Bo Zheng

Reinforced Preference Optimization for Recommendation

Enhancing Generative Auto-bidding with Offline Reward Evaluation and Policy Search

Tricks or Traps? A Deep Dive into RL for LLM Reasoning

Mixture-of-Visual-Thoughts: Exploring Context-Adaptive Reasoning Mode Selection for General Visual Reasoning

Expert Divergence Learning for MoE-based Language Models

ORION: Decoupling and Alignment for Unified Autoregressive Understanding and Generation

COMI: Coarse-to-fine Context Compression via Marginal Information Gain

ReWatch-R1: Boosting Complex Video Reasoning in Large Vision-Language Models through Agentic Data Synthesis

DESIGNER: Design-Logic-Guided Multidisciplinary Data Synthesis for LLM Reasoning

Asymmetric Proximal Policy Optimization: mini-critics boost LLM reasoning

HiDe: Rethinking The Zoom-IN method in High Resolution MLLMs via Hierarchical Decoupling

iTryOn: Mastering Interactive Video Virtual Try-On with Spatial-Semantic Guidance

Mesa and Mask: A Benchmark for Detecting and Classifying Deceptive Behaviors in LLMs

MeSH: Memory-as-State-Highways for Recursive Transformers

FORGE: Forming Semantic Identifiers for Generative Retrieval in Industrial Datasets

QuadSentinel: Sequent Safety for Machine-Checkable Control in Multi-agent Systems

Identity-Preserving Image-to-Video Generation via Reward-Guided Optimization

Beyond Refusals: Fine-grained Safety Alignment for Reasoning LLMs

From Pseudo-Balancing to True Specialization: Memory-Aware Routing for Mixture-of-Experts

RAVR: Reference-Answer-guided Variational Reasoning for Large Language Models

ICPRL: Acquiring Physical Intuition from Interactive Control

Reconstructing KV Caches with Cross-Layer Fusion for Enhanced Transformers

Universal embeddings via Reconstruction-Augmented Contrastive Learning

A Unified Multi-Task Learning Framework for Generative Auto-Bidding with Validation-Aligned Optimization

Beyond Safe Answers: A Benchmark for Evaluating True Risk Awareness in Large Reasoning Models

Enabling Agents to Communicate Entirely in Latent Space

Boosting Training-Free Composed Image Retrieval with Tools

A General Framework for Empowering Graph Neural Networks with Causal Invariance

VideoShield: A Unified Framework for Multimodal Risk Detection and Control in Video Generative Models

USB: A Comprehensive and Unified Safety Evaluation Benchmark for Multimodal Large Language Models

One Sample to Rule Them All: Extreme Data Efficiency in RL Scaling

Think before Recommendation: Autonomous Reasoning-enhanced Recommender

TranSUN: A Preemptive Paradigm to Eradicate Retransformation Bias Intrinsically from Regression Models in Recommender Systems

Towards Visualization-of-Thought Jailbreak Attack against Large Visual Language Models

Differentiable Solver Search for fast diffusion sampling

Differentiable Solver Search for Fast Diffusion Sampling

Minimal Impact ControlNet: Advancing Multi-ControlNet Integration

MTU-Bench: A Multi-granularity Tool-Use Benchmark for Large Language Models

M2rc-Eval: Massively Multilingual Repository-level Code Completion Evaluation

Can VLMs Play Action Role-Playing Games? Take Black Myth Wukong as a Study Case

DMM: Building a Versatile Image Generation Model via Distillation-Based Model Merging

Truly Safe & Truly Helpful: Achieving Harmonious Balance for Large Language Model

Action Sequence Planner: An Alternative For Offline Reinforcement Learning

Adaptive dense reward:Understanding the Gap Between Action and Reward Space in Alignment