An Wang

The paper introduces an A*-inspired framework to generate highly effective and efficient adversarial prompts that cause LLMs to hallucinate commonsense errors while maintaining the original prompt's intent.

The paper introduces Dr. DocBench, a difficulty-aware, comprehensive benchmark designed to rigorously test expert-level and challenging document parsing capabilities for VLMs, demonstrating that current state-of-the-art models fail on complex, domain-specific structures.

The paper introduces an LLM-agent framework to solve the 'last-mile forecasting' problem, bridging the gap between raw statistical predictions and business-ready forecasts by incorporating weakly structured contextual knowledge.

The paper proposes GloResNet, a lightweight 3D CNN that effectively predicts brain injury in preterm infants using T2-weighted MRI, achieving an average accuracy of 75.18%.

The paper introduces InsightVQA, a large-scale benchmark dataset designed for hierarchical visual question answering that assesses complex emotion understanding and cognitive reasoning beyond simple emotion recognition.

COMAP introduces a novel co-evolutionary framework that simultaneously updates textual world models and agent policies through closed-loop interaction, significantly improving long-horizon decision-making for LLM agents.

The paper introduces Repair-Augmented Constraint Learning (RACL), a framework that models contextual decisions by allowing systems to learn whether a candidate should be repaired before being vetoed, significantly reducing false vetoes compared to existing methods.

The paper proposes a novel render-free framework that conditions video diffusion models directly on compressed 3D human mesh tokens, enabling robust 3D-aware human motion control without relying on rendered 2D guidance.

The paper introduces AutoMedBench, a novel workflow-aware benchmark that evaluates autonomous medical-AI agents across a five-stage research process, revealing that agents struggle most with validation and submission.

The paper introduces SMH-Bench, a comprehensive benchmark built on a simulator to rigorously test LLM agents' ability to perform complex, environment-grounded reasoning and actions in realistic smart-home scenarios.

The paper introduces RadioMaster, a novel multi-agent system that successfully translates high-level user intents into physically viable, real-world radio signals, significantly outperforming existing methods.

MOSS-Audio is a unified audio-language model designed for comprehensive understanding of speech, environmental sounds, and music, achieving strong performance across various audio-grounded tasks.

EvoBrain proposes a dynamic, cross-task continual learning framework to overcome the limitations of task-specific EEG decoding, enabling unified and scalable brain-computer interfaces.

The paper proposes a hierarchical framework, PHF (Practice-Habitus-Field), inspired by Bourdieu's Theory of Practice, to improve LLM personalization by modeling user behaviors at three distinct levels.

The paper introduces EvoNote, a self-evolving agentic framework that significantly improves the generation of evidence-grounded health community notes by utilizing an accumulated memory of past misinformation correction experiences.

The paper introduces the threat model of sequential data poisoning, demonstrating that multiple, collaborating attackers can exploit compound vulnerabilities in LLM post-training pipelines that are invisible when analyzing individual stages.

Pepper is a novel, high-bandwidth anonymous broadcast protocol that achieves cryptographic sender anonymity and significantly improves messaging throughput compared to existing state-of-the-art systems.

The paper proposes OneReason, a framework that enhances the reasoning capability of generative recommendation models by focusing on improving item perception and structuring user behavior into coherent latent interests.

This paper introduces a novel full-space quantization-driven architecture (FQA) to create highly efficient and accurate hardware approximations of nonlinear activation functions using piecewise polynomial approximations (PPAs).

This paper proposes a training-free framework called ReasonAlloc to mitigate inference bottlenecks in large language models by recasting decoding-time key-value compression as a hierarchical budget allocation problem.