Zhang Huan
YOU?
Author Swipe
View article: Geometric-Disentangelment Unlearning
Geometric-Disentangelment Unlearning Open
Machine unlearning, the removal of a training subset's influence from a deployed model, is critical for privacy preservation and model reliability, yet gradient ascent on forget samples often harms retained knowledge. Existing approaches f…
View article: Geometric-Disentangelment Unlearning
Geometric-Disentangelment Unlearning Open
Machine unlearning, the removal of a training subset's influence from a deployed model, is critical for privacy preservation and model reliability, yet gradient ascent on forget samples often harms retained knowledge. Existing approaches f…
View article: Geometric-Disentangelment Unlearning
Geometric-Disentangelment Unlearning Open
Machine unlearning, the removal of a training subset's influence from a deployed model, is critical for privacy preservation and model reliability, yet gradient ascent on forget samples often harms retained knowledge. Existing approaches f…
View article: Fast and Certified Bounding of Security-Constrained DCOPF via Interval Bound Propagation
Fast and Certified Bounding of Security-Constrained DCOPF via Interval Bound Propagation Open
Security-Constrained DC Optimal Power Flow (SC DCOPF) is an important tool for transmission system operators, enabling economically efficient and physically secure dispatch decisions. Although CPU-based commercial solvers (e.g., Gurobi) ca…
View article: Fast and Certified Bounding of Security-Constrained DCOPF via Interval Bound Propagation
Fast and Certified Bounding of Security-Constrained DCOPF via Interval Bound Propagation Open
Security-Constrained DC Optimal Power Flow (SC DCOPF) is an important tool for transmission system operators, enabling economically efficient and physically secure dispatch decisions. Although CPU-based commercial solvers (e.g., Gurobi) ca…
View article: The complete mitogenome of <i>ceratina smaragdula</i> fabricius, 1787 (hymenoptera: apidae) and phylogenetic analysis of the family apidae base on mitogenome sequences
The complete mitogenome of <i>ceratina smaragdula</i> fabricius, 1787 (hymenoptera: apidae) and phylogenetic analysis of the family apidae base on mitogenome sequences Open
This study describes the mitogenome of Ceratina smaragdula (Hymenoptera: Apidae), which consists of 15,591 base pairs with an AT content of 85.7%. It contains 37 genes, including 13 protein-coding genes, two rRNA genes, 22 tRNA gene…
View article: Hierarchical Symbolic Pop Music Generation with Graph Neural Networks
Hierarchical Symbolic Pop Music Generation with Graph Neural Networks Open
Music is inherently made up of complex structures, and representing them as graphs helps to capture multiple levels of relationships. While music generation has been explored using various deep generation techniques, research on graph-rela…
View article: Bridging Paintings and Music - Exploring Emotion based Music Generation through Paintings
Bridging Paintings and Music - Exploring Emotion based Music Generation through Paintings Open
Rapid advancements in artificial intelligence have significantly enhanced generative tasks involving music and images, employing both unimodal and multimodal approaches. This research develops a model capable of generating music that reson…
View article: Hierarchical Symbolic Pop Music Generation with Graph Neural Networks
Hierarchical Symbolic Pop Music Generation with Graph Neural Networks Open
Music is inherently made up of complex structures, and representing them as graphs helps to capture multiple levels of relationships. While music generation has been explored using various deep generation techniques, research on graph-rela…
View article: Bridging Paintings and Music - Exploring Emotion based Music Generation through Paintings
Bridging Paintings and Music - Exploring Emotion based Music Generation through Paintings Open
Rapid advancements in artificial intelligence have significantly enhanced generative tasks involving music and images, employing both unimodal and multimodal approaches. This research develops a model capable of generating music that reson…
View article: AlphaOne: Reasoning Models Thinking Slow and Fast at Test Time
AlphaOne: Reasoning Models Thinking Slow and Fast at Test Time Open
This paper presents AlphaOne (α1), a universal framework for modulating reasoning progress in large reasoning models (LRMs) at test time. α1 first introduces α moment, which represents the scaled thinking phase with a universal parameter α…
View article: NEWSROOM DIVERSITY AND ITS CONTRIBUTION TO REDUCED INEQUALITIES (SDG 10)
NEWSROOM DIVERSITY AND ITS CONTRIBUTION TO REDUCED INEQUALITIES (SDG 10) Open
Diversity of newsrooms is also a key aspect of reinforcing the Sustainable Development Goal 10, which aims at lowering the inequalities within and among societies. Media is a key determinant of the formation of the common discourse, democr…
View article: HUMOUR IN SOCIAL CAMPAIGNS AS A TOOL FOR HEALTH AWARENESS & CLIMATE CHANGE (SDG 3 & 13)
HUMOUR IN SOCIAL CAMPAIGNS AS A TOOL FOR HEALTH AWARENESS & CLIMATE CHANGE (SDG 3 & 13) Open
Humor has now been realized to be one of the most powerful and creative tools of social campaigns and can change the serious and intimidating intentions like health awareness and climate change into based and stimulating conversations. Wit…
View article: NEWSROOM DIVERSITY AND ITS CONTRIBUTION TO REDUCED INEQUALITIES (SDG 10)
NEWSROOM DIVERSITY AND ITS CONTRIBUTION TO REDUCED INEQUALITIES (SDG 10) Open
Diversity of newsrooms is also a key aspect of reinforcing the Sustainable Development Goal 10, which aims at lowering the inequalities within and among societies. Media is a key determinant of the formation of the common discourse, democr…
View article: AUDIENCE ENGAGEMENT IN ONLINE JOURNALISM FOR QUALITY EDUCATION (SDG 4)
AUDIENCE ENGAGEMENT IN ONLINE JOURNALISM FOR QUALITY EDUCATION (SDG 4) Open
The online journalism will play the role in realizing extensive, fair and quality education to all the audience as spelled by Sustainable Development Goal 4 (SDG 4). The present digital form of journalism is not only a channel of the infor…
View article: HUMOUR IN SOCIAL CAMPAIGNS AS A TOOL FOR HEALTH AWARENESS & CLIMATE CHANGE (SDG 3 & 13)
HUMOUR IN SOCIAL CAMPAIGNS AS A TOOL FOR HEALTH AWARENESS & CLIMATE CHANGE (SDG 3 & 13) Open
Humor has now been realized to be one of the most powerful and creative tools of social campaigns and can change the serious and intimidating intentions like health awareness and climate change into based and stimulating conversations. Wit…
View article: AUDIENCE ENGAGEMENT IN ONLINE JOURNALISM FOR QUALITY EDUCATION (SDG 4)
AUDIENCE ENGAGEMENT IN ONLINE JOURNALISM FOR QUALITY EDUCATION (SDG 4) Open
The online journalism will play the role in realizing extensive, fair and quality education to all the audience as spelled by Sustainable Development Goal 4 (SDG 4). The present digital form of journalism is not only a channel of the infor…
View article: On The Fragility of Benchmark Contamination Detection in Reasoning Models
On The Fragility of Benchmark Contamination Detection in Reasoning Models Open
Leaderboards for LRMs have turned evaluation into a competition, incentivizing developers to optimize directly on benchmark suites. A shortcut to achieving higher rankings is to incorporate evaluation benchmarks into the training data, the…
View article: Genetic Programming with Model Driven Dimension Repair for Learning Interpretable Appointment Scheduling Rules
Genetic Programming with Model Driven Dimension Repair for Learning Interpretable Appointment Scheduling Rules Open
Appointment scheduling is a great challenge in healthcare operations management. Appointment rules (AR) provide medical practitioners with a simple yet effective tool to determine patient appointment times. Genetic programming (GP) can be …
View article: Improving Data Efficiency for LLM Reinforcement Fine-tuning Through Difficulty-targeted Online Data Selection and Rollout Replay
Improving Data Efficiency for LLM Reinforcement Fine-tuning Through Difficulty-targeted Online Data Selection and Rollout Replay Open
Reinforcement learning (RL) has become an effective approach for fine-tuning large language models (LLMs), particularly to enhance their reasoning capabilities. However, RL fine-tuning remains highly resource-intensive, and existing work h…
View article: Stealthy Backdoor Attack in Self-Supervised Learning Vision Encoders for Large Vision Language Models
Stealthy Backdoor Attack in Self-Supervised Learning Vision Encoders for Large Vision Language Models Open
Self-supervised learning (SSL) vision encoders learn high-quality image representations and thus have become a vital part of developing vision modality of large vision language models (LVLMs). Due to the high cost of training such encoders…
View article: Targeting ligand PDL1 for cardiotoxicity assessment and cardiac protection in immune-related myocarditis
Targeting ligand PDL1 for cardiotoxicity assessment and cardiac protection in immune-related myocarditis Open
The role of cardiac programmed cell death ligand 1 (PDL1) in immune checkpoint inhibitor (ICI) – related myocarditis (irMyocarditis) remains unclear. We aimed to investigate whether ligand PDL1 could serve as an early indicator and a poten…
View article: Targeting ligand PDL1 for cardiotoxicity assessment and cardiac protection in immune-related myocarditis
Targeting ligand PDL1 for cardiotoxicity assessment and cardiac protection in immune-related myocarditis Open
The role of cardiac programmed cell death ligand 1 (PDL1) in immune checkpoint inhibitor (ICI) – related myocarditis (irMyocarditis) remains unclear. We aimed to investigate whether ligand PDL1 could serve as an early indicator and a poten…
View article: Table 1 in A phylogeny with divergence-time estimation of planthoppers (Hemiptera: Fulgoroidea) based on mitochondrial sequences
Table 1 in A phylogeny with divergence-time estimation of planthoppers (Hemiptera: Fulgoroidea) based on mitochondrial sequences Open
Table 1. Estimated divergence and diversification times and 95% credibility intervals. NA: not applicable FamilyEstimated divergence–diversification times95% credibility intervalsDelphacidae173.00 Mya–165.15 Mya (Middle Jurassic)[163.28, 1…
View article: Table 1 in A phylogeny with divergence-time estimation of planthoppers (Hemiptera: Fulgoroidea) based on mitochondrial sequences
Table 1 in A phylogeny with divergence-time estimation of planthoppers (Hemiptera: Fulgoroidea) based on mitochondrial sequences Open
Table 1. Estimated divergence and diversification times and 95% credibility intervals. NA: not applicableFamilyEstimated divergence–diversification times95% credibility intervalsDelphacidae173.00 Mya–165.15 Mya …