arXiv:2610.28945v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly expected to interact with environments and improve their decisions during deployment. However, how anticipating future outcomes and drawing on accumulated interaction experience can jointly support decision-making remains underexplored. We propose a test-time framework, inspired by Model Predictive Path Integral (MPPI) control, that combines multi-branch predictive analysis with experience-based ad
arXiv:2610.28982v1 Announce Type: new Abstract: While large language models are increasingly deployed as tool-calling agents, their safety evaluations still focus primarily on their ability to refuse harmful requests in text. We investigate whether the refusal behavior in text transfers to tool use. We construct a benchmark of 2,304 prompts across four domains, consisting of matched triplets that present the same harmful or benign request in text-only, standard tool-enabled, and adversarially fr
arXiv:2610.28987v1 Announce Type: new Abstract: Persistent memory is a largely undefended surface for injection and poisoning, and the same risk limits memory sharing among agents. We present PerMIT, a memory architecture with provable integrity guarantees under an explicit trust assumption: only user-typed inputs and values confirmed at a gate are trusted. PerMIT mediates the release and use of untrusted environment-derived values through a finite memory interface, while preserving agent-driven
arXiv:2610.28989v1 Announce Type: new Abstract: Customized agent deployments define policies over tools, resources, and execution conditions. Providing these policies in the task agent's prompt does not reliably prevent violations. Training a guard requires examples of tool calls that comply with or violate those policies. We introduce AgentPolicyForge, a framework that uses executable environments to construct policy benchmarks and iteratively train policy guards. With this framework, we constr
arXiv:2610.29010v1 Announce Type: new Abstract: Radiation treatment planning requires balancing competing objectives to satisfy clinical criteria while accommodating physicians' preferences. This study aims to develop an evidence-driven, multi-agent compound AI system that integrates DRL-based automated treatment planning with LLM-based conversational plan refinement for proton PBS therapy. A policy network trained using PPO first generates multiple plans without human intervention, from which o
arXiv:2610.29014v1 Announce Type: new Abstract: Coding agents are rapidly advancing from single-file generation to long-horizon repository-scale engineering, yet the evaluation of their capabilities to author interactive extended reality (XR) applications remains an open challenge. Unlike traditional software, XR applications operate within dynamic 3D physical environments where runtime robustness depends jointly on continuous user motion, spatial input, and environmental sensing. We introduce G
arXiv:2610.29023v1 Announce Type: new Abstract: Large language model (LLM) agents are often improved by iterating until a fixed verifier is done. Once that verifier saturates, the loop has nothing left to climb. We instead score agents against the current population, and a frozen panel records whether anyone got better. The panel mean can rise inside the seeds by elimination alone. We therefore count a run as training only when its final mean beats the best generation-zero member. A merge is adm
arXiv:2610.29036v1 Announce Type: new Abstract: Explaining a time-series model requires both locating the evidence and extracting meaningful knowledge from it. Yet a single timestep is not a semantic unit: the meaningful patterns live across many samples, not in any one of them. Per-timestep saliency is therefore noisy, high-dimensional, and unstable. Existing methods address this by introducing fixed-size windows and treating them as independent players in a cooperative game. Evaluating a coali
arXiv:2610.29046v1 Announce Type: new Abstract: Large language models (LLMs) have shown considerable promise in synthesizing formal specifications from source code. However, existing approaches primarily focus on functional contracts and loop invariants, while largely neglecting frame conditions, which specify the memory locations that may or may not be modified during execution and constitute an essential part of formal specifications. To address this limitation, we propose an LLM-based approac
arXiv:2610.29057v1 Announce Type: new Abstract: Users increasingly rely on large language models (LLMs) to analyse collections of independent inputs, such as product reviews or social media posts, and return a single aggregated answer. This setting requires multi-instance processing (MIP), where a model combines information from multiple instances into one prediction. We study a task in which the model must infer a binary label for each instance, and identify the most frequent label among the fu
arXiv:2610.29091v1 Announce Type: new Abstract: Objective.To distinguish the contributions of acquisition context and photoplethysmography (PPG) to session-calibrated blood pressure (BP) tracking.Approach. The original comparison used three repeats of five-fold participant-level cross-validation in 653 AuroraBP participants, excluding calibration-labelled prediction targets. Matched analyses separated context and PPG and removed elapsed-time terms. Later exploratory follow-ups retained PPG, acce
arXiv:2610.29092v1 Announce Type: new Abstract: Multilingual large language models are increasingly deployed in settings where visible reasoning traces support verification, education, and safety auditing, yet standard evaluations typically measure only final-answer agreement or holistic response similarity. As a result, they establish outcome consistency without determining whether users across languages receive comparable supporting evidence. We introduce a trace-level divergence framework tha
arXiv:2610.29110v1 Announce Type: new Abstract: Merging different models into a single model is highly desirable for multi-task deployment, particularly in regulated domains like healthcare. In such settings, models are often locally fine-tuned due to strict data privacy policies, necessitating that aggregation operates exclusively on the fine-tuned weights. Existing methods apply a uniform merge across all tasks, ignoring the geometric relationships among their learned subspaces. When task upda
arXiv:2610.29121v1 Announce Type: new Abstract: A coding agent that rewrites its own scaffold has to choose which proposed edit to expand before that edit has been evaluated, and the search then has to return one agent. Each run fits one linear readout on the parent's steering-split solve rate and eight coordinates of the proposed diff, from 96 scaffold edits labeled by four evaluations of a 16-instance fit split, and then freezes the coefficients. HumanEval+ reserves 48 instances as the steerin
arXiv:2610.29122v1 Announce Type: new Abstract: Individual patient data are the gold standard for secondary analysis of time-to-event evidence, but they are rarely released, and reconstruction from a published Kaplan-Meier figure could help. Existing methods all claim high fidelity, each judged by whichever agreements it happens to achieve, and each returns a single dataset although the map from figure to data is not bijective. We define fidelity explicitly: a dataset is admissible when it repro
最近做了个小站，叫 PortGames 域名也简单粗暴，意思就是希望大家能玩到原生的高效的移植游戏，玩得开心 这个站主要用来收集整理社区里的游戏移植项目。可以按游戏和平台找项目，看看需要准备什么，再去原项目那里继续了解 起因是看到了 D2R Mobile 当时第一感觉就是：哇，终于要打破这面墙了嘛，大一统时代终于要来了，终于不用背着我厚重的 NS 出门了哈哈哈。深入了解后发现，它居然是个通用方案，靠 AI 打破了原本最费劲的那部分转译，一下子就井喷了 越看越上头,就做了这个小站 AI 真是个神兵利器，解决了很多以前不太好解决、或者需要花很大精力才能解决的问题。我现在 Codex 额度没用完，总归会觉得有点浪费。人才是 AI 整个工作流上最低效的一环呐哈哈哈 既然 AI 这么好用，相当于给了所有团长一艘船，可以出海找自己的 One Piece。大航海时代，去到哪里、能遇到什么样的伙伴、干成什么样的事情，比最后的宝藏还吸引人。就看大家的想象力和执行力了 看到如今井喷的这么多项目，我也观察了一下：其实更多大佬热衷于移植，而不是热衷于游戏本身。这真是一个很有意思的事情 这种整理项目的网站已经不
Hey HN! We built https://harnesshive.com , a browser based tool to help you design wiring harnesses together. HarnessHive has three key features 1) multiple engineers editing the same harness in real time 2) BOM and drawings that stay in sync automatically and 3) AI-native agent + MCP to assist designing. We hope HarnessHive can make harness design less painful than Excel and Visio for some of you :). Let me know if you have any questions/comments/feature requests!
Earlier this year, inspired by a spate of supply chain attacks (like Mini Shai-Hulud-style worms, which culminated in headlines like Microsoft's Azure tools themselves being compromised), I wrote on my blog about the Kessler Syndrome, applied to software. Ie: the rising likelihood that, because of how bad actors with agentic AI can do far more damage, more quickly with the secrets they exfiltrate from penetrated networks, we could get into a situation where essentially every network of medium or