Wei Wang

E2-Explainer reveals the hidden communication subgraphs that drive successful collaboration in LLM-based multi-agent systems, enabling both interpretability and cost efficiency.

Luna-TTS achieves unprecedented low error rates in real-time speech generation, outperforming leading commercial systems while enabling advanced features like zero-shot voice cloning and emotional control.

Targeted verification can boost reasoning accuracy by over 27 percentage points while using 37% fewer tokens, transforming how we evaluate model outputs.

KANResDiff significantly boosts ambiguous medical image segmentation performance by up to 16.8% through innovative stage-aware modeling techniques.

DARAD not only adapts to new remote sensing data but also preserves the integrity of historical retrieval performance, a dual capability that sets a new standard in continual learning.

FinanceHarness not only automates financial deep research but also reveals that even advanced LLMs struggle with specialized financial tasks, scoring below 40% on rigorous benchmarks.

ServerlessT2I can double the request rates of text-to-image workflows while slashing GPU resource consumption by up to three times.

StrataCL achieves up to 1.9x faster LLM inference and significantly reduces training times by streamlining communication in distributed AI systems.

Existing LLMs fail to effectively retain knowledge over time, with significant implications for their reliability in dynamic environments.

Operating four automated trucks with a single driver can slash freight costs by over 56%, revolutionizing the economics of long-haul trucking.

Safety and coordination in multi-agent robotic systems can be achieved simultaneously, even in complex environments with up to 25 agents.

LANav outperforms traditional Transformer-based navigation policies by 6.3 percentage points, revealing the power of structured state updates in complex environments.

Joint modeling of time series with vastly different scales can be achieved without sacrificing accuracy, thanks to a novel self-Adaptive Scale-handling module.

MaineCoon achieves a groundbreaking 47.5 FPS in real-time audio-visual generation, redefining the potential for social-interactive AI applications.

Tying intermediate reasoning directly to visual evidence allows VLMs to outperform larger models in spatial reasoning tasks.