Papers by Ruoyu Wang

16 papers
Mitigating Visual Knowledge Forgetting in MLLM Instruction-tuning via Modality-decoupled Gradient Descent (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing fine-tuning and continual learning methods compress visual representations and emphasize task alignment over visual retention.
Approach: They propose a modality-decoupled gradient descent (MDGD) that regulates gradient updates to preserve effective rank of visual features and explicitly disentangles visual learning from task-specific alignment.
Outcome: The proposed model reduces visual forgetting and improves visual retention . it disentangles visual learning from task-specific alignment and preserves effective rank .
FinBPM: A Framework for Portfolio Management-based Financial Investor Behavior Perception Model (2024.eacl-long)

Copied to clipboard

Challenge: a portfolio management framework based on reinforcement learning is needed to optimize stock price movements.
Approach: They propose a framework that takes irrational investment into account when calculating portfolio weights . they use financial text to analyze intrinsic value information of companies and time series data .
Outcome: The proposed framework gains 13.26% returns over state-of-the-art models while controlling for risk.
Second Language (Arabic) Acquisition of LLMs via Progressive Vocabulary Expansion (2025.acl-long)

Copied to clipboard

Challenge: In the evolving landscape of large language models, the predominant focus has been on English and Chinese.
Approach: They propose to utilize Arabic-specific vocabulary in the tokenizer to accelerate decoding.
Outcome: The proposed model achieves decent performance comparable to the best Arabic LLMs across various Arabic benchmarks.
MedDialog: Large-scale Medical Dialogue Datasets (2020.emnlp-main)

Copied to clipboard

Challenge: telemedicine is a medical practice that provides patient care remotely using video conferencing tools.
Approach: They build large-scale medical dialogue datasets to facilitate research . they pretrain several models on the Chinese MedDialog dataset and compare their performance .
Outcome: The proposed datasets show that models trained on MedDialog can generate doctor-like medical dialogues.
Improving Grammatical Error Correction via Pre-Training a Copy-Augmented Architecture with Unlabeled Data (N19-1)

Copied to clipboard

Challenge: Neural machine translation systems have become state-of-the-art approaches for Grammatical Error Correction (GEC) task.
Approach: They propose a copy-augmented architecture for the Grammatical Error Correction task by copying unchanged words from the source sentence to the target sentence.
Outcome: The proposed architecture outperforms all recently published state-of-the-art results by a large margin.
VCORE: Variance-Controlled Optimization-based Reweighting for Chain-of-Thought Supervision (2026.acl-long)

Copied to clipboard

Challenge: Empirical evaluations demonstrate that VCORE achieves the strongest overall average performance, with especially clear gains on lower-capacity models.
Approach: They propose a framework that reformulates supervision as a constrained optimization problem.
Outcome: Empirical evaluations show that VCORE achieves the strongest overall average performance, with especially clear gains on lower-capacity models.
Not All Citations Are Equal:Entropy-Guided Citation Selection for Noise-Resistant Medical LLM (2026.findings-acl)

Copied to clipboard

Challenge: Large language models have demonstrated extensive potential in medical applications . however, their practical deployment in healthcare faces significant challenges .
Approach: They propose a training-free multi-turn reasoning framework and a post-training methodology that provides external knowledge support for large language models.
Outcome: The proposed framework elicits internal thought, external thought, and fusion thought, with an entropy-based reward that encourages selective citation of beneficial external knowledge while penalizing noisy citations.
Constructing Flow Graphs from Procedural Cybersecurity Texts (2021.findings-acl)

Copied to clipboard

Challenge: Graph Convolution Network with BERT sentence embeddings outperforms BERT in all three domains.
Approach: They propose to identify relevant information from procedural texts and generate information flows between sentences from them.
Outcome: The proposed method outperforms BERT in two domains of cybersecurity.
AceGPT, Localizing Large Language Models in Arabic (2024.naacl-long)

Copied to clipboard

Challenge: Significant concerns emerge when addressing cultural sensitivity and local values.
Approach: They propose a localized Large Language Model (LLM) specifically for Arabic, a language imbued with unique cultural characteristics inadequately addressed by current mainstream models.
Outcome: The proposed model sets the state-of-the-art standard for open Arabic LLMs across various benchmarks.
Towards A Better Initial Policy Model For Scalable Long-CoT Reinforcement Learning (2025.findings-acl)

Copied to clipboard

Challenge: Long-CoT reasoning and reinforcement learning are demonstrating remarkable performance and scalability, however, there is a lack of systematic guidelines for obtaining a better initial policy model.
Approach: They propose a systematic guideline and a novel Re-RFT method to obtain more efficient reasoning patterns from different initial models.
Outcome: The proposed method surpasses DeepSeek-R1-Distill-Qwen-14B model by 4.6%, demonstrating its effectiveness and superiority.
Denoising based Sequence-to-Sequence Pre-training for Text Generation (D19-1)

Copied to clipboard

Challenge: PoDA pre-trains encoders and decoders by denoising noise-corrupted text . Unlike encoder-only or decode-only methods, it can be used for text generation tasks without using any task-specific techniques.
Approach: They propose a sequence-to-sequence (seq2sequ) pre-training method PoDA which denoises autoencoders by denoising noise-corrupted text.
Outcome: The proposed method improves model performance over strong baselines without using any task-specific techniques and significantly speed up convergence.
TeachMaster: Generative Teaching via Code (2026.acl-industry)

Copied to clipboard

Challenge: Existing methods for creating video content are limited by high costs and slow update cycles.
Approach: They propose a paradigm shifting educators from manual creators to high-level directors who focus on pedagogical intents while agents handle execution.
Outcome: The proposed framework reduces production costs to 0.3% of traditional course videos and provides a robust solution for scalable education.
QueryAttack: Jailbreaking Aligned Large Language Models Using Structured Non-natural Query Language (2025.findings-acl)

Copied to clipboard

Challenge: Existing methods to bypass security defenses of large language models (LLMs) are not effective, but QueryAttack can be jailbroken.
Approach: They propose a framework to examine generalizability of safety alignment by translating malicious queries into structured non-natural query languages.
Outcome: The proposed framework can achieve high attack success rates and jailbreak various defense methods on mainstream LLMs.
Plutus: Benchmarking Large Language Models in Low-Resource Greek Finance (2025.emnlp-main)

Copied to clipboard

Challenge: Greek is the dominant language of the world's merchant navy and is a key language for international trade.
Approach: They propose to develop a Greek financial evaluation benchmark and a financial LLM fine-tuned on Greek-specific financial data to bridge this gap.
Outcome: The proposed benchmarks surpass GPT-4 by 8.33%, GPT- 4o by 26.83%, and Deepseek-V3 by 67.74%.
Multi-Perspective Context Aggregation for Semi-supervised Cloze-style Reading Comprehension (C18-1)

Copied to clipboard

Challenge: Recent studies have shown that cloze-style reading comprehension is a popular task for measuring the progress of natural language understanding.
Approach: They propose a multi-perspective framework which can be seen as joint training of heterogeneous experts and aggregate context information from different perspectives.
Outcome: The proposed framework achieves new state-of-the-art over previous strong baselines on a recently released cloze-test dataset.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations