI index.
7 papers
How to Guide LLM Generation: Dual-Surrogate Guided Search for Automated Heuristic Design
Yuhan Wang, Chaoda Peng, Xingyu Wu +2
The paper introduces DualSurrogate Guided Search, a method that uses two surrogate models to predict the outcome and utility of candidate heuristic code before querying a large lan…
Semantics-Aware Bilevel Co-Evolution: Towards Automated Multicomponent Algorithm Design
Zhiyao Zhang, Shenghao Wu, Xingyu Wu +1
LLM-assisted evolutionary search (LES) has emerged as a promising paradigm for automated algorithm design. However, existing methods usually suffer from two inherent limitations wh…
Towards Adaptive Continual Model Merging via Manifold-Aware Expert Evolution
Haiyun Qiu, Xingyu Wu, Kay Chen Tan
Continual Model Merging (CMM) sequentially integrates task-specific models into a unified architecture without intensive retraining. However, existing CMM methods are hindered by a…
Fine-Grained Model Merging via Modular Expert Recombination
Haiyun Qiu, Xingyu Wu, Liang Feng +1
Model merging constructs versatile models by integrating task-specific models without requiring labeled data or expensive joint retraining. Although recent methods improve adaptabi…
Diversity-Aware Policy Optimization for Large Language Model Reasoning
Jian Yao, Ran Cheng, Xingyu Wu +2
The reasoning capabilities of large language models (LLMs) have advanced rapidly, particularly following the release of DeepSeek R1, which has inspired a surge of research into dat…
LLM Cannot Discover Causality, and Should Be Restricted to Non-Decisional Support in Causal Discovery
Xingyu Wu, Kui Yu, Jibin Wu +1
This paper critically re-evaluates LLMs' role in causal discovery and argues against their direct involvement in determining causal relationships. We demonstrate that LLMs' autoreg…