May 2026

I’m exploring whether “world model” ideas from RL can be adapted to manufacturing-style decision problems — an Industrial World Model for Manufacturing. I put together a small open-source synthetic benchmark around process-window recommendation. The idea is to model a manufacturing process as a state-transition problem under constraints, sparse observations, uncertainty, and next-experiment decisions. The current repo includes a runnable toy environment, simple baseline planners, uncertainty-aware recommendation logic, and an example visualization. It is not a production model and […]

Ver mais

I recently joined Alignerr for a different technical role and noticed they’re looking for people with hands-on MuJoCo / robotics simulation / reinforcement learning experience. The role seems best suited for people who have worked with MuJoCo, MJCF/XML, Gymnasium/dm_control, reward shaping, PPO/SAC/TD3, physics debugging, and robot control. It’s remote contractor work. I don’t want to oversell it because project availability can vary, but the listed rate is high and it may be worth checking out if you already […]

Ver mais

Multimodal Large Language Models (MLLMs) still struggle with fine-grained visual understanding, where answers often depend on small but decisive evidence in the full image. We observe a regional-to-global perception gap: the same MLLM answers fine-grained questions more accurately when conditioned on evidence-centered crops than on the corresponding full images, suggesting that many failures stem from difficulty to focus on relevant evidence rather than insufficient local recognition ability. Motivated by this observation, we propose Vision-OPD (Vision On-Policy Distillation), a […]

Ver mais

Post-training has split large language model (LLM) alignment into two largely disconnected tracks. Online reinforcement learning (RL) with verifiable rewards drives emergent reasoning on math and code but depends on a programmatic verifier that cannot reach open-ended tasks, while preference optimization handles open-ended generation yet forgoes the continuous exploration that powers online RL. Closing this gap requires a verifier for open-ended quality, but a scalar reward model is the wrong shape for the job. Quality is multi-dimensional, and […]

Ver mais

Unmanned Aerial Vehicles in dynamic environments face telemetry outages, structural vibrations, and regime-dependent noise that invalidate the stationary covariance assumptions of classical Kalman filters. The Sage-Husa Kalman Filter (SHKF) estimates noise statistics online, but its reliance on a static, scalar forgetting factor forces a strict compromise between steady-state stability and transient responsiveness. We introduce the N-Deep Recurrent Sage-Husa Filter (NDR-SHKF), which replaces this scalar parameter with a vector-valued memory attenuation policy learned by a hierarchical recurrent network operating […]

Ver mais

Blood-based biomarkers underpin clinical diagnosis and management, yet their interpretation relies largely on fixed population reference intervals that ignore stable, intra-patient variability. As such, population-based interpretation can mask meaningful deviation from an individual’s baseline, risking delayed disease detection. To remedy this, there have been increasing efforts to personalize blood biomarker interpretation using individual testing histories. However, these methods may overfit to sparse data, inflating false-positive rates and unnecessary follow-up, and can also unwittingly include unrecognized or subclinical disease. […]

Ver mais

Virtually all aspects of many-body atomic physics are challenging: experiments are technically demanding, datasets have become enormous, and the memory and CPU requirements for classical simulation of generic quantum systems often scale exponentially with system size. Machine learning (ML) methods are already assisting in each of these areas and are poised to become transformative. Here, we focus on two specific applications of ML to cold-atom-based quantum simulators. These devices generally generate data in the form of images; we […]

Ver mais