via Alignment Forum [999] — Writing up a probably-obvious point that I want to refer to later, with significant writing LLM writing help.TL;DR: 1) A common critique of AI safety evaluations is that they occur in unrealistic settings, such as excessive goal conflict, or are…

via ArXiv cs.AI [5] — Multi-modal large language models (MM-LLMs) have shown strong performance in medical image understanding and clinical reasoning. Recent medical agent systems extend them with tool use and multi-agent collaboration, enabling complex decision-making. However,…

via Alignment Forum [999] — A central AI safety concern is that AIs will develop unintended preferences and undermine human control to achieve them. But some unintended preferences are cheap to satisfy, and failing to satisfy them needlessly turns a cooperative situation into an…

via ArXiv cs.AI [5] — Inference-time alignment effectively steers large language models (LLMs) by generating multiple candidates from a reference model and selecting among them with an imperfect reward model. However, current strategies face a fundamental dilemma: ``optimistic''…

via ArXiv cs.AI [3] — The deployment of autonomous AI agents in derivatives markets has widened a practical gap between static model calibration and realized hedging outcomes. We introduce two reinforcement learning frameworks, a novel Replication Learning of Option Pricing…

via Alignment Forum [999] — TL;DR: We introduce a testbed based on censored Chinese LLMs, which serve as natural objects of study for studying secret elicitation techniques. Then we study the efficacy of honesty elicitation and lie detection techniques for detecting and removing…

via DeepMind Blog [4] — Ten years since AlphaGo, we explore how it is catalyzing scientific discovery and paving a path to AGI.

via ArXiv cs.AI [5] — Clinical image interpretation is inherently multi-step and tool-centric: clinicians iteratively combine visual evidence with patient context, quantify findings, and refine their decisions through a sequence of specialized procedures. While LLM-based agents…

via ArXiv cs.AI [3] — Real-time AI services increasingly operate across the device-edge-cloud continuum, where autonomous AI agents generate latency-sensitive workloads, orchestrate multi-stage processing pipelines, and compete for shared resources under policy and governance…

via Alignment Forum [999] — I originally wrote this as a private doc for people working in the field - it's not super polished, or optimized for a broad audience.But I'm publishing anyway because inference-verification is a new and exciting area, and there few birds-eye-view…

via ArXiv cs.AI [2] — Large Language Models (LLMs) are increasingly integrated into critical decision-making pipelines, a trend that raises the demand for robust and automated data analysis. Current approaches to dataset risk analysis are limited to manual auditing methods

via ArXiv cs.AI — Current AI agents can flexibly invoke tools and execute complex tasks, yet their long-term advancement is hindered by the lack of systematic accumulation and transfer of skills. Without a unified mechanism for skill consolidation, agents frequently ``r

via Alignment Forum [5] — Authors: Gerson Kroiz*, Aditya Singh*, Senthooran Rajamanoharan, Neel NandaGerson and Aditya are co-first authors. This work was conducted during MATS 9.0 and was advised by Senthooran Rajamanoharan and Neel Nanda.TL;DRUnderstanding why a model took an action is a key question in AI Safety. It is a

via ArXiv cs.AI [6] — Large language model (LLM) agents typically rely on reactive decision-making paradigms such as ReAct, selecting actions conditioned on growing execution histories. While effective for short tasks, these approaches often lead to redundant tool usage, un

via ArXiv cs.AI [8] — Everyone from AI executives and researchers to doomsayers, politicians, and activists is talking about Artificial General Intelligence (AGI). Yet, they often don't seem to agree on its exact definition. One common definition of AGI is an AI that can do

via ArXiv cs.AI [3] — Synthesizing high-quality training data is crucial for enhancing domain models' reasoning abilities. Existing methods face limitations in long-tail knowledge coverage, effectiveness verification, and interpretability. Knowledge-graph-based approaches s

via Alignment Forum — Also available in markdown at theMultiplicity.ai/blog/schelling-goodness. This post explores a notion I'll call Schelling goodness. Claims of Schelling goodness are not first-order moral verdicts like "X is good" or "X is bad." They are claims about a class of hypothetical coordination games in the

via ArXiv cs.AI [4] — Agile hardware design flows are a critically needed force multiplier to meet the exploding demand for compute. Recently, agentic generative AI systems have demonstrated significant advances in algorithm design, improving code efficiency, and enabling d

via ArXiv cs.AI [3] — Traditional software relies on contracts -- APIs, type systems, assertions -- to specify and enforce correct behavior. AI agents, by contrast, operate on prompts and natural language instructions with no formal behavioral specification. This gap is the

via Alignment Forum [5] — Authors: Aditya Singh*, Gerson Kroiz*, Senthooran Rajamanoharan, Neel NandaAditya and Gerson are co-first authors. This work was conducted during MATS 9.0 and was advised by Senthooran Rajamanoharan and Neel Nanda.MotivationImagine that a frontier lab’s coding agent has been caught putting a bug in