Xin Geng

Alignment through Meta-Weighted Online Sampling: Bridging the Gap between Data Generation and Preference Optimization

UNITE: Universal kNowledge Integration from Task-specific Experts

Towards Understanding Feature Learning in Parameter Transfer

DRDFL: Divide-and-conquer Collaboration for Efficient Ring-topology Decentralized Federated Learning

Toward Fine-Grained Domain Knowledge: Curriculum Pseudo-Labeling for Online Test-Time Adaptation

One-for-All Model Initialization with Frequency-Domain Knowledge

ALT: Adaptive Low‑Rank Transformation for Learngene-based Transformer Initialization

InfoAgent: Advancing Autonomous Information‑Seeking Agents

Constructing Variable-depth ViTs from Repeatable Low-bit Learngene

GeneLLM: Inheriting 1.25% of MoE-LLMs to Build Models of 8% Size that Retain 80% Performance

Chain-of-Learngene: A Scalable Learngene-based Paradigm for Building and Initializing Variable-Sized Language Models

Enriching Knowledge Distillation with Intra-Class Contrastive Learning

LADA: Enabling Adaptation of Black-Box LLMs to Dynamic Domain Changes at Test Time

Inheriting Generalizable Knowledge from LLMs to Diverse Vertical Tasks

Rethinking Toxicity Evaluation in Large Language Models: A Multi-Label Perspective

Preference Orchestrator: Prompt-Aware Multi-Objective Alignment for Large Language Models

ECO: Evolving Core Knowledge for Efficient Transfer

RankMatch: A Novel Approach to Semi-Supervised Label Distribution Learning Leveraging Rank Correlation between Labels

Reduction-based Pseudo-label Generation for Instance-dependent Partial Label Learning

Progressively Label Enhancement for Large Language Model Alignment

Selective Label Enhancement Learning for Test-Time Adaptation

Can Class-Priors Help Single-Positive Multi-Label Learning?

FlowPrune: Accelerating Attention Flow Calculation by Pruning Flow Network

KIND: Knowledge Integration and Diversion for Training Decomposable Models

SoPo: Text-to-Motion Generation Using Semi-Online Preference Optimization

Fast Large Language Model Collaborative Decoding via Speculation

Learngene Tells You How to Customize: Task-Aware Parameter Initialization at Flexible Scales

Progressively Label Enhancement for Large Language Model Alignment

Learngene Tells You How to Customize: Task-Aware Parameter Prediction at Flexible Scales

DFL$^2$G: Dynamic Agnostic Federated Learning with Learngene

Improved Noise Schedule for Diffusion Training

RankMatch: A Novel Approach to Semi-Supervised Label Distribution Learning Leveraging Inter-label Correlations

Enriching Knowledge Distillation with Intra-Class Contrastive Learning

Negative-Prompt-driven Alignment for Generative Language Model