arXiv:2603.23659v1 Announce Type: new Abstract: When large language models make ethical judgments, do their internal representations distinguish between normative frameworks, or collapse ethics into a single acceptability dimension? We probe hidden representations across five ethical frameworks (deontology, utilitarianism, virtue, justice, commonsense) in six LLMs spanning 4B–72B parameters. Our analysis reveals differentiated ethical subspaces with asymmetric transfer patterns — e.g., deontology probes partially generalize to virtue scenarios while commonsense probes fail catastrophically on justice. Disagreement between deontological and utilitarian […]

Ver mais

arXiv:2604.18732v1 Announce Type: new Abstract: Dynamic state estimation (DSE) is becoming increasingly important for monitoring inverter-dominated power systems. Due to their cascading control structures, inverter-based resources (IBRs) exhibit multi-timescale dynamics, leading to stiff system models that pose significant challenges for conventional DSE methods. In particular, explicit discretization schemes often require impractically small sampling intervals to maintain numerical stability, increasing computational and communication burdens. To address this issue, this paper proposes a stiffness-aware decentralized DSE method for inverter-dominated power […]

Ver mais

arXiv:2601.21027v1 Announce Type: new Abstract: This paper presents a global trajectory optimization framework for minimizing lap time in autonomous racing under uncertain vehicle dynamics. Optimizing the trajectory over the full racing horizon is computationally expensive, and tracking such a trajectory in the real world hardly assures global optimality due to uncertain dynamics. Yet, existing work mostly focuses on dynamics learning at the tracking level, without updating the trajectory itself to account for the learned dynamics. To address these […]

Ver mais

arXiv:2602.03968v1 Announce Type: new Abstract: This paper presents a safety-critical reinforcement learning framework for nonlinear dynamical systems with continuous state and input spaces operating under explicit physical constraints. Hard safety constraints are enforced independently of the reward through action shielding and reachability-based admissible action sets, ensuring that unsafe behaviors are never intentionally selected during learning or execution. To capture nominal operation and recovery behavior within a single control architecture, the state space is partitioned into safe and unsafe […]

Ver mais

Two years ago, building a production-grade voice bot meant hand-rolling WebSocket state machines, managing audio buffers yourself, and debugging race conditions between your speech-to-text and your LLM. Today you can wire it all together in about 100 lines of Python. This tutorial uses Pipecat — Daily.co’s open-source Voice AI framework — to build a real-time voice agent that listens, thinks, and speaks. We’ll use AssemblyAI’s Universal-3 Pro Streaming model for speech-to-text, GPT-4o for the language model, and Cartesia […]

Ver mais

At Serokell, we have been working hard for some time now to help add dependent types to Haskell. It is the primary goal of our dedicated GHC team, currently consisting of Vladislav Zavialov and Andrei Borzenkov. We have occasionally shared our progress on Discourse in the past, though in an irregular manner, which has left some wondering if we are still actively working on it. Yes, we are! However, retrofitting ergonomic dependent types is a very complex task. […]

Ver mais

Standard reward models typically predict scalar scores that fail to capture the multifaceted nature of response quality in non-verifiable domains, such as creative writing or open-ended instruction following. To address this limitation, we propose Rubric-ARM, a framework that jointly optimizes a rubric generator and a judge using reinforcement learning from preference feedback. Unlike existing methods that rely on static rubrics or disjoint training pipelines, our approach treats rubric generation as a latent action learned to maximize judgment accuracy. […]

Ver mais

arXiv:2602.10329v1 Announce Type: new Abstract: Human reasoning is shaped by resource rationality — optimizing performance under constraints. Recently, inference-time scaling has emerged as a powerful paradigm to improve the reasoning performance of Large Language Models by expanding test-time computation. Specifically, instruction-tuned (IT) models explicitly generate long reasoning steps during inference, whereas Large Reasoning Models (LRMs) are trained by reinforcement learning to discover reasoning paths that maximize accuracy. However, it remains unclear whether resource-rationality can emerge from such scaling […]

Ver mais