Si Liu

RoboInter: A Holistic Intermediate Representation Suite Towards Robotic Manipulation

Factuality Matters: When Image Generation and Editing Meet Structured Visuals

Genie Envisioner: A Unified World Foundation Platform for Robotic Manipulation

OSWorld-MCP: Benchmarking MCP Tool Invocation In Computer-Use Agents

LLaVA-CMoE: Towards Continual Mixture of Experts for Large Vision-Language Models

FLUX-Reason-6M & PRISM-Bench: A Million-Scale Text-to-Image Reasoning Dataset and Comprehensive Benchmark

PICABench: How Far are We from Physical Realistic Image Editing?

Coder-R3: Recognize, Review and Repair Defective Code with Finetuned LLMs in Practice

ExploraQA: Embodied Question Answering with Long-horizon Proactive Exploration

FACT: Mitigating Inconsistent Hallucinations in LLMs via Fact-Driven Alternating Code-Text Training

Towards Realistic Earth-Observation Constellation Scheduling: Benchmark and Methodology

Mixture Compressor for Mixture-of-Experts LLMs Gains More

Point Cluster: A Compact Message Unit for Communication-Efficient Collaborative Perception

LLaVA-MoD: Making LLaVA Tiny via MoE-Knowledge Distillation

Towards Realistic UAV Vision-Language Navigation: Platform, Benchmark, and Methodology