February 2026

In a matter of months, skills have taken the agent ecosystem by storm: major LLM agent platforms (e.g., Cursor, Claude Code, Antigravity, Coze) now support skills natively, and skill counts across domains are growing at a breakneck pace. Although skills initially served as on-demand prompts to avoid excessive prompt context length, they differ fundamentally from conventional prompts in three respects: (i) skills embody locality: the very presence of a skill implies repeated reuse; (ii) skills encode concrete scenarios, […]

Ver mais

Alibaba has entered the race to build AI that powers robots, not just chatbots. The Chinese tech giant this week unveiled RynnBrain, an open-source model designed to help robots perceive their environment and execute physical tasks.  The move signals China’s accelerating push into physical AI as ageing populations and labour shortages drive demand for machines that can work alongside—or replace—humans. The model positions Alibaba alongside Nvidia, Google DeepMind, and Tesla in the race to build what Nvidia CEO […]

Ver mais

Speech Large Language Models (SLLMs) enable high-level emotion reasoning but often produce ungrounded, text-biased judgments without verifiable acoustic evidence. In contrast, self-supervised speech encoders such as WavLM provide strong acoustic representations yet remain opaque discriminative models with limited interpretability. To bridge this gap, we introduce ADEPT (Agentic Decoding of Emotion via Evidence Probing Tools), a framework that reframes emotion recognition as a multi-turn inquiry process rather than a single-pass prediction. ADEPT transforms an SLLM into an agent that […]

Ver mais

Vertical Federated Learning (VFL) has emerged as a critical paradigm for collaborative model training in privacy-sensitive domains such as finance and healthcare. However, most existing VFL frameworks rely on the idealized assumption of full sample alignment across participants, a premise that rarely holds in real-world scenarios. To bridge this gap, this work introduces Split-MoPE, a novel framework that integrates Split Learning with a specialized Mixture of Predefined Experts (MoPE) architecture. Unlike standard Mixture of Experts (MoE), where routing […]

Ver mais

A practical look at autonomy, system design, and what changes when software starts making decisions. Agentic AI is increasingly ubiquitous in technology discourse. While most discussions focus on models, frameworks, or orchestration patterns, that framing misses the more consequential shift. Agentic AI does not change how systems are built; it changes how decisions are made inside those systems. When software is allowed to plan, act, and adapt autonomously, the hardest problems move upstream. The core challenge is no […]

Ver mais

Most startups don’t think about data platforms when they begin. They think about speed, survival, and shipping something that works. In the early days, this focus is not just reasonable — it is necessary. There is usually one database, one analyst or engineer who “knows the data,” and a handful of SQL queries that answer the most important questions. Reports are built manually, dashboards are updated when someone remembers to do it, and the entire system fits comfortably […]

Ver mais

This paper introduces a biomimetic framework and novel brain-inspired AI (BIAI) models based on spiking neural networks (SNNs) for emotion recognition from audio (speech), visual (face), and integrated multimodal audio-visual data. The developed framework, named eXCube2, uses a three-dimensional SNN that is spatially structured according to a human brain template. The BIAI models developed in eXCube2 are trainable on spatio- and spectro-temporal data using brain-inspired learning rules. Such models are explainable in terms of revealing patterns in data […]

Ver mais

The Deep Image Prior(DIP) suggests that it is possible to train a randomly initialized network with a suitable architecture to solve the inverse imaging problem by simply optimizing its parameters to reconstruct a single degraded image. However, the learning effect it seeks is often achieved with the most naive local convolution, which inevitably leads to the inverse imaging problem being limited by the model’s generative ability. Furthermore, image info is often not related to surrounding pixels but to […]

Ver mais

We investigate additive skip fusion in U-Net architectures for image denoising and denoising-centric multi-task learning (MTL). By replacing concatenative skips with gated additive fusion, the proposed Additive U-Net (AddUNet) constrains shortcut capacity while preserving fixed feature dimensionality across depth. This structural regularization induces controlled encoder-decoder information flow and stabilizes joint optimization. Across single-task denoising and joint denoising-classification settings, AddUNet achieves competitive reconstruction performance with improved training stability. In MTL, learned skip weights exhibit systematic task-aware redistribution: shallow skips […]

Ver mais