AI Networking Cookbook: Practical recipes for AI-assisted network automation and development
We have 8 of 62 papers
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
Emotional expression can influence the safety decisions of large language models (LLMs), offering a potential avenue for improving safety alignment. Existing studies have mainly focused on how emotional expressions facilitate attacks under harmful requests, while overlooking their effects on benign requests. We find th...
Ensuring the safe and reliable deployment of large language models (LLMs) remains a fundamental challenge. Existing safety alignment approaches either incur high computational cost or unintentionally disrupt the model's core knowledge, leading to degraded fluency and factual accuracy on benign tasks. This reveals a per...
As the capabilities of Vision Language Models (VLMs) continue to improve, they are increasingly targeted by jailbreak attacks. Existing defense methods face two major limitations: (1) they struggle to ensure safety without compromising the model’s utility; and (2) many defense mechanisms significantly reduce the model’...
The multi-view text-guided multimodal fusion adapter (MVFA) is proposed, a parameter-efficient framework that augments frozen LLMs with strong multimodal reasoning capability and achieves state-of-the-art performance on key metrics while updating only a small fraction of parameters.
The results support frozen verification as a training signal for evidence selection, while showing that strict boundary precision remains comparatively weaker.
SafeSteer is a lightweight, inference-time steering framework that effectively defends against diverse jailbreak attacks without modifying model weights, using the innovative use of Singular Value Decomposition to construct a low-dimensional safety subspace during inference.
The proposed MPFDock is a tightly coupled equivariant flow matching method for molecular docking guided by multimodal physical constraints in Cartesian space that consistently outperforms existing methods in terms of docking accuracy and physical realism on the evaluated benchmarks.
PhysMLLMs is a training-stage prior injection architecture that injects physics-inspired spatial continuity priors into Video MLLMs, demonstrating that the injected spatial prior improves video consistency without compromising image-level grounding or general multimodal capability.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.