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

+406新增论文
+306其中 AI 相关
+55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 11:42各信源进度 →

统计口径与数据新鲜度

统计 2026-11-15 12:59 — 2026-11-16 12:59(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 11:42库内论文最新更新时间:2026-11-14 01:29

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。

23,754 已收录论文19,886 AI 相关6,892 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 23,754 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)
版本更新 · v2arXiv:2606.12635

CD-RCM:面向反射共聚焦显微镜的可泛化连续深度新视角合成

CD-RCM: Generalizable Continuous-Depth Novel View Synthesis for Reflectance Confocal Microscopy

Tooba Imtiaz, Milind Rajadhyaksha, Kivanc Kose, Jennifer Dy

AI 速读 基于摘要

提出首个针对反射共聚焦显微镜(RCM)的前馈式新视角合成模型 CD-RCM,从稀疏深度采样的 RCM 层叠图像预测中间深度切面,实现连续深度、各向同性的三维可视化。

cs.CV

查看原文摘要

Reflectance confocal microscopy (RCM) provides noninvasive, cellular-resolution "optical biopsies" of human skin in vivo by acquiring en-face images at successive depths, forming a sparse z-stack. Due to optical limitations, these stacks are anisotropic 3D volumes with lateral resolution (0.5 $μ$m) $\sim$6 times higher compared to axial resolution, which is defined by the optical sectioning (3 $μ$m), limiting the interpretation of tissue. Our goal is to provide continuous-depth visualization by interpolating intermediate sections and making the 3D volume isotropic. Such a representation permits arbitrary-direction sectioning, including histopathology-like cross-sectional examination, without requiring per-patient optimization. To that end, we introduce the first RCM-specific novel-view synthesis (NVS) approach, CD-RCM, a feedforward model that predicts realistic, unseen depths from sparsely sampled RCM stacks. Classical neural rendering methods focus on reconstruction from surface-level multi-view observations. In contrast to surface-level camera views, RCM can acquire optically sectioned en-face images of tissue beyond the surface up to 200 $μ$m. However, during visualization of the RCM stacks, observations of the shallower sections (towards the surface) obscure the deeper ones. This unique axial imaging geometry and layer-dependent anatomical organization motivated our development of a tailored architectural and training framework that explicitly accounts for RCM's depth-resolved, occlusive imaging physics. Experiments demonstrate that CD-RCM achieves high-fidelity novel-view synthesis with sub-second inference time.

新论文 · v1arXiv:2611.14401

发现南亚季风降水热力学的方程

Discovering Equations for the Thermodynamics of South Asian Monsoon Precipitation

Savannah L. Ferretti, Jane W. Baldwin, Michael S. Pritchard, Tom Beucler

AI 速读 基于摘要

该研究用符号蒸馏从21年再分析资料中学习可解释的南亚季风降水方程,其解释方差约为常用浮力诊断的1.5倍,并保留约85%的神经网络技巧。

cs.LG · physics.ao-ph

查看原文摘要

South Asian monsoon precipitation depends on the thermodynamic structure of the lower troposphere, but how its ingredients combine to shape precipitation remains unclear, partly because this dependence is nonlocal in the vertical. We use symbolic distillation, which combines neural networks and symbolic regression, to discover interpretable precipitation equations from 21 years of June-August reanalysis data. Neural networks first learn the vertical thermodynamic structure most relevant to precipitation, condensing each variable into a single interpretable feature not tied to a particular vertical grid. Then, symbolic regression derives closed-form equations from these features and the surface state. Our best equation explains 1.5 times as much precipitation variance as a widely used buoyancy-based diagnostic and recovers about 85% of neural network skill while using about a thousand times fewer parameters. Unlike the buoyancy diagnostic, which adds moisture and instability, the equations treat them as competing controls, with the heaviest precipitation occurring when the lower free troposphere is humid. Over ocean, sensible heating is associated with additional precipitation mainly when near-surface air is warm and moist, improving predicted precipitation and its diurnal cycle. Although the learned features were optimized only for precipitation prediction, the equations largely respond to each feature as expected from convective physics, consistent with the reanalysis. These responses reduce the dependence of monsoon precipitation on the entire vertical column to a few simple, testable constraints that could guide data-driven precipitation models to make them more physically consistent.

版本更新 · v2arXiv:2606.30533

在细粒度负载波动下协调严格SLO与低成本

Reconciling Strict SLOs with Low Cost under Fine-Grained Load Fluctuations

Dilina Dehigama, Shyam Jesalpura, Zeyu Xu, Marton Nemeth, Shengda Zhu, Marios Kogias, Boris Grot

AI 速读 基于摘要

Spandana通过将SLO保障与成本优化解耦,在应用VM与FaaS之间逐请求调度,以在细粒度负载波动下兼顾严格SLO、高利用率与低成本。

cs.DC

查看原文摘要

Cloud-based online services face significant sub-second load fluctuations while needing to meet strict Service Level Objectives (SLOs). Cluster operators often over-provision resources to protect SLOs, sacrificing utilization and cost efficiency. Existing reactive and proactive autoscalers, serverless (FaaS) deployments, and VM/FaaS hybrid systems fail to reconcile strict SLO compliance with low cost and high utilization under fine-grained load fluctuation. We introduce Spandana, an architecture that addresses this trade off by decoupling SLO enforcement from cost optimization. A lightweight controller colocated with each application VM enforces SLOs by steering each arriving request between the VM and FaaS. Requests that can meet the SLO stay on the VM; the remaining requests are forwarded to a stock FaaS layer such as AWS Lambda. For cost optimization, Spandana's resource allocator determines the most-efficient VM provisioning by accounting for VM cost, FaaS cost, and traffic volatility, allowing the VM pool to run at high utilization. Our evaluation shows that Spandana maintains strict SLO adherence, achieves 76-86% CPU utilization, and reduces cost by 5-44% over three SOTA baselines.

新论文 · v1arXiv:2611.14399

REAL Loss:面向投机解码的真实期望接受长度优化

REAL Loss: Optimizing Real Expected Acceptance Length for Speculative Decoding

Yu Tian

AI 速读 基于摘要

作者提出 REAL 损失,用路径权重无偏估计并优化投机解码的真实期望接受长度(EAL),在受控损失级对比中优于若干已有目标。

cs.LG · cs.CL

查看原文摘要

Speculative decoding accelerates LLM inference by having a draft model propose multiple tokens for the target model to verify in one pass. The speedup is governed by the expected acceptance length (EAL), yet standard training objectives such as cross-entropy, KL divergence, and total variation focus only on distribution matching or single-step overlap. Recent studies instead propose block-level training objectives, aiming to optimize EAL directly. However, these objectives remain surrogates rather than exact formulations of EAL. For example, some average later acceptance over target-distributed continuations rather than the verification branches selected by accepted tokens, introducing a covariance gap that can reverse the gradient direction. We analyze this gap and derive REAL, a real EAL training loss that uses path weights for unbiased estimation and Rao-Blackwellization for variance reduction. In a controlled loss-level comparison with a Qwen3.6-35B-A3B target and a Domino drafter, REAL outperforms WTV, VSD, DBLast, and the static LK baseline on probabilistic serving acceptance length across 9 task subsets and 5 evaluation seeds. These gains require only a change to the training loss, with no modification to the drafter architecture or inference procedure, demonstrating the practicality of real EAL training. A reference implementation of the REAL loss is available at https://github.com/King-ty/speculators/tree/real-loss.

新论文 · v1arXiv:2611.14604

KernelMosaic: An Open Dataset and Agentic Framework for CUDA Kernel Generation

Ningyuan Xi, Yuhang Lai, Ning Miao

原文摘要

Training large language models to generate efficient CUDA kernels requires supervision spanning diverse computational requirements and iterative code optimization. Constructing this supervision is challenging because of heterogeneous specifications, and useful implementations emerge through iterative complicated testing and refinement. We introduce KernelMosaic, an open dataset pairing diverse CUDA tasks with verified solutions and trajectories. Human-guided review and diversity-aware selection curate tasks expressed as natural-language descriptions, reference programs, mathematical specifications, and their intersections. To build the solutions and trajectories, our agentic framework guides a strong LLM through implementation, correctness repair, and performance optimization with reusable CUDA guidance and execution feedback. Supervised fine-tuning of a small open-source model on KernelMosaic raises correctness from 4%, 0%, and 0% to 88%, 82%, and 16% on KernelBench L1, L2, and L3, respectively, while increasing the fraction of tasks with kernels faster than PyTorch eager at every level. Further training experiments show that supervision from the curated trajectories substantially outperforms direct responses and promotes sustained testing and revision. These results demonstrate that KernelMosaic enables a general-purpose model to achieve substantial gains in CUDA kernel generation.

cs.LG · cs.AI

新论文 · v1arXiv:2611.14397

从间隔视角理解 LoRA:对灾难性遗忘的鲁棒性与适配器合并(MaLoRA)

A Margin Perspective on LoRA: Robustness to Catastrophic Forgetting and Adapter Merging (MaLoRA)

Ziqing Xu, Hancheng Min, Lachlan Ewen MacDonald, Salma Tarmoun, Enrique Mallada, Weijie Su, Rene Vidal

AI 速读 基于摘要

该文在多类线性分类的近正交任务设定下,用间隔(margin)视角理论刻画最优 LoRA 适配器,并据此提出免训练的适配器合并规则 MaLoRA。

cs.LG

查看原文摘要

Low-rank adaptation (LoRA) is the de facto method for parameter-efficient fine-tuning of large neural networks, achieving strong task adaptation with minimal trainable parameters and low optimization cost. Beyond this efficiency, practitioners have observed two striking properties: \textit{robustness to catastrophic forgetting} and the ability to \textit{merge independently trained adapters} into a single model that performs competitively across multiple tasks. These properties are central to continual and multi-task learning, yet remain poorly understood. In this work, we provide a theoretical explanation through a unified \textit{margin-based perspective}. We analyze LoRA in multiclass linear classification under a near-orthogonal task regime with $\ell_2$-regularization, and characterize the optimal LoRA adapter across regularization regimes. Our analysis shows that, in an intermediate regime of regularization parameters, the optimal adapter aligns with the \emph{max-margin solution} on the fine-tuning data. Building on this characterization, we derive two key consequences. First, we obtain closed-form expressions for the margins on pre-training and fine-tuning data, revealing a precise margin trade-off: the regularization parameter controls the balance between retention and adaptation. Second, we analyze adapter merging, proving that merged models achieve positive margin on each task and deriving optimal mixing coefficients that balance margins across tasks and maximize the margin over their union. These results lead to a simple, training-free merging rule, which we term \textit{margin-based LoRA merging} (MaLoRA). Experiments on modern architectures and real datasets validate our theoretical predictions, showing that MaLoRA matches or outperforms several adapter-merging baselines across a range of vision and language classification tasks.

版本更新 · v3arXiv:2606.15893

BALTO:面向幻觉缓解的平衡令牌级策略优化

BALTO: Balanced Token-Level Policy Optimization for Hallucination Mitigation

Ning Li, Zixuan Guo, Yan Xu, Wenbo Fei, Yifan Niu, Chang Luo, Yasheng Wang, Weiwen Liu 等 10 位作者

AI 速读 基于摘要

提出 BALTO 框架,通过将声明级事实核验投影为令牌级标签并引入平衡的令牌级信用分配,缓解大语言模型在知识密集型场景中的幻觉问题。

cs.CL

查看原文摘要

Hallucinations remain a major obstacle to deploying large language models (LLMs) in knowledge-intensive settings, where generated responses must be faithfully grounded in provided evidence. Reinforcement learning (RL) is a promising direction for hallucination mitigation, but response-level faithfulness rewards suffer from a granularity mismatch: localized hallucinations can cause supported content to receive spurious penalties. Although recent work introduces fine-grained feedback such as claim-level verification and token-level rewards, unbalanced credit assignment can still induce length, verbosity, or optimization-noise biases. We propose BALTO, a Balanced Token-level Policy Optimization framework for hallucination mitigation. BALTO extracts checkable factual claims, verifies them against the reference context, and projects claim-level judgments to token-level labels. A balanced token-level credit assignment mechanism is introduced into the framework. This design redistributes probability mass from unsupported content toward faithful content, rather than suppressing the entire response. We systematically analyze the limitations of response-level rewards from a theoretical standpoint, and prove BALTO's advantages in training stability and optimization efficiency for hallucination mitigation. Experiments on ConFiQA, RAGTruth, and FinLLM-Eval show that BALTO achieves the highest faithfulness across all six model--benchmark settings and consistently outperforms existing post-training baselines in Q-Score, demonstrating a stronger faithfulness--informativeness trade-off. The code is released at https://github.com/NingLi670/BALTO.

版本更新 · v3arXiv:2603.13652

基于激活修补的因果归因

Causal Attribution via Activation Patching

Amirmohammad Izadi, Mohammadali Banayeeanzade, Alireza Mirrokni, Hosein Hasani, Mobin Bagherian, Faridoun Mehri, Mahdieh Soleymani Baghshah

AI 速读 基于摘要

提出 CAAP,通过在 ViT 中间层直接替换 patch 的内部激活来估计每个图像块对预测的因果贡献,作者报告其在多个骨干和指标上优于现有归因方法。

cs.CV

查看原文摘要

Attribution methods for Vision Transformers (ViTs) aim to identify image regions that influence model predictions, but producing faithful and well-localized attributions remains challenging. Existing attribution methods face several limitations, with gradient-based, relevance-propagation, and attention-based methods relying on local approximations, while perturbation- or optimization-based methods intervene on inputs, tokens, or surrogates rather than internal patch representations. The key challenge is that class-relevant evidence is formed through interactions between patch tokens across layers; methods that operate only on input changes, attention weights, or backward relevance signals may therefore provide indirect proxies for patch importance rather than directly testing the predictive effect of contextualized patch representations. We propose Causal Attribution via Activation Patching (CAAP), which estimates the contribution of individual image patches to the ViT's prediction by directly intervening on internal activations rather than using learned masks or synthetic perturbation patterns. For each patch, CAAP inserts the corresponding source-image activations into a neutral target context over an intermediate range of layers and uses the resulting target-class score as the attribution signal. The resulting attribution map reflects the causal contribution of patch-associated internal representations on the model's prediction. The causal intervention serves as a principled measure of patch influence by capturing semantic evidence after initial representation formation, while avoiding late-layer global mixing that can reduce spatial specificity. Across multiple ViT backbones and standard metrics, CAAP consistently outperforms existing methods in various settings and produces more faithful and localized attributions.

新论文 · v1arXiv:2611.14389

评估用于传达三维形状的触觉编码策略

Evaluating Tactile Encoding Strategies for Communicating 3-D Shape

MacKenzie C. Harnett, Rebecca F. Friesen

AI 速读 基于摘要

比较两种在二维触觉显示器上编码三维形状的方法(单层高度图与二值分层切片),并对比摩擦调制显示屏与针阵列的表现。

cs.HC

查看原文摘要

Two-dimensional haptic displays, such as programmable pin arrays and vibrotactile or friction-modulating touchscreens, enable tactile communication of complex shapes in a low-profile form factor. However, many shapes of interest, including CAD models, three-dimensional graphs, and map topographies, are inherently three-dimensional. This work investigates two methods for encoding 3-D information using a 2-D tactile display: (1) a single tactile height map providing coarse but quantifiable depth cues, and (2) a series of binary layers communicating sequential slices of a 3-D model, supporting more complex geometries such as overhangs and cavities. We further compare performance across two display types--a friction-modulating display and a pin array--to examine how the level of haptic abstraction affects users' ability to decode discrete 3-D information. A user study demonstrates that participants were able to effectively complete a 3-D assembly task using both strategies, achieving high accuracy across all display types. While the pin array supported faster task completion and higher-rated--though not statistically significant--user experience than the friction-modulating display for the binary-layer task, these differences did not translate to significant differences in overall usability (i.e., reassembly accuracy success). Together, these results identify design trade-offs and inform development of future surface haptic displays that support the communication and interpretation of discrete 3-D information in tactile graphics.

新论文 · v1arXiv:2611.14387

模糊块边界:面向高效扩散语言模型的重叠块去噪

Blurring the Block Boundaries: Overlapping Block Denoising for Efficient Diffusion Language Models

Amir Mann, Jonathan Fhima, Shaked Brody, Or Avitan, Asif Zamir, Shahar Tsiper, Niv Nayman

AI 速读 基于摘要

提出免训练、免超参的 B3 方法,让相邻两个块在同一次前向中并行去噪,以加速自编辑块扩散语言模型。

cs.CL · cs.AI

查看原文摘要

Diffusion Language Models (DLMs) have emerged as effective alternatives to autoregressive models for latency-critical tasks, enabling faster token generation. Recent advances employ block diffusion with self-editing capabilities, where each block refines tokens iteratively before advancing to the next one. Yet each block is only as fast as its slowest token, and the next block cannot start before it. We suggest Blurring the Block Boundaries (B3), a novel method with which two blocks are refined in parallel within a shared forward pass. B3 calibrates itself on the fly, opening the next block only once the model is ready to start generating it. It is a training-free and hyperparameter-free acceleration method which seamlessly integrates within any self-editing block diffusion model. Evaluated on state-of-the-art DLMs across 10 diverse datasets spanning varied tasks and difficulty levels, B3 results in on par generation quality and 11.5-14.5% speedup compared to the base model, pushing the Pareto frontier for latency-quality trade-offs in DLM inference.

版本更新 · v3arXiv:2610.33484

能否听见几何?用于文本到B-rep检索的规范不变谱描述子

Can One Hear Geometry? Gauge-Invariant Spectral Descriptors for Text-to-B-rep Retrieval

Hongzhu Yi, Hao Liu, Tao Yu, Tiankun Yang, Yujia Yang, Zhenghao Zhang, Tianyu Zong, Zhenyu Guan 等 15 位作者

AI 速读 基于摘要

提出 SpectraBRep 框架,用基于几何不变性与谱描述子的方法在跨模态对齐前消除 B-rep 的位姿、尺度与边界编码歧义,用于文本到三维实体检索。

cs.CV

查看原文摘要

Text-to-3D solid retrieval faces a fundamental representation gap: while natural language describes intrinsic shapes, Boundary Representation (B-rep) entangles precise geometry with arbitrary variations in pose, scale, and boundary encoding conventions. Existing methods either discretize B-reps at the cost of analytic precision, or learn features from topological graphs without explicit invariance guarantees. We propose **SpectraBRep** (*Spectral Geometry for Text-to-B-rep Retrieval*), a framework based on geometric invariance and spectral descriptors that mathematically eliminates representation ambiguities prior to cross-modal alignment. Formalizing translation, rotation, reflection, uniform scaling, and contour re-indexing via group actions, we exploit the planar-profile-and-extrusion structure of engineering solids to construct a unified spectral descriptor across cross-sectional regions, contour boundaries, and extrusion depths. On public benchmarks, our method achieves state-of-the-art recall performance across all evaluation metrics.

版本更新 · v5arXiv:2606.18048

在破碎阈值处对非拥挤超图进行列表染色

List coloring uncrowded hypergraphs at the shattering threshold

Abhishek Dhawan, Abhishek Methuku, Minh-Quan Vo

AI 速读 基于摘要

该论文证明了非拥挤超图的列表色数达到由优惠券收集启发式猜测的最优前导常数,并由此推出线性超图染色与调色板稀疏化结果。

math.CO · cs.DM · cs.DS

查看原文摘要

A foundational theorem of Ajtai, Komlós, Pintz, Spencer, and Szemerédi asserts that every $n$-vertex $k$-uniform uncrowded hypergraph (i.e., with girth at least $5$) of maximum degree at most $Δ$ contains an independent set of size $cn(\logΔ/Δ)^{\frac{1}{k-1}}$ for some absolute constant $c>0$. Improving an earlier bound of Frieze and Mubayi, Iliopoulos showed that every such hypergraph has list chromatic number at most $(1+o(1))(k-1)(Δ/\logΔ)^{\frac{1}{k-1}}$. Determining the optimal leading constant in this bound remains a major open problem. A natural target, suggested by the coupon-collector heuristic, is $(k-1)^{\frac{1}{k-1}}$, which also corresponds to the shattering threshold for coloring sparse random hypergraphs. Our main result shows that this target is attained: for every integer $k\ge2$, every $k$-uniform uncrowded hypergraph of maximum degree at most $Δ$ satisfies: $$χ_\ell(\mathcal{H})\le(1+o(1))\Big((k-1)\fracΔ{\logΔ}\Big)^{\frac{1}{k-1}}.$$ In particular, this confirms a 2019 conjecture of Molloy for uncrowded hypergraphs. Building on this result, we prove that every $k$-uniform linear hypergraph of maximum degree at most $Δ$ has chromatic number at most $(c_k+o(1))(Δ/\logΔ)^{\frac{1}{k-1}}$, where $c_k\to1$ as $k\to\infty$, thereby proving a conjecture of Verstraëte and Wilson on the independence number of linear hypergraphs in the stronger form of colorings. Our techniques also yield a palette sparsification theorem for uncrowded hypergraphs, partially answering a question of Casselgren and Eriksson.

版本更新 · v3arXiv:2609.29646

顺序 Phragmén 规则保证 2-近似核心稳定性

Sequential Phragmén Guarantees 2-Approximate Core Stability

Jianan Lin, Chenhao Wang

AI 速读 基于摘要

作者证明经典多项式时间投票规则 sequential Phragmén 在认可偏好下总能选出 2-稳定(2-核心)的委员会,并说明该界是紧的。

cs.GT

查看原文摘要

We study approximately stable committee selection under approval preferences, following Cheng et al.~[EC~2019] and Jiang, Munagala, and Wang~[STOC~2020]. In an approval-based committee election, $n$ voters each approve a subset of the candidates, and a committee of size $k$ is selected. A size-$k$ committee is $λ$-stable if, for every nonempty subset $T$ of candidates, the number of voters who strictly prefer $T$ is less than $λ$ times the proportional share $\frac{n|T|}{k}$. We consider sequential Phragmén, a classical polynomial-time voting rule, and prove that it always returns a $2$-stable committee. This implies that the $2$-core is always nonempty and improves upon the $3.651$-stability guarantee of Gao, Sun, and Vondrák~[EC~2026]. The guarantee also extends to weighted voters and is tight even for unweighted elections: for every $1\leλ<2$, sequential Phragmén can return a committee that is not $λ$-stable. Compared with the exact-core algorithm developed in concurrent work by Becker, Greger, and Peters~[arXiv~2026], sequential Phragmén is simpler to implement, and our proof of its $2$-stability is elementary.

新论文 · v1arXiv:2611.14383

基于能量的视角看机器学习

An Energy-Based Perspective on Machine Learning

Bo Dai, Yilun Du

AI 速读 基于摘要

这篇教程以能量模型(EBM)为统一视角,把判别模型、生成模型、表示学习、强化学习与推理时计算等机器学习方法组织成同一框架,并指出它们主要在三项选择上不同:能量、推断与学习。

cs.AI · cs.LG

查看原文摘要

Energy-based models (EBMs) define a distribution through a scalar energy, $p_θ(x)\propto\exp(-E_θ(x))$, and are usually presented as one model family among many. This tutorial has a broader goal: beyond introducing EBMs, it uses them as a single lens that connects the different aspects of machine learning. Any EBM is specified by three choices, the energy, which determines what the model represents; the inference procedure, which samples from or optimizes the energy; and the learning method, which fits the energy to data. We first develop these components, including sampling, variational inference, and learning by maximum likelihood, noise contrastive estimation, pseudolikelihood, score matching, and adversarial training, and show that the learning methods share one contrastive structure. We then show that discriminative models, generative models such as GANs, normalizing flows, diffusion and autoregressive models, representation learning methods such as contrastive learning, word2vec, BERT, denoising auto-encoders, and diffusion representations, reinforcement learning with EBM policies and world models, inverse reinforcement learning, and planning, as well as inference-time computation with verifiers, are all EBMs that differ in these three choices, mostly in how they handle the partition function and sampling. This view explains existing methods, organizes them into a common framework, and suggests new algorithms by replacing one component while keeping the others, with composition of energies as a principled way to combine learned models at inference time.

新论文 · v1arXiv:2611.14382

大语言模型“通过思考学习”机制的认知映射

Cognitive Mapping of Learning by Thinking Mechanisms in Large Language Models

Dong Han, Jin Zhang, Kunyue Yang

AI 速读 基于摘要

该文提出一个认知LbT框架,把四种“通过思考学习”的认知机制映射到大语言模型的计算过程,并讨论其能力边界与结构性失败。

cs.AI

查看原文摘要

Large language models (LLMs) increasingly exhibit capabilities that resemble human Learning by Thinking (LbT), the acquisition of new insight through internal processes such as explanation and reasoning, without new external input. A systematic account of the cognitive mechanisms behind these capabilities, and of their limits, remains underdeveloped. This paper proposes a Cognitive LbT Framework that maps four mechanisms (Reflective Reasoning, Synthetic Memory Formation, Internal Simulation, and Conceptual Self-Organization) onto computational processes in LLMs. Drawing on representative methods, including chain-of-thought prompting, self-consistency, self-reflection, Tree-of-Thoughts search, Self-Instruct, and self-taught reasoning, the analysis argues that selected methods can functionally approximate aspects of these mechanisms and can improve performance on some tasks without additional labeled examples, although this does not always constitute durable learning in model parameters. For example, Self-Instruct generated approximately 52,000 instructions from 175 seed tasks and reported a 33% absolute improvement on Super-NaturalInstructions (Y. Wang et al., 2023). The same mechanisms also produce characteristic failures, including hallucination diffusion, model collapse, concept drift, and high computational cost. These failures are not isolated technical defects but follow from four structural asymmetries between LLMs and human cognition: the identity of generator and monitor, closed learning loops without external grounding, symbolic rather than sensorimotor simulation, and the cost of inference-time computation. The framework yields testable propositions about when LbT in LLMs should succeed or fail and offers a structured basis for designing more reliable autonomous learning systems.

版本更新 · v2arXiv:2609.30746

面向极端事件数据受限仿真的机制感知集合条件化方法

Mechanism-Aware Ensemble Conditioning for Data-Limited Emulation of Extreme Events

Isabella S. Thiel, Juan Bello-Rivas, Yannis G. Kevrekidis, Themistoklis P. Sapsis

AI 速读 基于摘要

提出一种机制感知的条件化插件框架,把不完美模型的微扰集合转化为局部不稳定性几何的传感器,并以 FiLM 网络注入骨干模型,从而在数据受限下改进极端事件仿真。

cs.LG · math.DS · physics.ao-ph

查看原文摘要

Extreme events in chaotic systems are difficult to learn from short trajectories because they are controlled by finite-time instabilities rather than by frequently observed bulk dynamics. We propose a mechanism-aware conditioning plug-in framework that turns a nudged ensemble of an imperfect model into a non-intrusive sensor of the local instability geometry. In the small-noise regime, the ensemble covariance has the same finite-time deformation kernels that govern local instability, providing a Jacobian-free proxy for local amplifications around a trajectory. A small Feature-Wise Linear Modulation (FiLM) network injects the ensemble geometry into an otherwise unchanged backbone without modifying the low-fidelity model simulator. We demonstrate this interface in two settings: a Transformer-style residual-attention corrector for a controlled low-dimensional chaotic system and a probabilistic recurrent STORN corrector for topographic two-layer quasi-geostrophic (QG) flow. In the low-dimensional benchmark, leading ensemble-covariance directions identify transient instabilities that are also detected by Optimally Time-Dependent (OTD) modes. FiLM conditioning improves 99th-percentile exceedance-frequency errors over an otherwise identical unconditioned Transformer. In QG, an ensemble-conditioned FiLM-STORN model substantially improves long-horizon rare-event statistics in the data-limited regime, including density-tail errors, exceedance frequencies, and spatial exceedance-area distributions relative to an unconditioned STORN trained on the same data. On averaged high-threshold exceedance diagnostics, it also outperforms the unconditioned baseline STORN trained with 20 times more expensive trajectory data. These results show that local instability geometry is not merely interpretable post hoc, but an actionable conditioning signal for data-efficient rare-event emulation.

新论文 · v1arXiv:2611.14380

人类-AI协作下的集合级计数:哪些因素真正重要?

What Factors Matter for Collection-Level Counting with Human-AI Collaboration?

Seoyun Jeong, Logan Lawrence, Owen L. Martin, Benjamin Koger, Daniel Sheldon, Subhransu Maji

AI 速读 基于摘要

在统一的人在回路计数框架下跨12个任务比较模型、估计器与适配策略,并给出面向新部署的选型经验。

cs.CV

查看原文摘要

Many scientific and policy decisions require estimating the total number of target objects across a large image collection. Exhaustive human labeling is often infeasible, while raw model predictions can be biased, especially under domain shift. Recent human-AI estimation methods combine imperfect model predictions with limited human verification to obtain reliable collection-level estimates. However, existing work provides little guidance on how practitioners should choose among models, estimators, and adaptation strategies for a new deployment. We study these choices within a common human-in-the-loop counting framework across 12 tasks spanning biomedical microscopy, ground-level and aerial wildlife surveys, and overhead imagery of urban structures. Our results show that modern multimodal models provide a strong, training-free starting point for many tasks, while fine-tuned detectors become more advantageous at larger labeling budgets and across repeated deployments. Estimation efficiency improves further when model predictions are used to prioritize which units humans label, rather than sampling uniformly. We also find that each model's ranking quality can be estimated with a small pilot set, yielding practical guidance for selecting strategies on new counting tasks. In a field deployment on aerial pronghorn surveys, a task-specific adaptation of these methods meets the survey's $\pm 5\%$ tolerance after reviewing a median of only $3.2\%$ of collected images.

版本更新 · v2arXiv:2605.25399

面向端到端、删失感知的大语言模型生存分析

Towards end-to-end LLM-based censoring-aware survival analysis

Yishu Wei, Hexin Dong, Yi Lin, Jiahe Qian, Yi Liu, Yifan Peng

AI 速读 基于摘要

提出 LLMSurvival 框架,将带删失的时间-事件预测改写为可比个体间的成对排序,使未经修改的 LLM 能直接基于表格临床数据做生存分析。

cs.AI

查看原文摘要

Objective: Survival analysis is central to medical prediction, yet large language models (LLMs) are rarely used as end-to-end survival models because censoring prevents straightforward supervised fine-tuning. Here we present LLMSurvival, a framework that enables censoring-aware survival analysis with unmodified LLMs operating directly on tabular clinical data. Materials and Methods: LLMSurvival reformulates time-to-event prediction as pairwise ranking among comparable subjects, and derives test-time risk by aggregating comparisons against anchor individuals from the training cohort. Results: Across two clinical tasks (ICU mortality prediction in MIMIC-IV and fragility fracture prediction in a NewYork-Presbyterian/Weill Cornell Medicine cohort), LLMSurvival improves overall concordance over Cox proportional hazards modeling by 3.1% for ICU mortality and 0.5% for fracture risk, 2.1% on average for ICU mortality and 2.8% for fracture risk over three established deep learning survival models. Discussion: The results show that survival modeling with censoring can be made compatible with LLM fine-tuning through comparison-based reformulation. The framework demonstrates high portability and superior performance over expert curated scores like SAPS-II and FRAX scores across diverse clinical context. Furthermore, the framework supports local deployment, as compact, publicly available base models provide sufficient performance. Conclusion: The LLMSurvival framework serves as a proof of concept for an integrated, censoring-conscious approach to survival analysis via LLMs.

版本更新 · v2arXiv:2606.08243

用对称信念编辑构建比较性动机画像

Building Comparative Motivation Profiles with Symmetric Belief Editing

David Vella Zarb, Rustem Turtayev, Taywon Min, Jinghua Ou, Shi Feng

AI 速读 基于摘要

作者提出“比较性动机画像”与“对称信念编辑”,通过编辑模型对事实的信念并测量行为变化,来区分对齐伪装背后是“追踪训练后果”还是“追踪研究者预期”。

cs.CL

查看原文摘要

When a model behaves in unexpected ways, we want to know why, but its intent cannot be observed directly, and its self-reports may be rationalizations. We introduce comparative motivation profiles, which treat intent as propensity: we edit the model's beliefs about facts that a hypothesized motivation would depend on and measure how its behavior changes. To keep artifacts of the editing method from driving the result, we edit two minimally different hypotheses in the same way, with matched amplify and suppress edits; we call this symmetric belief editing. We apply it to alignment faking in four open-weight model organisms, contrasting scheming, which tracks the training consequences of compliance, with performative misalignment, which tracks what researchers expect to observe, and edit beliefs with synthetic document fine-tuning (SDF), activation steering, and prompting. Under SDF, amplifying performative beliefs raises alignment faking on every model, while scheming edits move it less or shift compliance in the monitored and unmonitored conditions alike. Steering on model organisms changes compliance along the expectation-tracking direction but mainly the reasoning along the consequence-tracking direction, and prompting agrees. In these organisms, alignment faking follows what the models infer their evaluators expect, which puts it closer to sycophancy than to deceptive alignment.

版本更新 · v2arXiv:2610.39941

插入的块交换下降:交换投影与最小掩码

Block-exchange descent for insertion: commuting projections and minimum masks

Haobo Ma, Wenlin Zhang, Charles E. Hughes

AI 速读 基于摘要

研究shuffle表示中最小插入度在块交换下的下降规律,并给出交换投影下的精确刻画与若干计算机辅助验证结果。

cs.FL · math.CO

查看原文摘要

For fixed nonempty words $u,v$, the minimum insertion degree of a shuffle is the least number of inserted-source runs among its representations. We prove that, after a block exchange from a minimum mask of degree $r$, the minimum over masks with commuting source projections is exactly $r-1$; a larger decrease occurs precisely when a noncommuting mask has fewer runs. If either source is unary, every such exchange is exact. We prove an explicit repair for an infinite periodic degree-three family. Computer-assisted results establish first-mask descent through combined length seventeen over every alphabet, degree-three first-mask descent through length thirty, and a uniform repair of all degree-three failures through length thirty, including all further letter identifications. Counterexamples to arbitrary, first, and last minimum-mask choices explain the need for alternative masks. Finally, we prove that minimum iteration depths have no gaps for arbitrary languages. General block-exchange descent and Singleton No-Gap remain open.

新论文 · v1arXiv:2611.14379

判定有限关系结构的Maltsev多态性存在性的复杂性

Complexity of Deciding Existence of Maltsev Polymorphisms of Finite Relational Structures

Dejan Delic, Ali Syed

AI 速读 基于摘要

作者报告判定有限关系结构是否具有Maltsev多态性可在多项式时间内完成,并借助二元关系矩形性给出归约证明。

cs.CC

查看原文摘要

In this article, we show that the problem of deciding whether a finite relational structure $\mathbf{A}$ admits a Maltsev polymorphism is a computational problem, which can be solved in polynomial time. We provide a proof of this result by reducing the problem to deciding whether a binary relational structure, associated with $\mathbf{A}$ admits such a polymorphism. In our construction of such an algorithm, we make an extensive use of a property of binary relations, which is implied by the existence of a Maltsev polymorphism, the so-called rectangularity property.

新论文 · v1arXiv:2611.14376

保留量子服务结果的引用更新操作极限

Operational limits of reference renewal for retained quantum-service results

Roberto Fernández-Barrios, Iker Pastor-López, Amaia Pikatza-Huerga, Pablo García Bringas

AI 速读 基于摘要

作者报告了一项按预设协议对保留的量子服务结果实施引用更新的实验,结果显示该更新未能确立恢复假设,并暴露出若干操作限制。

quant-ph · cs.SE

查看原文摘要

We implement reference renewal to reconsider retained physical quantum-service results under a prespecified protocol. Twelve transitions retained six results each; ten completed renewal, eight met the prespecified endpoint, and five succeeded using retained results alone. The exact one-sided p=0.0546875 does not establish the recovery hypothesis. Incomplete references, binding failures and sequential waits expose operational limits. Offline reconstruction recovers decisions from preserved counts. Retrospective binding-plus-oracle checking accepts all 60 usable-transition results without renewal. With the exact oracle also supplying calibration scores, the study does not demonstrate incremental numerical-correctness utility of renewal.

新论文 · v1arXiv:2611.14374

不确定性感知的混合离线强化学习:基于稀疏高斯过程评论家的在线动作选择

Uncertainty-Aware Hybrid Offline Reinforcement Learning with Online Action Selection Enabled by Sparse Gaussian-Process Critics

Marco Chan, Christos Evripidou, Liqiu Dong, Benoit Chachuat, Mehmet Mercangöz

AI 速读 基于摘要

该研究用稀疏变分高斯过程评论家替代HOFLON中的树集成评论家与对抗自编码器惩罚,在聚合反应器基准上考察不确定性调整动作选择的效果。

cs.LG · eess.SY

查看原文摘要

Start-ups and grade changes remain largely operator-driven, while developing and maintaining dynamic models for model-based control requires substantial engineering effort. Hybrid Offline Learning and Online Optimisation (HOFLON) uses one-step offline reinforcement learning to select actions from historical transition data. In its original implementation, the online objective combines the predicted return from a tree-ensemble critic with an adversarial-autoencoder (AAE) penalty for unfamiliar state-action pairs. We examine the effect of this penalty and replace the critic and AAE with a sparse variational Gaussian-process (sGP) critic. Actions maximise $μ- κσ$, an uncertainty-adjusted estimate of the return from one action followed by the historical operating practice. On the HOFLON polymerisation-reactor benchmark, removing the AAE from HOFLON led to 4 failed start-ups in 15 runs over five critics. At its published weight, the variation in the AAE penalty is four orders of magnitude smaller than the variation in the tree-ensemble critic. Nevertheless, it changes actions when the predicted returns of competing candidates are nearly equal, and paired runs with and without the penalty diverged from the first move. With a Matérn-5/2 sGP critic, $κ= 0$ led to failure in 13 of 15 runs, whereas with $κ= 3$ none of 15 runs (five critics, three noise realisations) failed. The median and lowest returns were 13,464 and 13,398 (HOFLON: 13,393 and 12,411). A squared-exponential sGP critic with learned inducing inputs needed $κ\geq 3$ to remain within the empirical data support in steady state, and then settled with a temperature offset of 2.0 K.

新论文 · v1arXiv:2611.14373

高斯过程增强的模型预测轮廓控制:用于全驱动AUV的六自由度轨迹跟踪

Gaussian Process-Augmented Model Predictive Contouring Control for 6-DoF Trajectory Tracking with a Holonomic AUV

David Dörner, Giancarlo Troni, Ignacio Torroba, Ivan Stenius

AI 速读 基于摘要

作者提出一种面向全驱动AUV的六自由度非线性模型预测轮廓控制方法,并用高斯过程学习残差动力学来补偿标称模型误差。

cs.RO

查看原文摘要

Small, holonomic Autonomous Underwater Vehicles (AUVs) enable close-range inspection of underwater structures from arbitrary viewpoints, requiring accurate position and attitude tracking under state and input constraints. Model Predictive Contouring Control (MPCC) naturally handles these requirements, but relies on an accurate dynamics model, which is difficult to obtain underwater. In this paper, we present a nonlinear MPCC in all six Degrees of Freedom (DoF) for a holonomic AUV and alleviate the inaccuracies in the dynamics by augmenting the nominal model employed with residual dynamics learned via Gaussian Process (GP) regression. The learned residual is embedded in the optimal control problem through a zero-order robust optimization scheme, which keeps the controller real-time capable at 25 Hz. A simulation ablation shows that residual selection is essential when designing the controller, and simulation and in-water experiments with the Multimodality, Observing, Low-cost, Agile (MOLA) AUV validate the resulting controllers against a nominal MPCC and a field proven Proportional-Integral-Derivative (PID) controller.

版本更新 · v2arXiv:2412.01968

论数据交换经济的理论基础

On the Theoretical Foundations of Data Exchange Economies

Hannaneh Akrami, Bhaskar Ray Chaudhury, Jugal Garg, Aniket Murhekar

AI 速读 基于摘要

提出无支付数据交换框架,要求互惠公平与核心稳定,并证明在单调连续效用下公平稳定交换存在,进而研究近似解的计算复杂度。

cs.GT

查看原文摘要

Organizations increasingly seek to share and access datasets to improve their ML models and derive insights. Despite the immense demand for quality data, data exchange and collaboration have not reached their full potential, as cited by the European Commission. One of the key reasons is the lack of reciprocity, where some participants perceive their contribution to others to be of higher value than what they receive in return. We propose a general framework for data exchange without payments, focusing on two main properties: (i) reciprocal fairness, which ensures that each agent receives utility proportional to their contribution to the total welfare -- contributions are quantifiable using well-established credit-sharing functions such as the Shapley share and proportional share, and (ii) core-stability, which guarantees that no coalition of agents can identify an exchange among themselves which they all unanimously prefer to the current exchange. Surprisingly, we show that fair and stable exchanges exist for all monotone continuous utility functions. Building on this, we investigate the computational complexity of finding approximate fair and stable exchanges. We present a local search algorithm for instances with monotone submodular utility functions and cross-monotone credit-sharing functions, e.g., Shapley share. We prove that this problem belongs to CLS ($=$ PPAD $\cap$ PLS) under mild assumptions. Our framework opens up several intriguing theoretical avenues for future research on data economics.

版本更新 · v2arXiv:2608.22731

当言语与身体不一致:实现言语与非言语的不匹配

When Words and Bodies Disagree: Realizing Verbal and Nonverbal Mismatch

Parisa Ghanad Torshizi, Stacy Marsella

AI 速读 基于摘要

提出言语—非言语不匹配的分类体系,并评估LLM能否依据社会语境选择恰当的不匹配行为,以及这些行为在虚拟人中对观察者感知的影响。

cs.AI · cs.HC · cs.RO

查看原文摘要

Traditional nonverbal behavior generation systems for virtual agents typically take utterances as input and generate behaviors that match or illustrate the spoken words. However, human nonverbal behavior is shaped by more than speech content. It is influenced by speaker roles, interpersonal relationships, social context, and cognitive and emotional states of interactants. Therefore, the nonverbal channel carries information or emotions hidden in words. Modeling these complex verbal-nonverbal mismatches is critical for realistic virtual agents, particularly in training applications. We propose a taxonomy of mismatches, strengthening, mitigating, and leaking, and evaluate whether LLMs can select contextually appropriate mismatched behaviors. Through human-subject studies, we assess whether these context-driven behaviors, realized in a virtual human framework, produce the intended effects. Our results demonstrate that LLMs, when prompted with social context, successfully select targeted mismatched behaviors (e.g., sarcasm). Furthermore, the human-subject evaluation confirms that these nonverbal cues significantly shift observers' perceptions of the agent's attitude and emotional valence. Participants successfully distinguished between distinct mismatch functions, such as strengthening, mitigating, and leaking, perceiving clear directional shifts in social meaning compared to matching baselines.

版本更新 · v3arXiv:2609.14743

量化语音-文本语言模型中的生成模态差距

Quantifying the Generation Modality Gap in Speech-Text Language Models

Ju-Chieh Chou, Jiawei Zhou, Karen Livescu

AI 速读 基于摘要

作者在匹配数据和生成设置下统一评测纯语音、纯文本与语音-文本语言模型,发现联合建模显著提升语义连贯性但并非所有指标都一致改善。

cs.CL

查看原文摘要

Pure speech language models often lag behind text and speech-text language models in generating coherent content, but this gap is difficult to quantify because speech and text systems are typically evaluated with different metrics and trained on different data. We study the speech-text modality gap in a family of spoken language models, based on flow matching for continuous acoustic feature generation. We construct a unified generation-based evaluation suite that compares speech-only, text-only, and speech-text language models trained on matched data distributions and evaluated in matched generation settings. We evaluate generated continuations along multiple dimensions: semantic coherence, measured by transcribing generated speech and scoring it with a reference language model; local phonetic structure, measured by phone n-gram distributional statistics; speaker consistency and acoustic quality; and emotion-based distributional metrics. Across datasets, we find that joint speech-text modeling substantially improves semantic coherence. However, the improvement is not uniform across metrics: phone-level metrics change only modestly, speaker similarity and predicted quality are lower for speech-text continuations, while emotion-based distributional metrics improve. Compared with larger-scale speech-only models, our speech-text model closes much of the scaling gap in transcript-based semantic coherence, suggesting that text provides an efficient semantic training signal for spoken language modeling.

版本更新 · v3arXiv:2606.24853

面向物联网边缘环境的轻量级开源网络数字孪生蓝图

A Lightweight, Open-Source Network Digital Twin Blueprint for IoT-Edge Environments

Josevany do Amaral, Rute C. Sofia

AI 速读 基于摘要

该教程基于IRTF参考模型,介绍如何用开源工具构建轻量级网络数字孪生,并讨论无线损伤建模与孪生保真度验证流程。

cs.NI

查看原文摘要

IoT-Edge-Cloud infrastructures are increasingly heterogeneous, and experimenting across the continuum is difficult and risks disrupting live systems. Network Digital Twins (NDT) are a pragmatic answer, yet practitioners lack a structured account of three things: how to build an NDT from open tools, how to model wireless impairments, and how to understand whether the twin is faithful to the physical network, or not. This tutorial introduces NDT through the IRTF reference model and walks through a lightweight open-source blueprint built on Containerlab, Open vSwitch, \textit{Open Network Operating System (ONOS)}, and Prometheus/Grafana. It then explains how to define wireless impairment profiles (static, model-based, trace-driven, AI-generated, and live), and what each approach can and cannot reproduce. Finally, it presents a workflow to validate the twin against the physical network. A proof-of-concept Edge case study illustrates the procedure, including the pitfalls encountered and the lessons learned. Directions for future research are outlined.

新论文 · v1arXiv:2611.14365

从视频帧到机器人元帧:面向世界动作模型的统一潜在接口

From Video Frames to Robot Meta-Frames: A Unified Latent Interface for World Action Models

Guoqiang Liang, Yiqi Lin, Yanzhe Chen, Kevin Yuchen Ma, Mike Zheng Shou

AI 速读 基于摘要

作者提出 MetaWAM,将机器人各类低维信号统一折叠进一个频域元帧,以对齐预训练视频扩散模型的潜在先验,并在仿真与真实机器人任务上报告了性能与效率结果。

cs.CV · cs.RO

查看原文摘要

Pretrained video diffusion models (VDMs) offer a powerful foundation for World Action Models (WAMs), which jointly predict robot actions and future observations. Building such models requires injecting low-dimensional robot signals (e.g. proprioception, actions, and values) into a latent space shaped by natural videos, and a key design choice is the interface through which these signals enter: existing schemes treat it as a formatting detail, spending a separate latent frame on each signal or bolting on dedicated encoders and decoders. We argue that signal injection is instead a distribution-alignment problem: how well a representation matches the prior of the pretrained VDMs during denoising. Using the relative reconstruction error of a frozen VDM denoiser as a probe, we find that conventional spatial encodings sit far off this prior, whereas frequency-domain packing yields latents that the denoiser corrects far less. Motivated by this diagnostic result, we propose MetaWAM, which folds all robot signals into a single unified meta-frame: each signal is assigned to a disjoint frequency-domain channel band and lifted to the spatial latent by an inverse DCT, yielding a decodable interface aligned with the pretrained VDM prior without any added module or architectural change. A fixed band layout together with slot-aware routing lets the same meta-frame support policy, world-model, and value prediction over a shared set of robot variables. On RoboCasa, MetaWAM reaches 68.9% success while collapsing the per-signal frames into one and cutting inference latency by 39%, and attains 85.0% success on three real-world Franka tasks. Code and models are available at https://github.com/showlab/MetaWAM

新论文 · v1arXiv:2611.14364

ARE-bench:在真实研究者会话与工作区中评测智能体的研究能力

ARE-bench: Benchmarking Agents' Research Ability in Real-World Researcher Sessions and Workspaces

Jiahang Lin, Wanying Ren, Wanghan Xu, Huayu Sha, Yang Shu, Zhonghang Lu, Chengjun Pan, Yi Dong 等 27 位作者

AI 速读 基于摘要

ARE-bench 用真实研究者的会话与未清理工作区构建 75 个任务,检验研究智能体能否在延续项目时遵守先前决策与工作区状态。

cs.CL

查看原文摘要

Research agents achieve increasingly strong benchmark scores, yet these scores do not necessarily translate into reliable delivery in ongoing research projects. Many research-agent benchmarks emphasize standalone tasks and relatively self-contained workflows in clean or curated workspaces. However, researchers ask agents to complete one step within an ongoing project, where requests depend on earlier session decisions and workspaces contain unfinished files, outdated versions, and failed attempts. We introduce ARE-bench to evaluate whether research agents advance ongoing projects while respecting prior decisions and existing workspace state, grounded in real session context and in-progress workspaces. ARE-bench includes 75 tasks built on agent sessions and full workspaces from recruited researchers' own projects, spanning seven research stages, 33 requirements, and 150 sub-requirements, with 1,173 binary checklist items in total. Each task is anchored at a real user request, preserves its requirements, and summarizes the earlier requests still in effect as background. The complete workspace is replayed to that boundary without cleanup, preserving artifacts from earlier turns. Each checklist is compiled from an intent-aligned rubric and credits any valid solution. Across seven frontier models, the strongest passes every checklist item on only 66.2% of tasks, and Design and Execution rank among the hardest stages for every model. Three models pass about 69% of tasks when only delivery contracts and general norms are checked, but just 28.4% to 40.0% once historical and material or workspace requirements are also checked. Evaluating delivery without these contextual requirements can therefore overestimate research agents and mask differences between them.