· ai
[Paper] Context as a Tool: Context Management for Long-Horizon SWE-Agents
Agents based on large language models have recently shown strong potential on real-world software engineering (SWE) tasks that require long-horizon interaction ...
-
- · ai
[Paper] A Frobenius-Optimal Projection for Enforcing Linear Conservation in Learned Dynamical Models
We consider the problem of restoring linear conservation laws in data-driven linear dynamical models. Given a learned operator widehat{A} and a full-rank constr...
- · ai
[Paper] Scaling Adversarial Training via Data Selection
Projected Gradient Descent (PGD) is a strong and widely used first-order adversarial attack, yet its computational cost scales poorly, as all training samples u...
- · ai
[Paper] Prefill vs. Decode Bottlenecks: SRAM-Frequency Tradeoffs and the Memory-Bandwidth Ceiling
Energy consumption dictates the cost and environmental impact of deploying Large Language Models. This paper investigates the impact of on-chip SRAM size and op...
- · ai
[Paper] StreamAvatar: Streaming Diffusion Models for Real-Time Interactive Human Avatars
Real-time, streaming interactive avatars represent a critical yet challenging goal in digital human research. Although diffusion-based human avatar generation m...
- · ai
[Paper] Toward Secure and Compliant AI: Organizational Standards and Protocols for NLP Model Lifecycle Management
Natural Language Processing (NLP) systems are increasingly used in sensitive domains such as healthcare, finance, and government, where they handle large volume...
- · ai
[Paper] Why Smooth Stability Assumptions Fail for ReLU Learning
Stability analyses of modern learning systems are frequently derived under smoothness assumptions that are violated by ReLU-type nonlinearities. In this note, w...
- · ai
[Paper] MAI-UI Technical Report: Real-World Centric Foundation GUI Agents
The development of GUI agents could revolutionize the next generation of human-computer interaction. Motivated by this vision, we present MAI-UI, a family of fo...
- · ai
[Paper] Backdoor Attacks on Prompt-Driven Video Segmentation Foundation Models
Prompt-driven Video Segmentation Foundation Models (VSFMs) such as SAM2 are increasingly deployed in applications like autonomous driving and digital pathology,...
- · ai
[Paper] From In Silico to In Vitro: Evaluating Molecule Generative Models for Hit Generation
Hit identification is a critical yet resource-intensive step in the drug discovery pipeline, traditionally relying on high-throughput screening of large compoun...
- · ai
[Paper] LibContinual: A Comprehensive Library towards Realistic Continual Learning
A fundamental challenge in Continual Learning (CL) is catastrophic forgetting, where adapting to new tasks degrades the performance on previous ones. While the ...
- · ai
[Paper] Patch-Discontinuity Mining for Generalized Deepfake Detection
The rapid advancement of generative artificial intelligence has enabled the creation of highly realistic fake facial images, posing serious threats to personal ...