Detecting out-of-distribution (OOD) objects is essential for ensuring the reliability and safety of object detectors in real-world deployment. We treat the decision on a detected box as a hypothesis test with the null hypothesis that the box covers an object of an in-distribution (ID) class. The null distribution changes with the size of the box and its contact with the image border, and the detector observes both. The test can therefore rank a box only against ID boxes of the same size and border contact, which gives a distribution-free bound on the false-alarm rate for each of them and uncovers the separation that one null distribution hides. The power of the test also rests on the layer it reads. Layers built only for the ID classes have no reason to keep what sets an unknown object apart, so the test reads the features before them. We propose \textbf{PLUM}, a training-free score that pools the last backbone stage inside the box and ranks its distance to the ID classes among ID boxes of the same size and border contact. On the standard benchmark, PLUM reaches the highest AUROC among the compared methods without external models, with 91.5 and 94.5 on PASCAL-VOC and 95.1 and 97.5 on BDD100K. Code: https://github.com/DDOZMI/PLUM
AI agents are trained on population-scale data to encode broad capabilities spanning those of many practitioners. Yet the artifacts they produce rarely meet the personal bar professionals need to stake their reputation on. On realistic, open-ended tasks where success criteria are heterogeneous and insufficiently documented, individual expertise lives precisely in the elevation and departure from the average. In practice, iterative human-agent interaction surfaces criteria that users cannot fully specify up front, yet apply repeatedly across tasks. We argue this cross-session interaction data is a rich, underused signal for closing the gap to individual expertise. In this work, we propose test-time adaptation through human-agent interaction (TAHI), which integrates these signals into agent context and weights, and crystallizes each user's training and evaluation criteria via an evolving rubric module. We adapt agents to 30 individuals in two high-utility domains, writing and visual creation, on a total of 600 tasks. Our agents improve solo task success by 4.5-20.9% within only tens of tasks. Meanwhile, our evolving rubric module serves as a scalable annotation tool, creating evaluation rubrics that catch 16.0-22.3% more failures than those from LMs or humans alone. While agents are adapted towards individuals, we show these personalized agents also produce improvements in success of up to 8.8% that generalize across users.
Beyond Blossom: Accelerating MWPM Decoding for Surface Codes With Local Predecoding and Architecture-Aware Optimizations
Jiayu Fu, Zirong Shen, Lin Gan, Yiming Zhang, Wenjing Han, Xiaoming Sun, Riling Li, Guangwen Yang
AI 速读 基于摘要
Beyond Blossom 是一个基于 CPU 的 MWPM 表面码解码器,它通过局部预解码、架构感知优化和流水线并行,在保持 MWPM 最优性的同时降低匹配工作量与执行开销。
cs.IT · cs.PF · quant-ph
查看原文摘要
Minimum-weight perfect matching (MWPM) is widely used to decode surface codes, but its computational cost increases with code distance. We present Beyond Blossom, a state-of-the-art CPU-based MWPM decoder for surface codes. It combines algorithmic optimization through local predecoding, architecture-aware optimization, and pipeline-wide parallelism to reduce matching workload and execution overhead while preserving MWPM optimality. Across rotated surface codes with distances 9-33 and circuit-level physical error rates of 0.05%-0.3%, on a single CPU core, Beyond Blossom achieves a 4.23$\times$ geometric-mean speedup and up to 5.58$\times$ speedup over Sparse Blossom in PyMatching. At distance 33 and a physical error rate of 0.1%, with 8 CPU cores, decoding a block of 33 measurement rounds takes 21.40 $μ$s on average, equivalent to 0.65 $μ$s per round.
Decoder-only Large View Synthesis Models (LVSMs) with KV-cache have recently achieved state-of-the-art quality by regressing novel views from neural networks without reconstructing 3D geometry, while Test-Time Training (TTT) layers save computation but sacrifice quality. We propose Test-Time Baking (TTB), a strategy that learns to bake global scene information from cross-view attention into a lightweight MLP, enabling fast and high-fidelity novel-view decoding. The first key is to repurpose the fast-weight MLP of TTT to learn the cross-view attention mapping explicitly from its input-output pairs, rather than replacing attention entirely, achieving the same rendering efficiency as TTT while delivering substantially higher quality. The second key addresses a modality misalignment between context prefilling and novel-view rendering: we disentangle context tokens into camera-only context rendering tokens to serve as TTB input, and context ground-truth tokens with image content to serve as the TTB mapping target, ensuring the baked MLPs are relieved of the modality generalization challenge between context and novel-view processing. Our method achieves up to **23x** rendering speedup with quality on par with the state-of-the-art across several benchmarks. Project page: https://test-time-baking.github.io
Applying machine learning systems to new domains introduces bottlenecks in both labeled data acquisition and manual pipeline engineering. While recent MLE agents show promise in automating pipeline design, they have been studied almost exclusively in fully supervised settings, leaving them untested in realistic regimes where labeled data is scarce but unlabeled data is abundant. In this work, we introduce Agentic Semi-Supervised Learning (AgentSSL), framing SSL as a program search problem and studying how agents navigate this space under limited supervision. We find that even under weak feedback and matched compute, agents can synthesize code that combines and adapts established SSL techniques, exceeding the strongest human-engineered baselines in 8 of 12 settings by up to +21.8%. However, our results identify program selection as a critical bottleneck in AgentSSL, where performance drops by as much as 47% when selecting based on the agent's feedback metric. To systematically study this bottleneck, we introduce ProgramSelectBench, a large-scale benchmark with a combined total of 16,977 agent-generated and human-engineered SSL programs, and evaluate both unsupervised and active program selection methods. We further evaluate AgentSSL on real-world ecology detection tasks, highlighting the opportunities and challenges of transferring agent-discovered strategies beyond controlled benchmarks.
Large Language Models (LLMs) exhibit strong informal mathematical reasoning but struggle to generate mechanically verifiable proofs in formal languages like Lean. We present LEAP, an agentic framework that enables general-purpose foundation models to achieve state-of-the-art performance on automated formal theorem proving. LEAP leverages foundation model capabilities, such as informal reasoning, instruction following, and iterative self-refinement. By decomposing complex problems into smaller units, the system bridges formal proof construction with informal blueprints through continuous interaction with the Lean compiler. To provide a rigorous evaluation beyond increasingly saturated benchmarks, we introduce Lean-IMO-Bench, a benchmark of IMO-style problems formalized in Lean, with short statements yet highly non-routine and multi-step proofs across a wide range of difficulty levels. Empirically, on the latest 2025 Putnam Competition, an annual mathematics competition for undergraduate students in North America, LEAP solves all 12 problems, matching recent breakthroughs by frontier formal mathematical models. On Lean-IMO-Bench, LEAP boosts the one-shot formal solve rate of general-purpose LLMs from below 10% to 70%, notably surpassing the 48% benchmark set by a specialized, gold-medal-caliber IMO system. Furthermore, we demonstrate LEAP's research-level utility by autonomously formalizing complex proofs for open combinatorial challenges, including a verified proof for a key subproblem in Knuth's Hamiltonian decomposition of even-order Cayley graphs.
We study the ε-capacity regions of mixed multiple-access channels (MACs) with general mixture and cost constraints by the information spectrum method. The starting point is a general formula for the ε-capacity region of general MACs with cost constraint, together with the random-coding and converse lemmas from which it is derived. For the quasi-static fading Gaussian MAC, with random fading coefficients and noise variance, we show that the ε-capacity region is the ε-outage capacity region, which gives the latter a Shannon-theoretic operational meaning. The direct part follows from the general formula, and the converse part from the strong converse property of the Gaussian MAC, which is stronger than the ε-strong converse, applied to each component. The ε-capacity regions coincide in all four scenarios of availability of channel state information (CSI), and the results extend to K users. With CSI at both terminals, variable-rate coding enlarges the 0-capacity region to the polymatroid whose rank function is the average of those of the components. For mixed MACs with general components and a countable parameter space, CSI at the receiver alone never enlarges the ε-capacity region, and the region with CSI at the transmitters is given by an information spectrum formula; unlike the Gaussian case, the regions with and without CSI at the transmitters may differ. For uncountable parameter spaces, only the formula for the region with CSI at both terminals survives in general. Parts II and III treat finite-alphabet mixed MACs and the strong converse for the K-user Gaussian MAC, respectively.
Federated learning (FL) enables independent data owners to train models collaboratively without centralizing raw data. However, identifying collaborators, preparing training tasks, and coordinating execution and incentives hinder its practical adoption by organizations without specialized expertise. We present Cortex-FL, an agent-native FL system that enables users to express learning objectives in natural language and delegate the collaboration required to train a model. Agents translate these objectives into shared tasks, discover collaborators, and negotiate participation agreements, connecting collaboration formation with distributed training and incentive management. The architecture separates agent coordination from local execution authority, allowing owners to delegate coordination while retaining control over their data, computation, and financial commitments. We demonstrate the complete workflow on a four-node deployment, in which a model requested in natural language proceeds through collaboration formation, contract-governed training, verified model delivery, and x402 settlement on a blockchain testnet.
作者用拓扑方法给出 Kneser 图 K(n,k) 的 boxicity 新下界,并首次给出关于 n 次线性的上界。
math.CO · cs.DM
查看原文摘要
For a graph $G=(V,E)$, the boxicity $\operatorname{box}(G)$ is the least integer $d$ such that $G$ is the intersection graph of axis-parallel boxes in $\mathbb{R}^d$. The Kneser graph $K(n,k)$ is the graph whose vertex set corresponds to the family of $k$ sized subsets of an $n$ sized ground set, and a pair of vertices are adjacent if and only if the corresponding sets are disjoint. In this short paper, we provide new bounds for the boxicity of $K(n,k)$. We first show that for $k\ge 2$ and $n\ge 2k+1$, $\operatorname{box}(K(n,k))\ge n-2k$, thereby improving the lower bounds of Caoduro and Lichev [Discrete Mathematics 346(5):113333 (2023)] when $n\geq 3k+1$ and $k\ge 3$. The proof of this result uses the topological method and follows a strategy similar to Greene's proof of the Kneser's conjecture on the chromatic number of $K(n,k)$. We also show that for $k\ge 2$, $t\ge 1$ and $n=2k+t$, $\operatorname{box}(K(2k+t,k)) \leq O( t^3\log(2k+t) ).$ In particular, for every $t=t(k)\ge1$ with $t^3\log k=o(k)$, $\operatorname{box}(K(2k+t,k))=o(k)$. This is the first upper bound on the boxicity of $K(n,k)$ that is sublinear in $n$. We also show that for large enough $k$, $\operatorname{box}(K(2k+1,k))\le \frac{3\ln (2k+1)}{\ln\ln (2k+1)}$.
Embodied interaction requires an agent to integrate multimodal perception with verbal and nonverbal behaviors to effectively communicate its intentions, including speech, body motion, and facial expressions. We present XBot-Omni, a Thinker--Talker--Actor architecture that extends the Omni paradigm to robotic modalities, including body motion and facial expressions, with ``X'' representing the extensibility to robotic-specific modalities. The Thinker processes text, images, audio, and video to formulate contextually appropriate responses, which are then realized by the Talker and Actor through speech and coordinated bodily expressions, respectively. In this report, we innovatively introduce the concept of ``Actor'' as a dedicated component for embodied agents: a coordinated, streaming module that leverages the native representations of the Thinker and Talker within an end-to-end pipeline, introducing only modest latency overhead. This unified architecture enables multimodal generation conditioned on omni-modal inputs directly to prevent information loss in staged designs. We comprehensively evaluate XBot-Omni across seven benchmarks, demonstrating strong multimodal understanding capabilities on both general-domain and robot-centric videos, as well as high-quality, natural, and well-coordinated motion generation. Furthermore, our results validate its capability to generate embodied multimodal responses directly from omni inputs, meeting the diverse requirements of embodied interaction. XBot-Omni enables robots not only to perceive the omni world but also to express themselves through speech and bodily behaviors, extending Omni models from conversational interaction to embodied communication in physical environments.
Large language models (LLMs) show growing promise in specialty medicine, yet existing evaluations largely focus on isolated clinical encounters rather than assessing the distinct clinical responsibilities along longitudinal care pathways. We introduce NPC-RubricBench, a phase-aware benchmark for evaluating multistage clinical reasoning across the cancer care continuum, using nasopharyngeal carcinoma as a defined disease context. Curated from real-world records of 405 patients at two cancer centers, it comprises 941 evaluation units across six sequential care stages, from work-up through follow-up, and 17,688 patient-specific criteria across five dimensions. Each assessment is restricted to clinical evidence available at the corresponding decision point. Across 30 general-purpose and medical LLMs, we find strong diagnostic scores did not ensure reliable downstream care. Decision Quality ranged from 56.8 to 94.7 in diagnosis on a 0-100 scale, but dropped to 8.9-56.0 in staging and spanned 32.4-94.2 in treatment. Crucially, failure modes differed across phases: explicit decision errors predominated in anatomical staging (48.2%), whereas decision omissions were more frequent during initial work-up (33.4%) and post-treatment follow-up (30.4%). Varying query granularity revealed a trade-off between decision quality and other clinical requirements. Broad open-ended requests reduced decision quality across all 30 models, whereas more detailed, checklist-like prompts improved decision quality but were accompanied by lower evidence faithfulness and safety adherence. These findings highlight the need to assess clinical reliability against the distinct clinical requirements of each phase along the longitudinal care pathway. Finally, NPC-RubricBench informs the design of stage-aware clinical guardrails and offers a reusable blueprint for evaluating LLMs across other cancers and chronic diseases.
作者提出 TuMAP 流程,结合双视角基础模型表征、形态型标注与患者级统计推断,从 PAS 染色肾活检切片中量化肾小管形态型组成及形态型内差异。
cs.CV · eess.IV
查看原文摘要
Quantifying disease-associated renal tubular remodeling is challenging because tubular appearances are heterogeneous. We introduce TuMAP (Tubular Morphotype-Aware Pathomics), a pipeline combining (1) dual-view foundation-model representations and provisional morphotype annotation, (2) compositional and within-morphotype pathomics, and (3) patient-level statistical inference. From 456,158 tubules in 358 PAS-stained kidney biopsy slides, TuMAP identified eight image-based morphotypes, seven of which showed stable subsampling assignments. Across 219 patients in six diagnostic groups, diabetic nephropathy (DN) and tubulointerstitial nephritis/acute tubular injury (TIN/ATI) each differed from minimal change disease in seven morphotype proportions, while IgA nephropathy differed in six. DN showed reduced nuclear solidity and increased eccentricity across all eight morphotypes. Area-normalized nuclear density exposed additional within-morphotype differences. TuMAP provides a framework for separating changes in tubular morphotype composition from phenotypic differences within morphotypes.
In the Online Transportation problem, introduced by Kalyanasundaram and Pruhs (1995), requests arrive sequentially in a metric space and must be assigned immediately and irrevocably to one of $m$ facilities with available capacity. The objective is to minimize the total distance between requests and their assigned facilities. The problem generalizes the Online Metric Matching problem, in which each facility has unit capacity, and inherits a $2m-1$ lower bound on the competitive ratio for all deterministic algorithms. For Online Metric Matching, Kalyanasundaram and Pruhs (1993) and Khuller, Mitchell, and Vazirani (1994) established an optimal deterministic competitive ratio of $2m-1$, attained by the Retrospective (Permutation) algorithm. Kalyanasundaram and Pruhs (1998) conjectured that the same ratio is achievable for online transportation. Recently, Seth, Das, and Raghvendra (2026) obtained a competitive ratio of approximately $6.66m-2.89$ using their Robustified Greedy algorithm. We give an algorithm with competitive ratio $ρ(m-1)+1$, where $ρ\approx5.64$. Our algorithm builds on minimum-spanning-tree search techniques developed by Harada and Itoh (2025) and Arndt, Moseley, Pruhs, and Uetz (2026), using priority-based virtual depth-first searches over the known facilities. However our algorithm uses a phase to shift the geometric search scales, to decide when to skip certain searches and optimize the thresholds that define what edges to include in a given search.
Large foundation models such as GPT-6 Astra and Claude Fable 5.1 have recently been shown to control robots directly from visual observations. They nevertheless remain brittle on long-horizon, contact-rich tasks in real-world environments. Further progress depends on two complementary resources: demonstrations of human behavior that supervise how tasks are performed, and interactive environments in which policies can be rolled out, evaluated, and improved in closed loop. Language models have advanced along well-characterized scaling laws; the corresponding question for physical AI is how capability scales with environments and behaviors, which first requires identifying which data to scale and how to produce and reuse such data continuously. To address this question, we introduce Ropedia Xperience 2, comprising 1,589 released metric-scale 3D scans and a collection of 4,230 egocentric recordings of 74 tasks (131 hours). We further show that each capture can be repurposed for robot learning: scenes are converted into physically interactive simulation environments, human motion is retargeted to whole-body robot trajectories, and additional robot behaviors are obtained through virtual teleoperation and automated long-horizon trajectory generation. We propose this joint capture of environments and behaviors as a gold standard for physical AI data collection, laying a data foundation for establishing scaling laws over environments and behaviors.
Hosted retrieval-augmented generation (RAG) and semantic search allow users to query valuable provider-held corpora, raising two competing demands: to hide each query and chosen result, yet reveal only the documents that the user is authorized to receive. Existing cryptographic approaches either make this costly by processing the entire corpus for every query, or sacrifice quality for efficiency by scanning a few clusters. We repurpose learned deep hashing as a private filter: a randomized binary code points the provider to a short candidate list, while encrypted reranking and oblivious key transfer protect the precise query and final selection. This shortlist short-circuits full-corpus cryptographic search without sacrificing retrieval quality: with 200-500 candidates, it closely matches full-corpus retrieval across five zero-shot corpora spanning 25K to 5.4M documents. On the full 2.68M-passage NQ corpus over a 10-Gbps link, our protocol only adds 0.38 seconds, or 6.5%, to a 128-token Qwen3-32B RAG pipeline. The released code satisfies directional metric differential privacy (DP) and substantially reduces embedding-inversion and property-inference leakage, demonstrating that a carefully learned shortlist can make private dense retrieval both accurate and practical.
LLM agents are increasingly used for open-ended, data-driven scientific discovery, yet most of them evaluate hypotheses pointwise, testing an empirical claim $h$ only against its complement $\neg h$. When evidence rejects $h$, the agent learns that one explanation is unlikely but not which alternative the evidence supports. To recover that alternative, we introduce Contrastive Discovery, which replaces the single hypothesis with a contrast set, a scientific question paired with mutually exclusive and collectively exhaustive competing explanations. Each experiment is scored by how its evidence redistributes belief across the set, and the question is resolved when belief concentrates on one explanation. Because contrast sets change only the unit an agent proposes and scores, an existing agent can use them to re-evaluate evidence it has already collected or to guide its search. Across six domains, $89\%$ of the pointwise candidates with the largest belief shifts are rejections. On the same evidence, contrast sets resolve $73\%$ of these candidates to a specific alternative. When contrast sets guide search, the agent resolves $67\%$ more questions than pointwise surprisal under the same experiment budget. In a blinded study, $38$ human experts chose the same explanation on $78\%$ of resolved questions and on $82\%$ of questions where the pointwise agent rejected its hypothesis and stopped. Together, these results support explicit competing explanations as a way for discovery agents to resolve more scientific questions from the same evidence and validation budget.
Fetal brain biometry provides a quantitative basis for assessing prenatal growth and identifying congenital brain abnormalities. However, comprehensive assessment remains limited by labor-intensive manual measurements and automated models that lose accuracy across imaging sequences and institutions. We introduce BioTTA, an anatomically guided test-time adaptation framework that measures 11 fetal brain biometrics from reconstructed three-dimensional magnetic resonance images. BioTTA adapts pretrained models to individual unlabeled volumes using geometric consistency and atlas priors, without access to source images. In a retrospective study across five centers in China and Switzerland, BioTTA improved measurement accuracy across three out-of-distribution datasets and two network architectures. It reduced mean absolute measurement errors by 46-82% relative to the corresponding models without adaptation, achieving errors similar to those of supervised fine-tuning. Its measurements were mapped onto normative growth trajectories derived from expert annotations, enabling gestational-age-specific assessment through a web-based reporting platform. Across four disease cohorts, the platform achieved 82.1-95.0% accuracy in binary detection of ventriculomegaly or cerebellar abnormalities, with performance similar to that of a senior radiologist. BioTTA provides a practical foundation for standardized fetal brain biometry across heterogeneous imaging settings, supporting scalable prenatal growth assessment and quantitative evaluation of brain abnormalities.
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.
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.
Rapidly evolving LLM architectures and heterogeneous mobile hardware make supporting new models in on-device inference engines a recurring and costly system challenge. Each model may require coordinated changes across conversion, graph construction, operators, and hardware backends. Existing coding agents often miss these cross-layer dependencies and struggle to diagnose failures across host--device execution paths. We present EngineAgent, an agentic system for automating on-device LLM model enablement. Its key insight is to separate reusable engine and hardware knowledge from model-specific implementation. Before enablement, EngineAgent constructs a hierarchy-aware engine index that captures engine abstraction layers and cross-layer dependencies, and pre-validates backend-specific build, execution, and debugging workflows using already supported models. For each new model, a builder uses this knowledge to coordinate cross-layer changes, while an auditor checks implementation completeness and drives iterative debugging through device-side validation. Across 12 tasks spanning three mobile LLMs and four engine--backend pairs on CPUs, GPUs, and NPUs, EngineAgent achieves 75% end-to-end success with either of two underlying LLMs, versus 33.3% for the strongest coding-agent baseline. Success requires compilation, on-device execution, and output-logit agreement with reference implementations. EngineAgent also reduces average per-task time and model-API cost relative to baselines using the same underlying LLMs.
Stochastic diffusion samplers often require substantially more neural function evaluations (NFEs) than deterministic samplers to achieve comparable sample quality. Although higher-order discretization accelerates deterministic sampling, its benefits for stochastic sampling remains unclear. We establish matching upper and worst-case lower bounds showing that stochastic samplers with higher-order approximations remain first-order convergent, even with exact score functions. Thus, increasing the approximation order does not generally improve the convergence order. Motivated by this limitation, we introduce forward-value discretization, a training-free framework that evaluates the data/noise prediction model at a lookahead estimate constructed from previous predictions. For deterministic sampling, the resulting Forward DPMSolver (F-DPMSolver), improves sample quality at low NFE budgets through its DDIM-based variant, despite remaining first-order convergent. For stochastic sampling, we develop Forward DDPM (F-DDPM), which evaluates the model at a cleaner lookahead point adjusted for subsequent noise injection. This estimate reuses both the previous model prediction and the previously injected noise. Both variants admit first-order convergence guarantees with one model evaluation per iteration. Across CIFAR-10, ImageNet, FFHQ, and LSUN, F-DDPM consistently achieves lower FID than the evaluated stochastic baselines at matched NFE budgets. With 1.5 times as many NFEs as deterministic baselines, it achieves competitive sample quality, narrowing the practical efficiency gap between stochastic and deterministic sampling. Code is available at https://github.com/Na-Li66/F-DDPM.
External memory enables LLM agents to improve from experience by distilling trajectories of past tasks into concise lessons for new tasks. However, memory can also lead to memory-induced degradation, where an agent fails on tasks it would solve without memory, often because it follows a lesson that is incorrect, or misapplies or misreads a valid one. This is because lessons are inferred from trajectories labeled only with task success or failure, yet are typically provided without the trajectories they came from, leaving the agent no way to check them. To address this, we introduce WikiTrace, an evidence-linked memory system designed to let agents check each lesson against its original trajectory. Specifically, motivated by the finding that a trajectory often still holds the evidence needed to catch an erroneous lesson, WikiTrace stores each lesson as a concise insight linked to a summary of why its trajectory succeeded or failed and to the raw trajectory itself. Then, on a new task, the agent starts from all insights and, whenever one lacks details needed to act, conflicts with another, or comes from a failed task, descends through its links with terminal tools to fill in missing details or correct the insight before acting. We validate WikiTrace on five agentic benchmarks, showing that it consistently outperforms existing memory approaches and a baseline given the same insights alone, while also reducing memory-induced degradation. Our code is available at https://github.com/yukyeongleee/wikitrace
Evaluating candidate parallelization strategies through distributed training runs makes strategy selection prohibitively expensive at scale. Performance simulation offers a lower cost alternative, yet controlled comparisons among existing simulators remain difficult because they differ in the training stages they model and the workloads on which they are evaluated. We introduce LM-SimBench, a unified framework, profiling dataset, and benchmark for evaluating LLM training simulators. The framework decomposes simulation into functional stages from model compilation to schedule level aggregation, connecting execution representations and cost estimators to measurable evaluation targets. Following this framework, we profile 23 model settings from 10 dense and Mixture-of-Experts (MoE) LLM families on an Ascend Neural Processing Unit (NPU) cluster. The resulting dataset contains 348 feasible configurations that jointly vary five parallelism dimensions and activation recomputation, with six views spanning configuration, rank, operator, event, and kernel granularity. Evaluations of representative training simulators demonstrate how LM-SimBench supports comparisons of prediction accuracy, assessment of configuration selection, and diagnosis of prediction errors. Matched GPU-NPU experiments validate the cross-platform applicability of the collection and evaluation protocol and demonstrate NPU-guided GPU configuration selection.
Large language model agents are increasingly deployed as multi-agent systems rather than as a standalone model operating within a single context. Recent agent harnesses allow the model to instantiate worker agents and to author the executable workflow that coordinates them, shifting the design of system structure from developers to the model at run time. The quality of this organization therefore constitutes a distinct capability, yet existing benchmarks either measure end-to-end task completion or evaluate coordination under structures specified in advance. We formalize this capability as workflow orchestration and introduce WorkflowArena, a benchmark of 145 long-horizon tasks spanning 8 professional domains, from software engineering and formal mathematics to legal analysis and 3D production. Each task specifies an initial workspace and an integrated deliverable composed of interdependent components, which is assessed by a hidden verifier. Agents determine autonomously whether and how to delegate, and all delegation is recorded, enabling joint evaluation of deliverable quality and the organization of the underlying work. Experiments with nine models show that no single configuration performs best across all domains and that systems exhibit distinct delegation strategies: at high effort, Claude models in Claude Code invoke workflows less frequently but with larger worker groups, whereas GPT models in Codex delegate more frequently in smaller units. When the executing model is held fixed on a challenging subset of tasks, workflows authored by Claude Opus 5.5 substantially improve weaker executors, indicating that workflow organization is itself a measurable determinant of task performance.
Multi-criteria decision making over large databases is fundamental to many real-world applications. Interactive regret minimization has emerged recently as a promising approach that combines the controllable output size of top-$k$ queries with the flexibility of skyline queries, without requiring users to explicitly specify a utility function. Instead, the system iteratively presents a small set of options, learns the user's implicit preferences from their selections, and progressively refines its recommendations. However, modern applications such as housing search and financial product selection often involve hundreds of attributes. Existing interactive methods either fail to scale to such high-dimensional data or require excessive user interaction, frequently exceeding $1000$ rounds. We address this limitation with FHDR (Fast High-Dimensional Reduction), a framework that exploits sparse linear preferences to discover key attributes through lightweight interactions. We also propose AttributeSubset, an efficient single-round algorithm that returns a compact recommendation set without further feedback. In our experiments on real datasets, FHDR completes in under $0.03$s and requires fewer than $30$ interaction rounds under the default setting. Extensive experiments on both synthetic and real-world datasets show that FHDR outperforms the best-known baselines by at least one order of magnitude in execution time and by up to several orders of magnitude in interaction cost. These results establish FHDR as a new state of the art for scalable interactive regret minimization.
Weeds compete with crops for light, water, and nutrients, reducing yield and crop quality. Efficient weed detection is essential for site-specific weed management (SSWM). Although deep learning models have been deployed on UAV-based edge systems, a systematic understanding of how different model architectures perform under real-world resource constraints is still lacking. To address this gap, this study proposes a deployment-oriented framework for real-time UAV-based weed detection on resource-constrained edge platforms. The framework integrates UAV data acquisition, model development, and on-device inference, with a focus on balancing detection accuracy and computational efficiency. A diverse set of state-of-the-art object detection models is evaluated, including convolution-based YOLO models (v8-v12) and transformer-based RT-DETR models (v1-v2). Experiments on three edge devices (Jetson Orin Nano, Jetson AGX Xavier, and Jetson AGX Orin) demonstrate clear trade-offs between accuracy and inference latency across models and hardware configurations. Results show that high-capacity models achieve up to 86.9% mAP50 but suffer from high latency, limiting real-time deployment. In contrast, lightweight models achieve 66%-71% mAP50 with significantly lower latency, enabling real-time performance. Among all models, RT-DETRv2-R50-M achieves competitive accuracy (79% mAP50) with improved efficiency, while YOLOv10n provides the fastest inference speed. YOLOv11s and RT-DETRv2-R50-M offer the best balance between accuracy and speed, making them strong candidates for real-time UAV deployment.
In finance, the answer that matters is usually the current one: prices, rates, and disclosures hold only as of a given day, so retrieving up-to-date facts is a core capability of a financial agent. Yet most financial benchmarks overlook timeliness. They fix each answer at annotation time, so the gold goes stale as markets move and, once leaked into training data, rewards recall rather than search and analysis. We introduce LiveSearch, a live benchmark that stores each question as executable verifier code: a typed program of calls to a professional financial data service, followed by a declarative extractor. The question is frozen, while its gold answer is determined by executing the program against professional financial data at evaluation time, and a daily job re-executes it with no human or language model in the loop. A memorized answer is thus likely outdated, and a model must reach the current one through tools and analysis. Construction is equally automatic: a language model only verbalizes programs enumerated from real tool responses and never chooses an answer. Under general web search, six contemporary tool-calling models reach 70.9% to 76.7% answer accuracy. Over three consecutive days, 70% of served questions change their gold value, and even under tolerance-aware matching, a literally reused earlier snapshot fails on a majority of questions within a single day. Per-question tracking uncovers dynamic shifts in correctness masked by aggregate accuracy.
Scenario-based simulation is central to evaluating autonomous-driving policies, but a scenario scripted before execution (i.e., with predetermined, fixed actor routes and speeds) may not preserve its intended interaction once the tested ego policy responds in closed loop. Scripted scenarios cannot adapt to unanticipated ego behavior, while learned traffic actors may remain reactive without reliably realizing a particular scenario intent. We introduce a scenario orchestrator that continuously monitors the joint traffic state and adapts non-ego actors through re-timing, re-routing, and role reassignment for creating a specific scenario intent. At each step, the orchestrator solves a constrained optimization problem that preserves the intended interaction while limiting interventions and maintaining traffic feasibility. We further distinguish scenario realization from policy failure, allowing a test to first be judged based on whether the intended interaction occurred and then by how the ego policy responded. Experiments in the CARLA simulator across multiple scenario families and driving policies show that orchestration realizes targeted interactions more reliably than scripted and learned actor-level baselines, particularly under variations in initial conditions and ego behavior. These results support closed-loop scenario orchestration as a tool for controlled and reproducible policy evaluation.
Zeolites are an important class of porous crystalline materials with broad applications in gas storage, separations, and catalysis. Although hundreds of thousands of experimentally realized and hypothetical zeolite structures are known, identifying materials optimized for a target application remains challenging because property labels are expensive to obtain and available only for limited structure-molecule pairs. While prior work has demonstrated that 3D neural representations can predict zeolite properties effectively, it remains unclear which representation-learning strategies are most effective in the low-data regime relevant to materials discovery. We present ZeoBench, a benchmark for self-supervised learning on dense 3D representations of all-silica zeolites, and evaluate the resulting features on six existing downstream adsorption-property prediction tasks, across a range of training set sizes. Our study compares self-supervised 3D pretraining on volumetric zeolite data, adaptations of image-pretrained 2D ConvNets and vision transformers to 3D via multiview and channel-inflation strategies, and 3D ConvNets pretrained on out-of-domain volumetric datasets, alongside hand-crafted descriptors and supervised baselines. Using our pretraining and evaluation framework, we find self-supervised 3D representations outperform existing methods and achieve state-of-the-art performance across both in-domain label regimes and for the unseen domain of metal organic frameworks. Learned representations consistently outperform hand-crafted descriptors, and convolutional architectures are generally more effective than vision transformers in this setting. These results establish ZeoBench as a practical benchmark for 3D representation learning in zeolites and provide guidance for data-limited adsorption property prediction.
LLM-based coding agents such as Claude Code are guided by repository instruction files (also called context files or agent configuration files) like AGENTS.md and CLAUDE.md, which are read in every agent session. As any documentation artifact, these files can fall behind the code and become stale (e.g., a test folder moves, but the file still points to the old one). In technical debt terms, such a stale instruction is documentation debt, and its interest is the cost it adds to every later agent session. Evaluations that check only task success can miss this cost, and the stale file can persist. In this study, we ask what leads agents to act on a stale instruction, how it affects their behavior and at what cost, and how they recover. We study three real-world stale instructions from active open-source repositories in depth, each a different kind of guidance: a test command, a code location, and a procedure for adding a component. Agents fix 22 maintenance tasks under three versions of each instruction (left stale, corrected to the current path, and removed), with the code and the task held fixed, and we analyze 732 trajectories of two models at the level of individual tool calls. We find that agents follow stale guidance when it is relevant to their tasks, sometimes even when correct information is available elsewhere. On the stale test command, Claude Sonnet 5.5 needed about two extra tool calls per run, a third more than with the corrected command; in the other cases, we could not detect a difference. Although agents generally recovered from the immediate failures, none repaired the underlying instruction file. Correcting the instructions eliminated operations on obsolete paths, while removing them prevented most. Our findings show that task success alone can hide the costs of documentation debt, highlighting the need to evaluate agent behavior beyond final outcomes.