Papers by Mengyang Li
CODERL+: Improving Code Generation via Reinforcement with Execution Semantics Alignment (2026.acl-long)
Copied to clipboard
Xue Jiang, Yihong Dong, Mengyang Liu, Deng Hongyi, Tian Wang, Yongding Tao, Zhi Jin, Wenpin Jiao, Ge Li
| Challenge: | Large Language Models excel at code generation by learning from vast code corpora, but a fundamental semantic gap remains between training on textual patterns and the goal of functional correctness . reinforcement learning with verifiable rewards (RLVR) approaches are inefficient for establishing a well-aligned connection between the textual representation of code and its execution semantics. |
| Approach: | They propose a novel approach that integrates execution semantics alignment into the RLVR training pipeline for code generation. |
| Outcome: | The proposed model outperforms baseline training and RLVR and shows strong applicability across RL and LLMs. |
MTA:A Merge-then-Adapt Framework for Personalized Large Language Models (2026.acl-long)
Copied to clipboard
Xiaopeng Li, Yuanjin Zheng, Wanyu Wang, Wenlin Zhang, Pengyue Jia, Yingyi Zhang, Haiying He, Mengyang Ma, Yiqi Wang, Maolin Wang, Xuetao Wei, Xiangyu Zhao
| Challenge: | Personalized Large Language Models (PLLMs) aim to align outputs with individual user preferences . current methods of fine-tuning a separate module for each user are unscalable . |
| Approach: | They propose a Merge-then-Adapt framework for Personalized Large Language Models . they construct a shared Meta-LoRA bank and propose an Adaptive LoRA Fusion stage . |
| Outcome: | The proposed framework outperforms existing SOTA methods on the LaMP benchmark. |
Learning Temporally-Aware Sample Weights for Preference Optimization (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for preference optimization rely on static functions of instantaneous model states and ignore temporal learning dynamics. |
| Approach: | They propose a framework that meta-learns adaptive weights using three temporal features: reward margin evolution, learning volatility, and reference deviation. |
| Outcome: | The proposed framework achieves statistically significant improvements over baselines on models ranging from 7B to 70B parameters. |
KG-RAG: Enhancing GUI Agent Decision-Making via Knowledge Graph-Driven Retrieval-Augmented Generation (2025.emnlp-main)
Copied to clipboard
Ziyi Guan, Jason Chun Lok Li, Zhijian Hou, Pingping Zhang, Donglai Xu, Yuzhi Zhao, Mengyang Wu, Jinpeng Chen, Thanh-Toan Nguyen, Pengfei Xian, Wenao Ma, Shengchao Qin, Graziano Chesi, Ngai Wong
| Challenge: | Recent advances in GUI agents have limited app-specific knowledge of complex mobile tasks. |
| Approach: | They propose a Knowledge Graph-driven Retrieval-Augmented Generation framework that transforms fragmented UTGs into structured vector databases for efficient real-time retrieval. |
| Outcome: | The proposed framework outperforms existing methods in a 75.8% success rate and 84.6% decision accuracy test across mobile apps. |
What Do LLMs Learn First? Asymmetric Learning Dynamics of Input Complexity and Output Ambiguity in Preference Alignment (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods treat all preference pairs uniformly during training. |
| Approach: | They propose a training framework that maintains separate, adaptive pacing schedules for each dimension. |
| Outcome: | The proposed training framework outperforms curriculum baselines by 2.1% and 0.21 points . it achieves 42.3% length-controlled win rate on AlpacaEval 2.0 and 7.66 on MT-Bench . |
What Tokens Truly Matter? The Logit Conflation Problem in LLM Sampling (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for large language models filter tokens based on logit magnitudes or derived statistics, under the implicit assumption that high-logit tokens are desirable. |
| Approach: | They propose to isolate the Logit Conflation Problem by using attention-weighted attribution to extract prompt-relevance from token logits. |
| Outcome: | The proposed method improves on LLaMA-3 and is training-free and low latency. |