Qiwei Ye

Mr. Qiwei Ye

orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 March 2025 in Nature Chemical Biology
Verified
Journal Article
Closed access
PDF file
YZYaqing Zhang
Yaqing Zhang
YJYuan Jiang
Yuan Jiang
DKDavid Kuster
David Kuster
QYQiwei Ye
Qiwei Ye
WHWenhao Huang
Wenhao Huang
AJAndres Jäschke
Andres Jäschke

25 March 2026 in IEEE Transactions on Computational Social Systems
Verified
Journal Article
Closed access
MLMing Lu
Ming Lu
CLChaozhuo Li
Chaozhuo Li
PWPengbo Wang
Pengbo Wang
CWChenxu Wang
Chenxu Wang
LZLitian Zhang
Litian Zhang
PYPhilip S. Yu
Philip S. Yu
Edgar Allan Poe noted, “Truth often lurks in the shadow of error,” highlighting the deep complexity intrinsic to the interplay between truth and falsehood, notably under conditions of cognitive and informational asymmetry. This dynamic is strikingly evident in large language models (LLMs). Despite their impressive linguistic generation capabilities, LLMs sometimes produce information that appears factually accurate but is, in reality, fabricated, an issue often referred to as “hallucinations.” The prevalence of these hallucinations can mislead users, affecting their judgments and decisions. In sectors such as finance, law, and healthcare, such misinformation risks causing substantial economic losses, legal disputes, and health risks, with wide-ranging consequences. In our research, we have methodically categorized and analyzed the causes, detection methods, and solutions related to LLM hallucinations. Our efforts specifically focus on elucidating the root causes of hallucinations and evaluating current strategies, thereby laying the groundwork for developing robust mitigation techniques.

1 June 2023 in Entropy
Verified
Journal Article
Open access
PDF file
XLXin Lai
Xin Lai
YLYang Liu
Yang Liu
RQRui Qian
Rui Qian
YLYong Lin
Yong Lin
QYQiwei Ye
Qiwei Ye
Graph-structured data, operating as an abstraction of data containing nodes and interactions between nodes, is pervasive in the real world. There are numerous ways dedicated to extract graph structure information explicitly or implicitly, but whether it has been adequately exploited remains an unanswered question. This work goes deeper by heuristically incorporating a geometric descriptor, the discrete Ricci curvature (DRC), in order to uncover more graph structure information. We present a curvature-based topology-aware graph transformer, termed Curvphormer. This work expands the expressiveness by using a more illuminating geometric descriptor to quantify the connections within graphs in modern models and to extract the desired structure information, such as the inherent community structure in graphs with homogeneous information. We conduct extensive experiments on a variety of scaled datasets, including PCQM4M-LSC, ZINC, and MolHIV, and obtain a remarkable performance gain on various graph-level tasks and fine-tuned tasks.

4 March 2025 in Briefings in Bioinformatics
Verified
Journal Article
Open access
JLJiashan Li
Jiashan Li
XCXi Chen
Xi Chen
HHHe Huang
He Huang
MZMingliang Zeng
Mingliang Zeng
JYJingcheng Yu
Jingcheng Yu
QYQiwei Ye
Qiwei Ye
Protein pre-training has emerged as a transformative approach for solving diverse biological tasks. While many contemporary methods focus on sequence-based language models, recent findings highlight that protein sequences alone are insufficient to capture the extensive information inherent in protein structures. Recognizing the crucial role of protein structure in defining function and interactions, we introduce $\mathcal{S}$able, a versatile pre-training model designed to comprehensively understand protein structures. $\mathcal{S}$able incorporates a novel structural encoding mechanism that enhances inter-atomic information exchange and spatial awareness, combined with robust pre-training strategies and lightweight decoders optimized for specific downstream tasks. This approach enables $\mathcal{S}$able to consistently outperform existing methods in tasks such as generation, classification, and regression, demonstrating its superior capability in protein structure representation. The code and models can be accessed via GitHub repository at https://github.com/baaihealth/Sable.

2 May 2025 in bioRxiv
Verified
Preprint
Open access
PDF file
TTThe OpenComplex Team
The OpenComplex Team
QYQiwei Ye
Qiwei Ye
Recent advances in deep learning have revolutionized biomolecular structure prediction, yet a significant challenge remains: biological function emerges from dynamic equilibria across multiple conformational states rather than a single molecular structure. Here, we present OpenComplex2 (OC2), a generative model that efficiently sample biomolecular conformational ensembles. OC2 integrates a unified graph representation capturing diverse molecular inputs with diffusion-based sampling through our FloydNetwork architecture, which operates simultaneously across atomic, residue, and motif hierarchies. This multi-level approach balances fine-grained precision with computational efficiency, enabling conformational landscape sampling in hours rather than the weeks required by traditional simulations. Through comprehensive evaluation, we demonstrate that OC2 accurately reproduces experimentally determined protein and RNA ensembles, generates conformational distributions comparable to millisecond-scale molecular dynamics simulations, and captures functionally relevant transitions in important biological scenarios including allostery, induced-fit mechanisms, and cryptic pocket formation. OC2 also effectively models small molecule conformations with high validity, identifies multiple binding sites, and maintains competitive accuracy on structure prediction benchmarks, while scaling to ultra-large molecular assemblies exceeding 15,000 residues. By bridging static structure prediction and dynamic ensemble characterization, OC2 advances our understanding of how biomolecular dynamics give rise to biological function.

11 February 2026 in arXiv
Verified
Preprint
Open access
PDF file
CWChenxu Wang
Chenxu Wang
CLChaozhuo Li
Chaozhuo Li
SLSongyang Liu
Songyang Liu
ZCZejian Chen
Zejian Chen
JHJinyu Hou
Jinyu Hou
PYPhilip S. Yu
Philip S. Yu
The emergence of multi-agent systems built from large language models (LLMs) offers a promising paradigm for scalable collective intelligence and self-evolution. Ideally, such systems would achieve continuous self-improvement in a fully closed loop while maintaining robust safety alignment--a combination we term the self-evolution trilemma. However, we demonstrate both theoretically and empirically that an agent society satisfying continuous self-evolution, complete isolation, and safety invariance is impossible. Drawing on an information-theoretic framework, we formalize safety as the divergence degree from anthropic value distributions. We theoretically demonstrate that isolated self-evolution induces statistical blind spots, leading to the irreversible degradation of the system's safety alignment. Empirical and qualitative results from an open-ended agent community (Moltbook) and two closed self-evolving systems reveal phenomena that align with our theoretical prediction of inevitable safety erosion. We further propose several solution directions to alleviate the identified safety concern. Our work establishes a fundamental limit on the self-evolving AI societies and shifts the discourse from symptom-driven safety patches to a principled understanding of intrinsic dynamical risks, highlighting the need for external oversight or novel safety-preserving mechanisms.