arXiv subjects

Matthew Hart

Publications and source records attributed to Matthew Hart.

3 recordsLinked to original sources

arXiv2026

A Modular Agentic Framework for Synthetically Constrained Multi-Objective Hit-to-Lead Optimization

Hit-to-lead optimization requires iterative design of hit analogs across competing potency, selectivity, physicochemical, pharmacokinetic, safety, and synthetic constraints. We present SABLE (Synthetically-accessible Agentic Bayesian Ligand Exploration), an open-source framework that employs natural-language orchestration to guide chemical structure optimization. SABLE uses an LLM to interpret user-defined goals and route tasks, while specialized tools perform reaction-templated analog enumeration, physicochemical and ADMET property prediction, structure-based affinity scoring, and Bayesian optimization. The resulting workflow is a computational twin of the analytical and prioritization stages of the design-make-test-analyze cycle, providing provenance of each numerical output. Across single, and multi-objective optimization studies, SABLE enriches candidate sets for user-defined computational objectives while evaluating only a subset of the enumerated search space. Its modular architecture allows tools and characterization backends to be replaced by editing a simple config file, without modifying operational logic. SABLE provides an extensible decision-support framework for prioritizing synthetically constrained analogs in early-stage drug discovery.

cs.AI↗

arXiv2026

OpenJarvis: Personal AI, On Personal Devices

Personal AI stacks, like OpenClaw and Hermes Agent, are becoming central to daily work, yet they route nearly every query (often over sensitive local data) to cloud-hosted frontier models. Replacing frontier models with local models inside existing stacks does not work: swapping Claude Opus 4.6 for Qwen3.5-9B drops accuracy by 25-39 pp across personal AI tasks like PinchBench and GAIA. Existing stacks bundle agentic prompts, tool descriptions, memory configuration, and runtime settings around a specific cloud model. Only the prompts can be tuned, and state-of-the-art prompt optimizers close just 5 pp of the local-cloud gap on their own. This motivates a decomposed personal AI stack: one that exposes individual primitives which can be optimized individually or jointly to close the local-cloud gap. We present OpenJarvis, an architecture that represents a personal AI system as a typed spec over five primitives: Intelligence, Engine, Agents, Tools & Memory, and Learning. Each primitive is an independently editable field, making the stack end-to-end optimizable and measurable against accuracy, cost, and latency. Towards closing the local-cloud gap without surrendering local-model properties, OpenJarvis introduces LLM-guided spec search, a local-cloud collaboration in which frontier cloud models propose edits across the spec at search time, only non-regressing edits are accepted, and the resulting spec runs entirely on-device at inference time. With LLM-guided spec search, on-device specs match or exceed cloud accuracy on 4 of 8 benchmarks and land within 3.2 pp of the best cloud baseline on average. They also reduce marginal API cost by ~800x and end-to-end latency by 4x.

cs.LG↗

arXiv2013

Performance of an LPD prototype detector at MHz frame rates under Synchrotron and FEL radiation

A MHz frame rate X-ray area detector (LPD - Large Pixel Detector) is under development by the Rutherford Appleton Laboratory for the European XFEL. The detector will have 1 million pixels and allows analogue storage of 512 images taken at 4.5 MHz in the detector front end. The LPD detector has 500 mm thick silicon sensor tiles that are bump bonded to a readout ASIC. The ASICs preamplifier provides relatively low noise at high speed which results in a high dynamic range of 10^5 photons over an energy range of 5-20 keV. Small scale prototypes of 32x256 pixels (LPD 2-Tile detector) and 256x256 pixels (LPD supermodule detector) are now available for X-ray tests. The performance of prototypes of the detector is reported for first tests under synchrotron radiation (PETRA III at DESY) and Free-Electron-Laser radiation (LCLS at SLAC). The initial performance of the detector in terms of signal range and noise, radiation hardness and spatial and temporal response are reported. The main result is that the 4.5 MHz sampling detection chain is reliably working, including the analogue on-chip memory concept. The detector is at least radiation hard up to 5 MGy at 12 keV. In addition the multiple gain concept has been demonstrated over a dynamic range to 10^4 at 12 keV with a readout noise equivalent to <1 photon rms in its most sensitive mode.

physics.ins-det↗