Yongkang Liu

Publications

Arxiv

High-Rank Structured Modulation for Parameter-Efficient Fine-Tuning
Liu, Y., Li, X., Zhao, M., Zhang, S., Wang, Z., Li, Q., Feng, S., Ren, F., Wang, D., & Schutze, H.

Good Teachers, Better Students: A Survey of Reward Models for LLM
L Wang, Z Wang, J Lin, Q Zeng, J Zhang, T Wu, X Zhang, J Li, Z Wang, Liu, Y. …

A survey on entropy mechanism in large reasoning models
Z Wang, X Xu, H Wang, Y Ye, Y Li, L Wang, H Tan, P Wang, S Feng, …

Scaling Intelligence Through Model Merging: A Comprehensive Survey
Z Wang, Liu, Y., Y Luo, M Wang, Z Song, S Feng, X Yang, D Lin, D Wang, …

2026

  • From Parameter Dynamics to Risk Scoring: Quantifying Sample-Level Safety Degradation in LLM Fine-tuning （CCF A ICML)
    Wang, X., Zhang, Y., Liu, Y, Yang, X., Wang, Z., Feng, S., & Wang, D.
  • A Systematic Analysis of the Impact of Persona Steering on LLM Capabilities （CCF B Cognitive Science)
    Chen, J., Wang, M., Xie, T., Feng, S., & Liu, Y
  • NEAT: Neuron-Based Early Exit for Large Reasoning Models （CCF A ACL Findings)
    Liu, K., Liu, Y., Yang, X., Wang, P., Zhang, W., Feng, S., Zhang, Y., & Wang, D.
  • Look Within or Look Beyond? A Theoretical Comparison Between Parameter-Efficient and Full Fine-Tuning （CCF A ACL Main)
    Liu, Y., Xu, X., Nie, E., Wang, Z., Feng, S., Wang, D., Li, Q., & Schutze, H.
  • Why do more experts fail? a theoretical analysis of model merging (CCF A ACL Main)
    Wang, Z., Xu, X., Liu, Y., Zhang, Y., Lin, P., Feng, S., Wang, D.,Yang, X.,& Schutze, H.
  • SAD: A Large-Scale Strategic Argumentative Dialogue Dataset （CCF A ACL Main)
    Liu, Y., Yu, J., Wang, M., Zhang, Y., Nie, E., Feng, S., Wang, D., Song, K., & Schutze, H.
  • PlaM: Training-Free Plateau-Guided Model Merging for Better Visual Grounding in MLLMs （CCF A ACL Findings)
    Wang, Z., Liu, Y., Wang, M., Nie, E., Chen, D., Zhao, Z., Feng, S., Wang, D., Yang, X., Zhang, Y., & Schutze, H.
  • SAFE-QAQ: End-to-End Slow-Thinking Audio-Text Fraud Detection via Reinforcement Learning （CCF A ACL Main)
    Wang, P., Ma, Z., Dai, X., Liu, Y., Feng, S., Yang, X., Hu, W., Wang, Z., Pan, M., Yuan, L., & Wang, D.
  • MoLAN: A Unified Modality-Aware Noise Dynamic Editing Framework for Multimodal Sentiment Analysis （CCF A ACL Findings)
    Xu, X., Liu, Y., Cai, D., Feng, S., Yang, X., Wang, D., & Zhang, Y.

2025

  • MEKiT: Multi-source Heterogeneous Knowledge Injection Method via Instruction Tuning for Emotion-Cause Pair Extraction （CCF B Cognitive Science) S Mu, Y Liu, S Feng, X Yang, D Wang, Y Zhang
  • Muse: A multimodal conversational recommendation dataset with scenario-grounded user profiles （CCF A ACL Findings) Z Wang, X Yang, Y Liu, S Feng, D Wang, Y Zhang
  • Enhancing llm-based recommendation through semantic-aligned collaborative knowledge
    Z Wang, J Lin, X Yang, Y Liu, S Feng, D Wang, Y Zhang
  • SolEval: Benchmarking Large Language Models for Repository-level Solidity Code Generation （CCF B EMNLP Main) Z Peng, X Yin, R Qian, P Lin, Y Liu, C Ying, Y Luo
  • Pixel-level reasoning segmentation via multi-turn conversations （CCF A ACL)
    D Cai, X Yang, Y Liu, D Wang, S Feng, Y Zhang, S Poria
  • SSMLoRA: Enhancing Low-Rank Adaptation with State Space Model （CCF B NAACL)
    J Yu, Y Zhang, B Wang, P Lin, Y Liu, S Feng

~2024

  • ChatZero: Zero-shot Cross-Lingual Dialogue Generation via Pseudo-Target Language （ECML CCF B）
    Y Liu, F Shi, D Wang, Y Zhang, H Schütze
  • A Unified Data Augmentation Framework for Low-Resource Multi-Domain Dialogue Generation (CCF B ECML-PKDD)
    Y Liu, E Nie, S Feng, Z Hua, Z Ding, D Wang, Y Zhang, H Schütze
  • HiFT: A Hierarchical Full Parameter Fine-Tuning Strategy (** CCF B EMNLP)
    **Y Liu    , Y Zhang, Q Li, S Feng, D Wang, Y Zhang, H Schütze
  • Evaluate What You Can’t Evaluate: Unassessable Quality for Generated Response
    Y Liu, S Feng, D Wang, Y Zhang, H Schütze
  • PVGRU: Generating Diverse and Relevant Dialogue Responses via Pseudo-Variational Mechanism (CCF A ACL Main)
    Y Liu, S Feng, D Wang, H Schütze, Y Zhang
  • DialogConv: A Lightweight Fully Convolutional Network for Multi-view Response Selection (CCF B EMNLP Main) Y Liu, S Feng, W Gao, D Wang, Y Zhang
  • MulZDG: Multilingual Code-Switching Framework for Zero-shot Dialogue Generation (CCF B COLING Main) Y Liu, S Feng, D Wang, Y Zhang
  • A graph reasoning network for multi-turn response selection via customized pre-training (CCF A AAAI Main) Y Liu, S Feng, D Wang, K Song, F Ren, Y Zhang
  • Deep understanding based multi-document machine reading comprehension (TALLIP)
    F Ren, Y Liu, B Li, Z Wang, Y Guo, S Liu, H Wu, J Wang, C Liu, B Wang
  • An understanding-oriented robust machine reading comprehension model (TALLIP)
    F Ren, Y Liu, B Li, S Liu, B Wang, J Wang, C Liu, Q Ma
  • Techkg: A large-scale Chinese technology-oriented knowledge graph
    F Ren, Y Hou, Y Li, L Pan, Y Zhang, X Liang, Y Liu, Y Guo, R Zhao, …
  • A Multiple Utterances based Neural Network Model for Joint Intent Detection and Slot Filling.
    L Pan, Y Zhang, F Ren, Y Hou, Y Li, X Liang, Y Liu
  • An Enhanced ESIM Model for Sentence Pair Matching with Self-Attention.
    Y Liu, X Liang, F Ren, Y Li, Y Hou, Y Zhang, L Pan
  • Neural relation classification with text descriptions (CCF B COLING) F Ren, D Zhou, Z Liu, Y Li, R Zhao, Y Liu, X Liang
  • N-Reader: A Machine Reading Comprehension Model Based on Dual-Layer Self-Attention
    Xiaobo Liang, Feiliang Ren, Yongkang Liu, Lingfeng Pan, Yining Hou, Yi Zhang, Yan Li

Source: