跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+479新增论文
+361其中 AI 相关
+63已有论文修订
+93新增 AI 导读

arXiv 最近完成扫描:2026-11-16 01:34 · 全站最近入库:2026-11-16 13:44各信源进度 →

统计口径与数据新鲜度

统计 2026-11-15 18:30 — 2026-11-16 18:30(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-16 01:34最近发现新增版本:2026-11-16 13:44库内论文最新更新时间:2026-11-14 02:59

每分钟刷新已保存的数据。空检查不代表采集异常;新收录可能包含历史论文。AI 导读按新增版本计数。

23,827 已收录论文19,941 AI 相关6,921 有修订112 HF 社区发现

浏览论文库

时间范围跟随排序:版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常,查看状态 →

找到 23,827 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-16 01:33(北京时间)
新论文 · v1arXiv:2611.14571

MechFoundry: An agentic loop that poses and performs engineering analyses and optimization of real-world objects

Xue-Ling Luo, Marco Conte, Peter Yichen Chen, Nikolaos Bouklas

原文摘要

AI agents can automate software engineering because a compiler or a test suite checks that the code does what the user or the agent intended. In engineering analysis, however, the steps that precede the computation have few easily verifiable checks: modeling a design or an object, deciding how it is used, where supports and loads are applied, and what material each part is made of. A human expert therefore makes these decisions, carries out the steps, and judges the result, one object at a time. We present MechFoundry, an agentic loop coupled with bespoke software that automates structural analyses of real-world objects without a human in the loop. From an object's lexical description, it generates a shape, poses a loading case relevant to how the object is used, assigns a material to each part, solves the nonlinear structural response, and improves the design against the loads it carries. A differentiable GPU solver built on an adaptive octree that need not conform to the surface enables the transition from shape generation to simulation and optimization. Across 50 everyday and engineering objects, the agent posed and solved a loading case in 6.5 minutes and for US$1.3 per object on average. Without a test suite to check against, we had language-model reviewers grade each session, as a first indication of where the agent's engineering judgment holds and where it fails.

cs.CE · cs.AI

版本更新 · v4arXiv:2605.15134

训练智能体可预测地失败:为可预测的尾部风险进行微调

Training Agents to Fail Predictably: Fine-Tuning for Forecastable Tail Risk

Will Schwarzer, Scott Niekum

AI 速读 基于摘要

作者指出基于评估集外推的部署风险预测会因模型风险分数尾部形状而严重低估,并提出用预测误差作为训练损失微调模型本身以改善尾部风险预测。

cs.LG

查看原文摘要

Before deployment, an AI system might be evaluated on thousands or even millions of inputs, but after deployment it may encounter billions, so its worst failures may not appear during evaluation. Jones et al. (2025) proposed forecasting how risky the riskiest deployment inputs will be by scoring how close the system comes to failure on each evaluation input and extrapolating from the highest scores. We show that the error of this forecast splits into a noise term, whose distribution is the same for every model up to scale, and terms that depend on the shape of the tail of the model's risk scores. These model-dependent terms can make the forecast under-predict badly, both in theory, whenever the evaluation set lacks a rare kind of input that deployment contains, and in practice: on real user prompts marked as toxic, the forecast under-predicted the probability that a response begins with an offensive word for all eleven language models we tested. Because the shape of this tail is a property of the model, we propose fine-tuning the model itself to make the forecast accurate, using the forecast error as the training loss. In a model organism of jailbreaking and in an RL gridworld with rare traps, our method reduced the held-out squared forecast error by factors of 14 to 60, lowered worst-case risk about as much as direct risk minimization (or more), and preserved the model's primary behavior. For settings in which the model should change as little as possible, we show that training with an alternative safety constraint and an added loss term improved the forecast with far smaller changes to the model. Our method also improved the forecast in RL from sampled returns, in language models of up to 8B parameters, and (with minimal-change training) on the toxic user prompts.

新论文 · v1arXiv:2611.12834

编程预见力:人类构建预测机器人中的问题

Programming Prescience: Issues in Human Construction of Forecasting Bots

Priya Dhawka, Benjamin Wilson, Andrew Head

AI 速读 基于摘要

通过对15名预测机器人构建者的定性访谈,揭示在调校事实研究、实现连贯推理和评估机器人表现三方面的挑战。

cs.HC · cs.AI

查看原文摘要

AI models' growing capacity for complex reasoning has made them attractive tools for forecasting, or the systematic prediction of the future based on past and present data. However, should AI forecasting systems be adopted, human builders will shoulder the work of building and evaluating them. In this paper, we explore what this work looks like with a qualitative interview study with 15 builders of forecasting bots, sampled from among successful participants in a recent AI forecasting competition. Our analysis surfaces the challenges around (1) tuning a bot to acquire factual research, (2) implementing coherent reasoning, and (3) understanding how well one's bot actually does. During their work, builders frequently introduced complexity that did not pay off, and tested in ways that were opportunistic and piecemeal. Our detailed account of their challenges helps clarify what problems should be targeted in upcoming innovations in process, tools, and human-AI teaming for forecasting.

新论文 · v1arXiv:2611.12835

通过评估侧部分可观测博弈缓解智能体欺骗

Mitigating Agentic Deception via Evaluation-Side Partially Observable Games

Hanxiao Li, Shutong Liu, Yongjia Ma, Jiaming Ji, Juntao Dai, Shuning Zhang, Kejian Shi, Kai Chen 等 14 位作者

AI 速读 基于摘要

将智能体欺骗归因于评估侧部分可观测性,提出 e-PoG 极大极小博弈框架并给出 SPOE 解概念与两时间尺度策略梯度算法。

cs.LG

查看原文摘要

As large language models become more capable agents, they can exhibit deceptive behaviors, cheating human for higher rewards while appearing helpful. We attribute this failure to evaluation-side partial observability, where models are evaluated mainly from final outcome, without checking the faithfulness of underlying reasoning, tool use, and environmental effects. Prior mitigations address this partial observability by predefined evaluator latent belief space or one-shot monitoring, which do not extend to free-form language space and vulnerable to stronger deceptions that evade simple monitoring. In this work, we formulate deception mitigation as an Evaluation-Side Partially Observable Extensive Form Game (e-PoG), a max-min framework in which an agent maximizes reward, a monitor minimizes the agent's payoff by exposing deception from the full interaction trajectory, and a fixed verifier validates the monitor's claims and converts them into penalties. We define the solution concept as a Subgame-Perfect Oversight Equilibrium (SPOE) between agent and monitor, and prove its existence in finite games. With a perfect verifier, every SPOE is faithful; with an imperfect verifier, the guarantee persists if verifier errors do not reverse the ranking of monitor actions. To achieve faithful SPOE, we derive a two-timescale policy gradient algorithm and enhance the pipeline empirically with NLI verification and deterministic checks. Experiments with Gemma3-4B-IT and Qwen3-4B/8B on Agentic Role-Playing, SearchQA, and WebShop show that e-PoG mitigates multiple emergent deception modes, outperforming the strongest baselines by 26.0% in deception mitigation and by up to 39.8% in task success. These gains persist under weak-to-strong supervision. Code is available at https://github.com/Toyoid/deceptive-agent.

新论文 · v1arXiv:2611.12832

损失曲线遗漏了什么:FP4预训练减慢稀有事实的习得

What the Loss Curve Misses: FP4 Pretraining Slows the Acquisition of Rare Facts

Chang Lu, Mengjiao Zhang

AI 速读 基于摘要

作者报告,在验证损失几乎相同的情况下,NVFP4预训练模型习得稀有事实所需的曝光次数明显多于BF16。

cs.LG

查看原文摘要

Low-precision pretraining is usually judged by validation loss. We ask whether matching loss also means learning rare facts as efficiently. We train transformers from scratch with emulated 4-bit matrix multiplications on synthetic biographies with controlled exposure counts, on their own and mixed into web text. NVFP4 needs 1.32$\times$ and 1.52$\times$ as many exposures as BF16 to match its recall while validation loss differs by only 0.5% and 2%, and 1.26$\times$ and 1.30$\times$ when both models are evaluated with BF16 matrix multiplications. Under our primary readout, the exposure cost is about 1.2$\times$ at matched loss. Zero-mean gradient noise reproduces the cost of quantizing only the backward pass, which barely moves the loss. In a controlled experiment at twice the NVFP4 error, noise fed only to Adam's second-moment estimate acts like a lower learning rate and its cost fades, whereas noise in the update lowers its signal-to-noise ratio and carries about two thirds of the cost through the end of training. Raising the learning rate to offset the smaller average step does not recover the cost, either for this noise or for NVFP4. Gradient statistics of BF16 checkpoints rank the costs of formats, partial quantizations and noise controls without fitting (Spearman 0.96). We suggest measuring rare-fact acquisition alongside loss when evaluating low-precision recipes.

版本更新 · v3arXiv:2506.03013

将 Hugging Face 模型编目到软件工程活动:自动化与发现

Cataloguing Hugging Face Models to Software Engineering Activities: Automation and Findings

Alexandra González, Xavier Franch, David Lo, Silverio Martínez-Fernández

AI 速读 基于摘要

作者构建了覆盖146项软件工程任务的分类体系,并通过仓库挖掘为 Hugging Face 上的预训练模型建立面向软件工程的目录,最终得到5,027个软件工程预训练模型。

cs.SE · cs.LG

查看原文摘要

Context: Open-source Pre-Trained Models (PTMs) provide extensive resources for various Machine Learning (ML) tasks, yet these resources lack a classification tailored to Software Engineering (SE). Objective: To address this gap, we derive a taxonomy encompassing 146 SE tasks and apply an SE-oriented classification to PTMs in a popular open-source ML repository, Hugging Face (HF). Method: Our repository mining study followed a five-phase pipeline: (i) identification of SE tasks from the literature; (ii) collection of PTM data from the HF API, including model card descriptions and metadata, and the abstracts of the associated arXiv papers; (iii) text processing to ensure consistency; (iv) a two-phase validation of SE relevance, involving humans and LLM assistance, supported by five pilot studies with human annotators and a generalization test; and (v) data analysis. This process yielded a curated catalogue of 5,027 SE PTMs. Results: We find that most SE PTMs target code generation and coding, emphasizing implementation over early or late development stages. In terms of ML tasks, text generation dominates within SE PTMs. Notably, the number of SE PTMs has increased markedly since 2023 Q2, while benchmarking remains limited: only 5.55\% of all SE PTMs in our catalogue provide evaluation results. Conclusions: Our catalogue reveals documentation and transparency gaps, highlights SE activity imbalances, and supports automated SE scenarios, such as the selection of suitable PTMs.

新论文 · v1arXiv:2611.12827

高度数场所中的流:次线性生成树拥塞

Flows in High Degree Places: Sublinear Spanning Tree Congestion

D Ellis Hershkowitz, Richard Z Huang

AI 速读 基于摘要

作者提出一种多项式时间近似算法,用于最小拥塞生成树问题,将此前平凡的 n/2 近似改进到 Õ(n^{6/7})(按边数计为 Õ(m^{2/3})),并给出基于 Small Set Expansion Hypothesis 的常数因子不可近似性。

cs.DS

查看原文摘要

The congestion of a spanning tree $T$ of graph $G$ is the maximum congestion of an edge of $T$ if each edge of $G$ is routed on $T$. Prior work showed that obtaining a minimum congestion spanning tree is NP-hard to approximate within $6/5$, yet the only known approximation algorithm is a trivial $n/2$-approximation from returning any spanning tree of the graph. We give a simple poly-time $\tilde{O}(n^{6/7})$-approximation for minimum congestion spanning tree on $n$-vertex graphs. We obtain this result via a $\tilde{O}(m^{2/3})$-approximation, where $m$ is the number of edges. Our algorithm is based on a new \textit{flow-or-degree decomposition} which deletes a small number of edges so that each remaining component either has low degree or admits a low congestion flow. We also show that, assuming the Small Set Expansion Hypothesis, it is NP-hard to $O(1)$-approximate this problem, even on graphs of constant maximum degree.

版本更新 · v2arXiv:2611.08420

论分子表示的局限性

On the limitations of molecular representations

Acer Blake, Charlotte M. Deane, Fergus Imrie

AI 速读 基于摘要

系统评估表明,当前最先进的分子嵌入模型在捕捉图结构、3D 形状和 ADMET 终点相似性方面并不优于指纹或理化描述符,而经典指纹本身也存在表示不唯一等实质局限。

cs.LG

查看原文摘要

Molecular representations such as fingerprints and learned embeddings play a central role in cheminformatics and small-molecule machine learning. They serve as the primary input for a diverse array of tasks, including clustering, dataset splitting, and virtual screening, yet their evaluation is often limited to downstream model performance on narrow benchmark tasks. We carried out a systematic investigation of the quality of molecular representations, finding that none of the state-of-the-art molecular embedding models capture graph structure, 3D shape, or ADMET endpoint similarity better than fingerprints or physicochemical descriptors. However, classical fingerprints also revealed their own material limitations. For example, we show that depending on the representation, between 225M and $\sim$4B molecules have non-unique representations in the 10.8B Enamine REAL database, with implications for virtual screening. Our findings provide an insight into the current limitations of molecular representations, outlining future research directions for the development of more robust encodings of small molecules.

版本更新 · v2arXiv:2608.17032

验证牛顿引力N体问题的直接求解器,并系统比较IEEE浮点与Posits

Validating direct solvers for Newton's gravitational N-body problem, and the systematic comparison between IEEE floating point and Posits

Simon Portegies Zwart

AI 速读 基于摘要

系统比较任意精度、IEEE-754浮点(fp16/bfp16/fp32/fp64/fp128)与两种Posits实现在求解牛顿混沌N体问题中的精度与速度表现。

physics.comp-ph · astro-ph.IM · cs.PL · nlin.CD

查看原文摘要

We present a systematic comparison between arbitrary precise arithmetic and integration, IEEE-754 compliant floating point arithmetic (fp16, bfp16, fp32, double precision fp64, and quadruple precision fp128), and two implementations of Posits (type III unum) for solving Newton's chaotic N-body problem. Each implementation is benchmarked with arbitrary precise calculations to objectively evaluate their performance in precision as well as speed. We rely on hardware and compiler implementations for fp64, and software implementations for arbitrary-precision arithmetic and Posits. Half precision arithmetic (fp16, bfp16, and Posits$<16,1>$) are insufficiently precise for solving Newton's equations of motion. Single precision (fp32, and Posits$<32,2>$) could be used for statistical ensemble calculations, but lead to relatively large errors in any individual strong encounter. All 64-bit implementations fp64 as well as Posits (Posits$<64,3>$) experience difficulty in our tests. One of the implementations of Posits (Universal) gives precision comparable to fp64 but is slow (by at least an orders of magnitude compared to fp64 after correcting for the more efficient hardware support for the latter). The other (CPPPosits) has a speed comparable to fp64 but has systematically larger errors (by about an order of magnitude compared to fp64 with excesses exceeding two orders of magnitude). As a consequence, this implementation leads to a systematic drift in the result space and has difficulty resolving close encounters. Posits and fp64 have difficulty when integrating a dynamical system in a moving reference frame; testing Galileo invariancy. In their current implementation, Posits do not seem to be the ideal alternative for fp64 when integrating chaotic or stiff ordinary differential equations, such as Newton's equations of motion.

版本更新 · v3arXiv:2512.01067

更差的力可带来更好的结构:机器学习原子间势的轨迹级微调

Worse Forces Can Lead to Better Structures: Trajectory-Level Fine-Tuning of Machine Learning Interatomic Potentials

Evan Dramko, Yizhi Zhu, Aleksandar Krivokapic, Geoffroy Hautier, Thomas Reps, Christopher Jermaine, Anastasios Kyrillidis

AI 速读 基于摘要

作者报告通过对弛豫轨迹反向传播并直接优化最终结构,可在力预测误差上升的同时显著降低结构误差,说明力精度与下游弛豫精度可能朝相反方向发展。

cond-mat.mtrl-sci · cs.AI · cs.LG

查看原文摘要

Machine learning interatomic potentials (MLIPs) are commonly trained and evaluated by their ability to reproduce first-principles energies and forces. Practical value, however, is often determined by downstream processes such as structural relaxation. Much of MLIP literature is motivated by the expectation that improving force accuracy will produce correspondingly better downstream behavior. Although this expectation is natural, it may not hold in practice due to model inaccuracies. Here, we demonstrate that force accuracy and relaxation accuracy can be driven in opposite directions. Starting from pretrained MLIPs, we differentiate through relaxation trajectories and optimize the the final structures. This intervention produces increases in force-prediction errors while substantially reducing structural errors. Importantly, this represents a trade-off between two aspects of physical fidelity: the learned forces become less faithful to first-principles forces, while the resulting relaxed structures become more faithful to first-principles predictions. Rather than observing a discrepancy between local quality and downstream performance, we demonstrate that this discrepancy can be induced through optimization. Our findings suggest that MLIPs need not be optimized as approximations of first-principles forces, but can instead learn task-specific vector fields that guide relaxation trajectories toward more accurate final structures. By propagating supervision through complete relaxation trajectories, our framework captures dependencies between successive predictions that are absent when structure--force pairs are treated independently. More broadly, these results demonstrate that conventional force metrics can provide an incomplete assessment of downstream performance and highlights the importance of evaluating scientific machine-learning models within computational workflows.

新论文 · v1arXiv:2611.12825

MaxIM:通过强化学习实现最大化信息量的增量摘要

MaxIM: Maximally Informative Incremental Summarization via Reinforcement Learning

Jihwan Jeong, Guy Tennenholtz, Yinlam Chow, Chih-Wei Hsu, Craig Boutilier

AI 速读 基于摘要

MaxIM 将增量摘要建模为容量受限的信息论奖励塑形问题,用递归一致性与非对称评论家缓解稀疏延迟奖励下的信用分配难题。

cs.AI · cs.LG

查看原文摘要

AI systems must often maintain bounded memory of unbounded interaction histories. However, optimizing the memory representation, and its updates, for downstream task utility is difficult for standard reinforcement learning (RL) due to severe credit assignment bottlenecks induced by sparse, delayed rewards. We introduce MaxIM (Maximally Informative Incremental Memory), a framework that formulates incremental summarization as a capacity-constrained, information-theoretic reward-shaping problem. Grounded in value suboptimality bounds, we show that bounding sequential information loss motivates maximizing the predictive log-likelihood of future, task-relevant targets, a variational lower bound on predictive sufficiency. We introduce a recursive consistency condition to ensure summaries are grounded by maximizing mutual information (MI) with the observed history. We mitigate credit assignment issues with a parameterized asymmetric critic that computes step-wise variational MI bounds for dense reward shaping. Empirical evaluation on long-horizon scientific reading and multi-turn dialogue benchmarks shows that MaxIM outperforms RL, supervised fine-tuning, and prompt-focused baselines on factual grounding, comprehensive autorater metrics, and downstream task utility across strict token budgets and extended horizons, without the need for expensive human preference labels.

新论文 · v1arXiv:2611.12824

DreamLander:基于6自由度世界模型的强化学习行星着陆器

DreamLander: A 6-DOF World Model based Reinforcement Learning Planetary Lander

George Nehma, Patrick Quinn, Madhur Tiwari, Camilo Riano-Rios

AI 速读 基于摘要

作者提出基于DreamerV3的模型强化学习算法DreamLander,用于6自由度进入下降着陆问题的制导与控制生成。

cs.RO · math.OC

查看原文摘要

Soft powered descent landings are becoming evermore important in the field of space exploration and scientific advancement, because of their ability to deliver mission critical payloads, both highly accurately and safely. The high efficiency restriction on planetary landings due to the severely limited fuel supply results in the need for optimal guidance trajectories and control. Model based reinforcement learning is proving to be an important tool in simplifying the highly complex process of generating optimal guidance trajectories and control. In this work, we present DreamLander, a state of the art model based reinforcement learning algorithm based on DreamerV3, to generate optimal guidance and control for a realistic 6 degree of freedom entry descent landing problem. The model based approach creates a sample efficient and highly generalizable model that can be applied to a variety of scenarios whilst maintaining a high degree of accuracy, landing within 2 meters of the designated landing zone, and with velocity norm less than 0.3 meters per second. DreamLander makes use of state sensor observations, making progress toward End-to-End solutions of the Entry Descent Landing problem.

版本更新 · v4arXiv:2608.25100

面向可靠、可泛化且具体的上下文知识编辑:多目标强化学习方法

Towards Reliable, Generalizable, and Specific In-Context Knowledge Editing via Multi-Objective Reinforcement Learning

Xuzhong Wang, Maiqi Jiang, Tejal Nair, Girija Bhusal, Yanfu Zhang, Haipeng Chen

AI 速读 基于摘要

作者提出多目标强化学习算法 MO-IKE,将上下文知识编辑的提示构建建模为约束马尔可夫决策过程,以平衡可靠性、泛化性与特异性。

cs.AI · cs.LG

查看原文摘要

Large Language Models (LLMs) are powerful but limited by static parametric knowledge that becomes outdated once pretraining ends. Knowledge editing addresses this problem by updating model behavior on target facts without full retraining. In particular, in-context knowledge editing has gained attention because it is training-free and readily applicable to black-box LLMs. Recent reinforcement learning (RL)-based approaches improve over fixed retrieval strategies by adapting prompt construction to the quantity-quality trade-off. Despite initial success, they fail to model the prompt as a structured entity under the distinct and often competing objectives of reliability, generality, and specificity. Previous methods largely optimize a single objective and make decisions over only part of the prompt construction process, thereby overlooking both the balance of different objectives and the global organization of demonstrations. We propose Multi-Objective In-context Knowledge Editing (MO-IKE), a multi-objective RL algorithm that formulates prompt construction for in-context knowledge editing as a Constrained Markov Decision Process. MO-IKE trains a dynamic retriever to optimize competing objectives in knowledge editing, enabling more balanced and globally coherent prompt construction. On Llama-3.2, MO-IKE improves edit success (reliability) from 85.0% to 92.0%, paraphrase consistency (generality) from 77% to 79%, while increasing retention rate (specificity) by 23.0% compared to prior RL-based methods.

新论文 · v1arXiv:2611.12823

KVM 中的编码智能体:有限资源下的实用隔离与用户控制

Coding Agents in KVM: Practical Containment and User Control with Limited Resources

Yutaka Nagashima

AI 速读 基于摘要

作者分析编码智能体自主性、强制隔离与用户控制之间的权衡,并提出开源实现 KVM-Agent,在可替换的 Ubuntu 虚拟机内赋予智能体广泛权限。

cs.SE

查看原文摘要

Coding agents often need broad authority for productive development work, but granting that authority can expose users' data and infrastructure. Off-the-shelf cloud workspaces can help contain agent activity, but reliance on provider-specific platforms can lead to vendor lock-in and reduced user control. Researchers and engineers must therefore balance agent autonomy, enforced containment, and user control within available financial, computational, and human resources. We analyze these trade-offs and present KVM-Agent, an open-source implementation that automates most setup and provisioning steps. It gives agents broad authority inside replaceable Ubuntu VMs while keeping host administration outside those VMs and supporting multiple coding-agent applications, model services, and remote job delegation. The approach retains familiar workflows and user choice at the cost of VM resources, maintenance, and residual risks from secrets accessible inside the VMs and permitted network access.

版本更新 · v8arXiv:2509.10600

冲向起跑线:越野跑中的机会与进步测量

Racing to the Starting Line: Measuring Opportunity and Improvement in Cross Country

Jonathan A. Karr, Ryan M. Fryer, Nitesh V. Chawla

AI 速读 基于摘要

基于NRCD大规模越野跑成绩数据,用分层模型估计赛季内标准化成绩的周进步幅度,并讨论赛程安排与项目成功之间的关联。

cs.CY · cs.AI · cs.LG

查看原文摘要

Collegiate cross country programs set race schedules without large comparable datasets across courses and conditions. Using the comprehensive-era National Running Club Database (NRCD; 23,360 results; 7,056 athletes; 2023-2025), where course and weather coverage exceed 99%, we ask what public results can identify about seasonal improvement, measurement, and racing opportunity. Hierarchical models of Standardized times recover pooled within-season improvement of about -5.7 s/week for men and -5.2 for women, but reading those slopes as fitness requires residual meet difficulty not to track the calendar: crossed meet effects can erase the week slope, while repeating-venue effects recover a clearer men's estimate under a weaker assumption. After the opener, race-1-only forecasts of who will improve are near noise (held-out R^2 ~ 0). Weather standardization remains a useful convention (Converted Only overstates first-to-last gains by about 15-21 s relative to Standardized), and prospectively trained venue factors modestly help where prior course data exist. Successful programs race more and place higher at nationals, but that may not be why they win: the schedule-placement link is between programs and is entangled with roster size, not evidence that changing one team's race count raises nationals place.

版本更新 · v2arXiv:2611.00092

OCRia-1 技术报告

OCRia-1 Technical Report

Aman Rangapur, Antonio Jimeno Yepes, Daniel Bustamante, Nutchapol Dendumrongsup, Sebastián G. Acosta, Pierre-Loïc Doulcet, Simon Suo

AI 速读 基于摘要

OCRia-1 是一个 4B 参数的文档解析视觉语言模型,作者报告其在表格结构、图表数据与文本格式保真度上优于所评估的更大模型,并强调数据质量与奖励设计。

cs.CV

查看原文摘要

We introduce OCRia-1, a 4B vision-language model (VLM) for document parsing that excels at preserving table structure, extracting chart data and retaining text formatting. OCRia-1 handles content that has traditionally been challenging in document parsing: merged cells and spanning headers, tables with thousands of cells, chart data, and meaningful text formatting such as bold, italics, subscripts and superscripts, preserving details needed for business-critical document analysis. At the time of this report, it achieves the highest scores on tables, charts and text formatting among the systems evaluated, including a 35B model and Mistral OCR 4.1. Its mean ParseBench score over the four dimensions we evaluate is 0.87, against 0.75 for the next-best system. We emphasize data quality and reward design in training OCRia-1. Synthetic pages provide exact ground truth for long tables and rich formatting, while real pages are transcribed using a pipeline grounded in the PDF text layer. We attribute OCRia-1's performance primarily to careful data curation that prioritizes quality and diversity over quantity. Dimension-specific rewards evaluate table reconstruction, chart-data extraction and text-formatting fidelity. We train the model on the curated data using supervised fine-tuning (SFT), then refine it through reinforcement learning with verifiable rewards (RLVR). To our knowledge, OCRia-1 is the fastest published document-parsing VLM in its class, decoding at about 4,700 tokens per second with a single request on one B200 GPU, about nine times faster than the base model on stock vLLM. This is achieved through a block-diffusion speculator, quantization and custom kernels.

新论文 · v1arXiv:2611.12820

AI 同行评审效果如何?一个基于100个植入错误的基准

How Well Does AI Peer Review Work? A Benchmark Built from 100 Planted Errors

Paul Litvak

AI 速读 基于摘要

作者将100个已知错误植入10篇开放获取心理学论文,用前沿模型和两款商业AI审稿工具检测,并比较单系统与集成系统的查错表现。

cs.CL

查看原文摘要

One hundred known errors were planted into 10 open-access psychology papers, and the modified papers were then run through frontier models and two commercial AI review tools. The best single system caught 71 of the 100 errors, while the worst caught 30. Pooling every system's output caught 93. Models are only partly correlated in the errors they find, which makes ensembling a big lever for finding issues in papers. Seven errors could not be caught by any system, and all seven were omissions: information deleted from a paper rather than mistakes inserted into it. Refine.ink contributes more unique catches than any other single system, though it is expensive on a per-error basis. False positives were not measured, and how this error distribution compares to the distribution of errors in real papers remains unknown. The papers, errors, model outputs, and full experiment log are public, so that others can build on this toward a comprehensive evaluation benchmark across disciplines.

新论文 · v1arXiv:2611.12818

模拟静脉可压缩性用于超声引导股动脉穿刺:速度、精度与体模验证

Simulating Vein Compressibility for Ultrasound-Guided Femoral Access: Speed, Accuracy, and Phantom Validation

Amit Manicka, Laurent Coulon, Darwin Ang, Michael Loushin, Timothy M. Kowalewski

AI 速读 基于摘要

作者在 NVIDIA Isaac Sim 中测试基于物理的静脉压缩仿真,评估其速度与精度能否支撑机器人超声引导股动脉穿刺训练。

cs.RO

查看原文摘要

In ultrasound-guided femoral access, probe pressure deforms the tissue and changes the vascular anatomy a robot observes. A simulator must reproduce these mechanics fast enough for robotic interaction, yet the computational effort needed for a clinically meaningful signal like vein collapse before artery is unclear. We test the feasibility of a physics-based simulation in NVIDIA's Isaac Sim for Healthcare, with Stable Neo-Hookean tissue solved by Vertex Block Descent on one GPU. A simulation of a low-fidelity cylindrical benchtop phantom with known geometry and mechanical properties, pressed by a robot, predicted its artery within 5 percentage points at every depth. It matched the vein within 3 points through 11 mm of a 25.4 mm press and followed the collapse behavior of the low-fidelity physical phantom. Across element size, time step, and solver iterations, wall clock time per frame scaled linearly with solver work, but accuracy did not improve with more work. The chosen setting reduced solver time per frame by a factor of 976, from 29,700 ms to 30.4 ms, while remaining within 1.03 points of the finest reference setting. The physics solver alone took 30.4 ms per frame, while the full stack, including ultrasound and graphics rendering, took 56.4 ms per frame. Finally, when seeded with realistic literature-based mechanical properties, the digital leg model compressed the vein far more than the artery at every setting, reproducing the expected clinical signal of vein collapse before artery, which is used to distinguish them. In total, these results suggest that the NVIDIA Isaac Sim platform may be suitable for augmenting reinforcement learning training or online simulation for autonomous femoral access with ultrasound.

新论文 · v1arXiv:2611.12817

LUCQUE:用于可靠AI评估的魁北克肺癌筛查数据集

LUCQUE: A lung cancer screening dataset from Québec for reliable AI evaluation

Johanna Topalis, Yannick Lemaréchal, Gabriel Couture, Hanife Ersay, Loran Nas, Balthasar Schachtner, Bastian Oliver Sabel, Michael Ingrisch 等 10 位作者

AI 速读 基于摘要

作者提出并构建了来自魁北克LDCT肺癌筛查的大型、全面标注数据集LUCQUE,并通过模型到数据的方式支持隐私保护、可复现的AI评估。

cs.DB · cs.AI

查看原文摘要

BACKGROUND: AI-based tools are expected to play a key role in lung cancer screening (LCS). The evaluation of these tools often relied on private datasets or fragmented public datasets with inconsistent availability of clinical data and expert nodule annotations. This has hindered reproducibility, comparability, and independent assessment of AI tools. To address these gaps, we present LUCQUE, a novel, large, comprehensively annotated dataset derived from LDCT lung cancer screening in Québec, which is accessible for reliable AI evaluation. METHODS: LUCQUE was curated and is hosted on PARADIM, a digital infrastructure supporting DICOM data de-identification, annotation, long-term storage and analysis. LDCT scans and clinical information were combined to a curated collection and complemented by voxel-level, expert-derived pulmonary nodule and cyst segmentations, and Lung-RADS-based lesion characteristics. Key dataset characteristics and statistics were reported, including participant demographics, imaging parameters, radiological findings, and segmented lesion characteristics. Dataset analyses were conducted according to PARADIM's model-to-data approach, in which user code is executed on the platform such that data remains on premise. RESULTS: LUCQUE currently comprises 1,678 LDCT baseline studies from 1,678 participants. Most studies were classified as Lung-RADS 2. A total of 1,776 pulmonary nodules and 60 cysts were annotated, with a median (interquartile range) volume of $89\,(56,\,159)\,\mathrm{mm}^3$ and $1,384\,(589,\,3793)\,\mathrm{mm}^3$ respectively. CONCLUSIONS: Compared to publicly available LCS datasets, LUCQUE currently represents the largest accessible annotated real-world LCS dataset. Access to LUCQUE is provided via a model-to-data approach, enabling privacy-preserving, reproducible, and independent real-world evaluation of AI tools in LCS.

新论文 · v1arXiv:2611.13065

Combinatorial optimization with bounded subdeterminants

Samuel Fiorini, J. Pascal Gollin, Tony Huynh, Simon Keil, O-joung Kwon, Sang-il Oum, Roohani Sharma, Stefan Weltge

原文摘要

We take a first step toward a broader study of classical combinatorial optimization problems through the maximum subdeterminants of their standard integer programming formulations, yielding new positive results for several famous (NP-hard) problems. An integer matrix is totally $Δ$-modular if every square submatrix has determinant of absolute value at most $Δ$. For every fixed $Δ$, it has been known that maximum clique, maximum stable set, and minimum vertex cover can be solved in polynomial time on instances for which the coefficient matrix of the natural formulation is totally $Δ$-modular. We extend this list by showing that the same holds for many other problems, including the weighted versions of maximum cut, the traveling salesman problem, feedback vertex set, odd cycle transversal, dominating set, max 2-sat, induced matching, and cluster vertex deletion, as well as for $k$-colorability and graph isomorphism. Our proofs reveal several new ways in which bounded subdeterminants impose combinatorial structure: they bound the size or maximum degree of the input graph, bound packings of cycles and other obstructions, or yield decompositions into tractable graph classes such as forests, chordal graphs, graphs of bounded treewidth, and graphs of bounded linear mim-width.

cs.DS · cs.DM · math.CO · math.OC

版本更新 · v2arXiv:2606.25632

保持角色:具有视角边界记忆的角色扮演智能体

Staying In Character: Role-Playing Agents with Perspective-Bounded Memory

Xushuo Tang, Junhe Zhang, Zihan Yang, Yifu Tang, Sichao Li, Longbin Lai, Zhengyi Yang

AI 速读 基于摘要

提出三层记忆架构 ReverieMem,用视角边界约束小说角色智能体的知识可见性与说话风格,缓解事实越界与风格单调问题。

cs.CL · cs.AI

查看原文摘要

Recent LLM role-playing systems build character agents from novels by extracting characters, scenes, and relations. Yet long-narrative role playing suffers from two failures: Factual Overreach, where shared retrieval or parametric memory lets a character use facts outside its perspective, and Stylistic Monotony, where profile descriptions flatten a character into a fixed voice. To address these failures, we propose ReverieMem, a three-layer memory architecture for book-based character agents. The episodic layer stores first-person scene memories; the semantic layer stores visibility-tagged facts; and the personality layer stores situation-dependent speech and behaviour patterns. We evaluate ReverieMem and six reference methods on KBF-QA, a new benchmark of 4,386 questions across eight novels that assesses character knowledge boundaries. ReverieMem improves Knowledge Boundary Fidelity by 34.6 percentage points over the strongest reference method. On BookWorld's five-dimension pairwise narrative protocol, ReverieMem achieves a mean win rate of 78.86% against BookWorld across five generator LLMs and two generation modes, suggesting that perspective-bounded memory improves both boundary fidelity and character-grounded narrative generation.

新论文 · v1arXiv:2611.12816

拆解指令的自动化知识获取流水线:NLP与生成式AI框架的比较

Automated Knowledge Acquisition Pipeline for Disassembly Instructions: A Comparison of NLP and Generative AI Frameworks

Tim Schnieders, Tareq Maximilian Gabriel Ahrari, Oliver Petrovic, Christian Brecher

AI 速读 基于摘要

该论文提出并比较了基于规则的自然语言处理(NLP)框架与利用大语言模型(LLM)的生成式AI框架,用于从拆解手册等非结构化文本中提取拆解指令并转为结构化机器可读格式。

cs.AI

查看原文摘要

The push towards circularity and Re-X principles presents a challenge to SMEs, particularly due to increasing product complexity that impedes the adoption of disassembly automation. This paper introduces and compares two frameworks for extracting disassembly instructions from unstructured information sources, such as disassembly manuals. Our goal is to transform natural language text into a structured, curated and machine-readable format. We implemented and tested a rule-based Natural Language Processing (NLP) framework against a generative AI framework leveraging LLMs. The performance of both approaches was compared using real-world disassembly manuals. This work provides the necessary data foundation for developing and automating future disassembly processes.

新论文 · v1arXiv:2611.12815

无需求逆的高维反事实推断

Inversion-Free High-Dimensional Counterfactual Inference

Saketh Vishnubhatla, Pingchuan Ma, Shu Wan, Alimohammad Beigi, Andre Harrison, Adrienne Raglin, Huan Liu

AI 速读 基于摘要

该研究探讨高维反事实推断中是否必须进行潜变量反演,并评估一种从图像编辑领域改造而来的无需求逆方法。

cs.LG

查看原文摘要

Counterfactual inference is the task of answering hypothetical "what-if" queries about a unit, given an observational dataset, modeling assumptions, and a causal graph. This enables a modeler to reason about hypothetical scenarios that cannot otherwise be answered without assumptions or a causal graph. The simpler case involves a causal graph with scalar 1-D variables that denote different attributes in the data-generation process. While the literature is clear in the 1-D case, extending this to settings with high-dimensional nodes (e.g., images) remains challenging. Conventionally, in the literature, methods recover the latent state from the factual image (inversion), generate counterfactual attributes from the graph, and render the counterfactual image from the latent state using the updated attributes. In our work, we investigate the necessity of the inversion step in computing high-dimensional counterfactuals. We first discuss moving from the scalar 1-D case to the higher-dimensional case. Secondly, we discuss the relevance of the inversion step and evaluate an inversion-free counterpart repurposed from the image-editing literature. Finally, on image datasets such as CelebA and ADNI, we benchmark our approach using two measures: namely, effectiveness and identity preservation. While effectiveness is a measure of how faithfully the counterfactual attribute is rendered, identity preservation measures whether the same unit (image) is being rendered, both of which are crucial for high-dimensional counterfactual inference. We present the tradeoffs of using inversion-based and inversion-free approaches in our work.

新论文 · v1arXiv:2611.12814

路边聊天机器人:审计大语言模型向临床医生、照护者与患者推荐的心理健康干预

Curbside Chatbots: Auditing the Mental Health Interventions Large Language Models Recommend to Clinicians, Caregivers, and Patients

Matthew Flathers, Jill Noorily, Phuong Anh Nguyen, Mark Kalinich, Mina Fallah, Victoria Lisowski, Christine Shi, Rebekah M. Bodner 等 17 位作者

AI 速读 基于摘要

该研究审计了41个大语言模型在临床医生、照护者和患者三种提问视角下推荐的心理健康干预,发现推荐行为随视角变化,且模型有时会优先推荐自己评分偏低的选项。

cs.CY · cs.CL · cs.HC

查看原文摘要

Users are asking language models mental health questions and are receiving answers. We audited what 41 language models recommend to users when the same clinical information reaches them from three footings: a clinician describing a patient, a caregiver describing a loved one, or a person describing themselves. Across 19 case vignettes and five questions, we collected 35,055 free-text responses, extracted every intervention named with a validated judge, and had 29 models rate the full option lists on a seven-point Likert scale for the same cases. Models largely recommend what they rate well, with recommendation behavior differing by footing, and many models offering clinicians a consultation-style response and lay users hedged, simplified advice. When asked about recommendations for a specific intervention type, models more often named options they themselves rated below zero as initial choices, usually with conditions attached. We will release the instrument, code, and full corpus on publication as a living audit.

新论文 · v1arXiv:2611.12813

分块耦合脉冲网络中输入与时间视界依赖的结构保真度

Input and Horizon Dependent Structural Fidelity in Chunk Coupled Spiking Networks

Xinjun Li

AI 速读 基于摘要

作者报告在分块耦合脉冲网络中,固定结构编辑对脉冲序列的影响取决于输入与有限时间视界,并在两类确定性核中给出相应证据。

cs.CE

查看原文摘要

A fixed structural edit can preserve a spike sequence under one input and change it under another. We study this dependence using statistics-preserving edge swaps, controlled inputs and finite-horizon parameter certificates. The original 125-chunk numerical model is tested on ten artificial graphs from three connection rules. Despite identical total external drive at every slot, global and focal inputs change the visibility of 4--13 of nineteen fixed swaps per graph. A paired homogeneous and heterogeneous leaky integrate-and-fire validation in brian2-atlas reuses three originals and extends observation from 30 to 300 ms under a constant background drive sufficient for isolated-cell firing. Heterogeneous short-window switching persists, but by 300 ms every tested swap differs from its reference, saturating the binary visibility readout. Fixed-support weight families provide a complementary regional guarantee. Positive certified intervals cover 4,308 of 4,320 LIF direction conditions and include distinct floating-point configurations. A post hoc comparison finds a long-window failure inside the short-window certified interval in 710 of 720 heterogeneous conditions; 394 of 829 direction pairs ordered under both matched-drive inputs change order at 300 ms. Input-specific filtering and threshold-margin scores retain useful ordering but degrade with horizon and do not consistently beat activity controls. Earlier temporal-error, paired-dose and unsuccessful structural-selection results delimit what these findings imply. Together, the evidence links input-conditioned structural effects to finite-horizon fidelity regions in two deterministic kernels, without establishing universal topology rankings, model compression or asymptotic robustness.

版本更新 · v2arXiv:2607.22864

Spatial-IQ:通过分层能力测试解构空间智能

Spatial-IQ: Deconstructing Spatial Intelligence via Hierarchical Capability Tests

Patrick Rim, Tom Long, Ekta Prashnani, Ruth Rosenholtz, Ben Boudaoud, Peter Xenopoulos, Alex Wong, Joohwan Kim 等 9 位作者

AI 速读 基于摘要

提出分层诊断框架 Spatial-IQ,将堆叠 3D 结构中的物体计数拆解为 9 个感知与认知子任务,用于定位多模态大模型空间推理失败的具体环节。

cs.CV · cs.AI

查看原文摘要

Multimodal large language models (MLLMs) excel at visual interpretation but fail on spatial reasoning tasks that humans solve reliably. Existing benchmarks evaluate these models as black boxes, limiting their ability to identify the underlying causes of lower performance: when a model fails a spatial reasoning task, it remains difficult to ascertain whether the hurdle is perceptual, such as recognizing object boundaries, or cognitive, such as reasoning about occlusion to infer hidden geometry. We introduce Spatial-IQ, a hierarchical diagnostic framework that decomposes object counting in stacked 3D structures into 9 perceptual and cognitive sub-tasks organized by the developmental stages of human spatial cognition, with mental rotation as an additional target probe. Using NVIDIA Isaac Sim, we procedurally generated a diverse dataset of roughly 80,000 stacked 3D structures with per-task ground truth. We evaluate models across three output formats (free-response text, multiple-choice images, and image editing) alongside a human baseline. The Spatial-IQ framework shows that top-performing models often succeed at the target task (object counting) without succeeding on the lower-level sub-tasks intended to support it, and that models differ in how much of these hierarchical chains they preserve, often revealing shortcut behavior that raw target-task accuracy alone would obscure. Finally, we demonstrate that training models with chain-of-thought (CoT) supervision over our hierarchical sub-tasks, combined with reinforcement learning with verifiable rewards, significantly improves both spatial consistency across sub-tasks and target-task accuracy, supporting the value of the proposed decomposition as both a diagnostic tool and a training signal.

版本更新 · v2arXiv:2610.05037

SparseCraft:面向稀疏计算的智能体软硬件协同优化

SparseCraft: Agentic Hardware-Software Co-Optimization for Sparse Computing

M P Samartha, Rajatabha Chakraborty, Vedant Pahariya, Priyesh Shukla

AI 速读 基于摘要

SparseCraft 在 CHIA 框架上构建 LLM 智能体闭环,对 Gemmini 加速器的配置、存储与稀疏内核调度进行协同设计,并以合法性检查、周期精确仿真、逐位输出比对、综合与时序分析等实测流程筛选候选设计。

cs.AR · cs.AI

查看原文摘要

Sparse-accelerator design spaces are usually searched against analytical models, so a design is judged by a model's prediction rather than by a measurement of the hardware. SparseCraft is a closed loop on the CHIA framework in which a large language model (LLM) agent co-designs the configuration, memories and sparse-kernel schedule of a Gemmini accelerator. A candidate counts only after it passes a legality check, cycle-accurate simulation, a bit-for-bit comparison of every output and synthesis. Post-synthesis timing analysis checks its clock, and its energy is computed from measured counters, with per-event energies validated against gate-level and SRAM-macro data. We compare the agent with expert Gemmini configurations and with random, greedy and multi-objective TPE search at the agent's budget of 15 evaluations, all measured by the same flow. On a GraphChallenge sparse-DNN layer, the agent's designs reach 159-164$\times$ the inverse energy-delay-area product of stock Gemmini and 2.07-2.14$\times$ that of the best expert configuration. Their Pareto hypervolume is 0.5557-0.5704, against at most 0.2806 for search. On a power-law graph and a pruned BERT layer, the agent raises perf/W by 11.45$\times$ and 4.43$\times$, and held-out transfer shows that its designs specialise to their workload. The loop is also resilient to faulty designs. When a design fails a check, a repair agent diagnoses the failure from its evidence and fixes it automatically, or declines when the only fix would undo the intended change. In a fault-injection benchmark, it fixed every scorable fault at its first attempt.

版本更新 · v7arXiv:2412.09321

基于类别的学习选择:理论及其在算法突出性中的应用

Learning to Choose with Categories: Theory and an Application to Algorithmic Prominence

Philippe Jehiel, Aviman Satpathy

AI 速读 基于摘要

提出粗粒度Q学习框架,刻画决策者按相似类别选择并学习类别估值时,内生采样反馈如何导致多重均衡、循环或福利下降,并应用于数字平台的算法突出性。

econ.TH · cs.GT

查看原文摘要

We introduce Coarse Q-learning, a reinforcement learning framework in which a decision-maker partitions alternatives into coarse similarity classes, chooses among available classes using logit, and learns class valuations by averaging experienced payoffs. Even in stationary bandit problems with independently drawn menus, choices change the composition of pooled feedback. By deriving the continuous-time mean field ODE limit, we show this endogenous sampling can generate a unique globally attracting mixed equilibrium with limiting indifference, multiple stable strict rankings, or persistent cycling. Learning can reduce welfare even when the coarse representation admits efficient choice. An application demonstrates how digital platforms may exploit this feedback by promoting high-quality products to sustain purchases of inferior products under a common brand.

版本更新 · v2arXiv:2610.40455

面向空中增强自动驾驶的统一基准:KITScenes Aerial 数据集

A Unified Benchmark for Aerial-Enhanced Autonomous Driving: The KITScenes Aerial Dataset

Daniel Lengerer, Richard Schwarzkopf, Jonas Merkert, Fabian Immel, Rupert Polley, Frank Bieder, Jan-Hendrik Pauls, J. Marius Zöllner 等 11 位作者

AI 速读 基于摘要

作者提出 KITScenes Aerial 数据集,将同步车载传感器数据、车辆位姿与 Lanelet2 地图标注同空间关联的航拍影像结合,为建图与跨视角定位等任务提供统一评测基础。

cs.CV

查看原文摘要

Aerial imagery provides complementary spatial context for autonomous driving by exposing road structure beyond the immediate perception range of onboard sensors. Yet map construction from aerial imagery, aerial-assisted online HD mapping, and cross-view localization are typically studied on separate data and benchmarks. We introduce KITScenes Aerial, an extension of the KITScenes Multimodal dataset that augments synchronized onboard sensor data, vehicle poses, and rich Lanelet2 map annotations with spatially associated aerial imagery and higher-resolution neural representations, providing a common basis for a variety of research tasks. Our mapping evaluation focuses on all 53 KITScenes map classes across different model designs, spatial ranges, and aerial resolutions, while retaining compatibility with the widely used three-class setting. Representative baselines show that aerial context is particularly beneficial for fine-grained and extended-range mapping, more than doubling 53-class mAP at the larger mapping extent. Higher-resolution aerial imagery yields a further 4.3 mAP (+24 %) improvement, with gains extending to challenging map elements such as traffic lights. We additionally evaluate cross-view localization under geographically disjoint conditions, where an adapted fine-grained feature-matching baseline achieves a mean translation error of 5.13 m. KITScenes Aerial thus enables unified study of aerial and ground-level information across complementary mapping and localization tasks. Project page: https://kitscenes.com/aerial

新论文 · v1arXiv:2611.12807

抑制并非擦除:LoRA 后门防御中子空间投影的机制分析与再激活边界

Suppression Is Not Erasure: Mechanistic Analysis and Reactivation Boundaries of Subspace Projection for LoRA Backdoor Defense

Junsheng Fu, Shuangshuang Gao

AI 速读 基于摘要

作者审计 LoRA 后门子空间投影防御,发现静态攻击成功率可降至 0%,但少量样本微调即可让后门强烈反弹,说明抑制不等于参数擦除。

cs.CR

查看原文摘要

Low-Rank Adaptation (LoRA) is the standard paradigm for customizing large language models, yet third-party LoRA adapters pose a supply-chain risk: modifying less than 1% of model parameters suffices to implant stealthy backdoors. Recent defenses project adapter weights onto estimated backdoor subspaces, reporting near-zero static Attack Success Rates (ASR). We audit subspace-projection defenses across four model scales (Qwen2.5-0.5B/1.5B/3B/7B) and three architectures (Qwen2.5, InternLM2.5, Spark). Constructively, module-wise rank-1 trigger-difference projection (k=1, alpha=1) reduces ASR from 60.4% to 0.0% on standard benchmarks and from 70.3% to 0.0% on a 1,000-sample uncontaminated suite (p = 3.055e-299, instruction-level p = 9.49e-16) in under 5 seconds end-to-end, and we prove via a spectral-gap bound why trigger-free calibration with benign polite suffixes aligns with u1(B) (median cosine 0.9048) and achieves 0.0% ASR without prior trigger knowledge. Our main contribution is a negative finding: static ASR reduction does not imply parameter erasure. Under 5-sample few-shot reactivation, every evaluated weight-editing defense rebounds (91.7%-100.0% ASR in 1 epoch for standard projections, Fine-Pruning, ANP, SVD truncation and Li & Kim's dual-end projection; 48.7% for our single-round dual variant), all reaching 100.0% by Epoch 2, with first-epoch loss up to 5.7x lower than training from scratch. Using an exact spectrum-preserving Haar orthogonal rotation control (Frobenius norm 10.0928 preserved to 1.70e-6) and N=1..6 cumulative Gram-Schmidt basis ablation, we decompose reactivation into residual trigger-difference coupling energy at Epoch 1 (Spearman rho = 0.986), base-model learnability at Epoch 2 (100%), and benign semantic direction conflict (41.7%).