Related papers: From Theory to Protocol: Executable Frameworks for…
While the phenomenon of grokking, i.e., delayed generalization, has been studied extensively, it remains an open problem whether there is a mathematical framework that characterizes what kind of features will emerge, how and in which…
We propose a theoretical framework for studying behavior cloning of complex expert demonstrations using generative modeling. Our framework invokes low-level controllers - either learned or implicit in position-command control - to stabilize…
Choreographic models support a correctness-by-construction principle in distributed programming. Also, they enable the automatic generation of correct message-based communication patterns from a global specification of the desired system…
Steering vectors offer a training-free mechanism for controlling reasoning behaviors in large language models, but constructing effective vectors requires identifying genuine behavioral signals in the model's hidden states. For behaviors…
Diffusion models revolutionize image generation by leveraging natural language to guide the creation of multimedia content. Despite significant advancements in such generative models, challenges persist in depicting detailed human-object…
As agentic platforms scale, agents are evolving beyond static roles and fixed toolchains, creating a growing need for flexible, decentralized coordination. Today's structured communication protocols (e.g., direct agent-to-agent messaging)…
Autonomous agents executing human instructions must operate reliably even when instructions are incomplete. While recent approaches improve detection of missing information, detection alone is insufficient: agents often proceed to execution…
Ante-hoc interpretability methods based on prototypes provide highly accurate explanations by utilizing the intuitive "this looks like that" reasoning paradigm. On the other hand, post-hoc models can explain predictions for a single image…
Unrestricted adversarial attacks aim to fool computer vision models without being constrained by $\ell_p$-norm bounds to remain imperceptible to humans, for example, by changing an object's color. This allows attackers to circumvent…
Traditional target tracking pipelines including detection, mapping, navigation, and control are comprehensive but introduce high latency, limitting the agility of quadrotors. On the contrary, we follow the design principle of "less is…
Protocol narrations are widely used in security as semi-formal notations to specify conversations between roles. We define a translation from a protocol narration to the sequences of operations to be performed by each role. Unlike previous…
As artificial intelligence (AI) models are scaled up, new capabilities can emerge unintentionally and unpredictably, some of which might be dangerous. In response, dangerous capabilities evaluations have emerged as a new risk assessment…
Population protocols are a distributed computation model in which a collection of anonymous, finite-state agents interact in randomly chosen pairs and update their states according to a fixed transition function. The computation is defined…
With the rapid advancement of game and film production, generating interactive motion from texts has garnered significant attention due to its potential to revolutionize content creation processes. In many practical applications, there is a…
Large language models make it easy for students to delegate writing, analysis, and problem-solving to automated systems, bypassing the effortful engagement that produces lasting understanding. We introduce a practical framework that helps…
Transformer-based encoder-decoder models that generate outputs in a left-to-right fashion have become standard for sequence-to-sequence tasks. In this paper, we propose a framework for decoding that produces sequences from the "outside-in":…
As a pivotal technique for improving the defense of deep models, adversarial robustness transfer via distillation has demonstrated remarkable success in conventional image classification tasks. However, this paradigm encounters critical…
This paper identifies a structural limitation in current multimodal AI architectures that is topological rather than parametric. Contrastive alignment (CLIP), cross-attention fusion (GPT-4V/Gemini), and diffusion-based generation share a…
As large language models (LLMs) become more powerful and are deployed more autonomously, it will be increasingly important to prevent them from causing harmful outcomes. Researchers have investigated a variety of safety techniques for this…
Reasoning protocols such as Chain of Thought (CoT) and Tree of Thought (ToT) organize internal deliberation but lack an explicit mechanism for external questioning that elicits self-revision. We present FOR-Prompting (From Objection to…