Deep learning training is non-deterministic: identical code with different random seeds produces models that agree on aggregate metrics but disagree on individual predictions, with per-class AUC swings exceeding 20 percentage points on rare clinical classes. We present a framework for verified bit-identical training that eliminates three sources of randomness: weight initialization (via structured orthogonal basis functions), batch ordering (via golden ratio scheduling), and non-deterministic GPU operations (via architecture selection and custom autograd). The pipeline produces MD5-verified identical trained […]

Ver mais

arXiv:2601.10955v1 Announce Type: new Abstract: The agent-tool communication loop is a critical attack surface in modern Large Language Model (LLM) agents. Existing Denial-of-Service (DoS) attacks, primarily triggered via user prompts or injected retrieval-augmented generation (RAG) context, are ineffective for this new paradigm. They are fundamentally single-turn and often lack a task-oriented approach, making them conspicuous in goal-oriented workflows and unable to exploit the compounding costs of multi-turn agent-tool interactions. We introduce a stealthy, multi-turn economic DoS attack that […]

Ver mais

arXiv:2601.08882v1 Announce Type: new Abstract: Deploying geospatial foundation models on resource-constrained edge devices demands compact architectures that maintain high downstream performance. However, their large parameter counts and the accuracy loss often induced by compression limit practical adoption. In this work, we leverage manifold-constrained optimization framework DLRT to compress large vision transformer-based geospatial foundation models during transfer learning. By enforcing structured low-dimensional parameterizations aligned with downstream objectives, this approach achieves strong compression while preserving task-specific accuracy. We show that […]

Ver mais

It’s refreshingly honest There is no shortage of AI bots on Telegram. Perplexity AI’s bot will answer your questions with sourced references. TelegramGPT wraps ChatGPT in a convenient chat interface. Dozens of others offer variations on the same theme: you type a prompt, you get a response, you move on. The Portal One+ Telegram bot assistant, built on the OpenClaw runtime, aims to be fundamentally different. It doesn’t just answer questions – it lives on your machine, remembers […]

Ver mais

arXiv:2601.03272v1 Announce Type: new Abstract: The versatility of Large Language Models (LLMs) in vertical domains has spurred the development of numerous specialized evaluation benchmarks. However, these benchmarks often suffer from significant semantic redundancy and impose high computational costs during evaluation. Existing compression methods, such as tinyBenchmarks depend critically on correctness labels from multiple historical models evaluated on the full test set, making them inapplicable in cold-start scenarios, such as the introduction of a new task, domain, or model […]

Ver mais

If you have browsed my blog you may have noticed that each article has a different square image. A pattern of cubes that resembles a QR code but is not one. Colors that change from one post to another. And in the center, always the same kanji: 忍. Those images were not manually designed. They were generated by the content itself of each article. How it works When I publish an article the system takes two pieces of […]

Ver mais

arXiv:2601.10828v1 Announce Type: new Abstract: High-order Lie derivatives are essential in nonlinear systems analysis. If done symbolically, their evaluation becomes increasingly expensive as the order increases. We present a compact and efficient numerical approach for computing Lie derivatives of scalar, vector, and covector fields using the MATLAB ADTAYL package. The method exploits a fact noted by R”obenack: that these derivatives coincide, up to factorial scaling, with the Taylor coefficients of expressions built from a Taylor expansion about a […]

Ver mais

arXiv:2601.00148v1 Announce Type: new Abstract: School Campus emergency communication systems are vital for safeguarding student safety during sudden disasters such as typhoons, which frequently cause widespread paralysis of communication infrastructure. Traditional Delay-Tolerant Network (DTN) protocols, such as Direct Delivery and First Contact, struggle to maintain reliable connections in such scenarios due to high latency and low delivery rates. This paper proposes the Activity-based First Contact (A-FC) protocol, an innovative routing scheme that leverages real-world social roles to overcome […]

Ver mais