Author name cluster

Jinyang Wu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

1 author row

AAAI Conference 2026 Conference Paper

AStar: Boosting Multimodal Reasoning with Automated Structured Thinking

Jinyang Wu
Mingkuan Feng
Guocheng Zhai
Shuai Zhang
Zheng Lian
Fangrui Lv
Pengpeng Shao
Ruihan Jin

Multimodal large language models excel across diverse domains but struggle with complex visual reasoning tasks. To enhance their reasoning capabilities, current approaches typically rely on explicit search or post-training techniques. However, search-based methods suffer from computational inefficiency due to extensive solution space exploration, while post-training methods demand substantial data, computational resources, and often exhibit training instability. To address these challenges, we propose **AStar**, a training-free, **A**utomatic **S**tructured **t**hinking paradigm for multimod**a**l **r**easoning. Specifically, we introduce novel "thought cards", a lightweight library of high-level reasoning patterns abstracted from prior samples. For each test problem, AStar adaptively retrieves the optimal thought cards and seamlessly integrates these external explicit guidelines with the model’s internal implicit reasoning capabilities. Compared to previous methods, AStar eliminates computationally expensive explicit search and avoids additional complex post-training processes, enabling a more efficient reasoning approach. Extensive experiments demonstrate that our framework achieves 53.9% accuracy on MathVerse (surpassing GPT-4o's 50.2%) and 32.7% on MathVision (outperforming GPT-4o's 30.4%). Further analysis reveals the remarkable transferability of our method: thought cards generated from mathematical reasoning can also be applied to other reasoning tasks, even benefiting general visual perception and understanding. AStar serves as a plug-and-play test-time inference method, compatible with other post-training techniques, providing an important complement to existing multimodal reasoning approaches.

PDF Details DOI

AAAI Conference 2026 Conference Paper

From Imitation to Discrimination: Toward a Generalized Curriculum Advantage Mechanism Enhancing Cross-Domain Reasoning Tasks

Changpeng Yang
Jinyang Wu
Yuchen Liu
Shuai Zhang
Yang Li
Qiliang Liang
Hongzhen Wang
Shuai Nie

Reinforcement learning has emerged as a paradigm for post-training large language models, boosting their reasoning capabilities. Such approaches compute an advantage value for each sample, reflecting better or worse performance than expected, thereby yielding both positive and negative signals for training. However, existing approaches often mix them indiscriminately, especially in the early stages, leading to ambiguous guidance and limited gains. To address this issue, we propose CAPO (Curriculum Advantage Policy Optimization), an adaptive curriculum mechanism based on advantage signals. The proposed mechanism bootstraps imitation learning with positive-only advantage samples to establish robust foundations, and subsequently introduces negative signals to cultivate discriminative capabilities, thereby improving generalization across complex scenarios. Compatible with diverse optimization methods including GRPO, PPO, RLOO, and Reinforce++, our method consistently achieves stable and significant improvements in mathematical reasoning tasks, and further generalizes effectively to multimodal Graphical User Interface (GUI) reasoning scenarios, establishing itself as a versatile and robust optimization paradigm.

PDF Details DOI

AAAI Conference 2025 Conference Paper

Code-switching Mediated Sentence-level Semantic Learning

Shuai Zhang
Jiangyan Yi
Zhengqi Wen
Jianhua Tao
Feihu Che
Jinyang Wu
Ruibo Fu

Code-switching is a linguistic phenomenon in which different languages are used interactively during conversation. It poses significant performance challenges to natural language processing (NLP) tasks due to the often monolingual nature of the underlying system. We focus on sentence-level semantic associations between the different code-switching expressions. And we propose an innovative task-free semantic learning method based on the semantic property. Specifically, there are many different ways of languages switching for a sentence with the same meaning. We refine this into a semantic computational method by designing the loss of semantic invariant constraint during the model optimization. In this work, we conduct thorough experiments on speech recognition, speech translation, and language modeling tasks. The experimental results fully demonstrate that the proposed method can widely improve the performance of code-switching related tasks.

PDF Details DOI

JBHI Journal 2023 Journal Article

BertNDA: A Model Based on Graph-Bert and Multi-Scale Information Fusion for ncRNA-Disease Association Prediction

Zhiwei Ning
Jinyang Wu
Yidong Ding
Ying Wang
Qinke Peng
Laiyi Fu

Non-coding RNAs (ncRNAs) are a class of RNA molecules that lack the ability to encode proteins in human cells, but play crucial roles in various biological process. Understanding the interactions between different ncRNAs and their impact on diseases can significantly contribute to diagnosis, prevention, and treatment of diseases. However, predicting tertiary interactions between ncRNAs and diseases based on structural information in multiple scales remains a challenging task. To address this challenge, we propose a method called BertNDA, aiming to predict potential relationships between miRNAs, lncRNAs, and diseases. The framework identifies the local information through connectionless subgraph, which aggregate neighbor nodes’ feature. And global information is extracted by leveraging Laplace transform of graph structures and WL (Weisfeiler-Lehman) absolute role coding. Additionally, an EMLP (Element-wise MLP) structure is designed to fuse pairwise global information. The transformer-encoder is employed as the backbone of our approach, followed by a prediction-layer to output the final correlation score. Extensive experiments demonstrate that BertNDA outperforms state-of-the-art methods in prediction assignment and exhibits significant potential for various biological applications. Moreover, we develop an online prediction platform that incorporates the prediction model, providing users with an intuitive and interactive experience. Overall, our model offers an efficient, accurate, and comprehensive tool for predicting tertiary associations between ncRNAs and diseases.

Details DOI