# Research Papers — 2026-09 · bundle page 258 of 266 > Full-text markdown of papers #12851–12900 of 13298 published in 2026-09, newest first. AI-readable mirrors of the full 13298-paper 2026-09 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-09/page/258 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/257.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/259.md) ### Learning Evidence Sufficiency Boundaries for Selective Answering in Grounded Multi-Hop QA Authors: Haruto Sato, Yuki Tanaka, Ren Nakamura, Aoi Kobayashi, Mei Ito Published: 2026-09-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.01687v1 Grounded question answering systems should answer only when the supplied evidence supports the answer. In multi-hop QA, this requirement is difficult because partial evidence can make an unsupported answer appear plausible. We study selective answering through evidence sufficiency boundaries: for the same question, a model should abstain under unsupported or partially supported context, answer when the context first becomes sufficient, and keep the answer stable when redundant evidence is added. We introduce Evidence Sufficiency Boundary Training, a generation-native training framework that constructs ordered evidence chains and supervises the abstain-to-answer transition directly. The method combines level supervision, a boundary flip margin, post-boundary stability, and answer recall protection. We build evidence chains from HotpotQA, 2WikiMultiHopQA, and MuSiQue, then evaluate models with chain metrics, raw QA utility, and unsupported-answer rates on external non-answerable sets. With Qwen2.5-3B-Instruct and LoRA adaptation, Evidence Sufficiency Boundary Training gives the strongest boundary localization among the tested systems, with flip accuracy of 0.807 compared with 0.781 for a token-level abstention baseline. It also achieves the lowest overall unsupported-answer rate on external non-answerable evaluation, 0.095 compared with 0.101 for the same baseline, while retaining competitive raw QA F1. The results show that grounded selective answering improves when training marks the evidence level where refusal should give way to answering. --- ### Meta-ethics and AI: exploring the novel meta-ethical questions in the era of AI Authors: Shang Lu Published: 2026-09-01 Categories: cs.AI, cs.CY arXiv: https://arxiv.org/abs/2609.01685v1 With the development of artificial intelligence (AI), the landscape of meta-ethics, which has largely centred on human ethics, faces pressures that may significantly reconfigure it. In particular, if future AI systems were to exhibit sufficiently integrated capacities for moral reasoning, moral intentionality, and moral reflection, novel meta-ethical questions would arise concerning what I call "AI's own ethics", as distinct from ethical principles merely imposed on AI by human designers. This paper offers a conditional and methodological framework for identifying the questions that would emerge if such AI systems were to arise. On that basis, the paper distinguishes four domains of meta-ethical inquiry in the era of AI: questions about the nature of human ethics from the human perspective; questions about the nature of AI's own ethics from the human perspective; questions about the nature of human ethics from the AI perspective; and questions about the nature of AI's own ethics from the AI perspective. The paper then considers how some existing mainstream meta-ethical theories (such as cognitivism and non-cognitivism, error theory and success theory, relativism, and objective realism) might illuminate these domains, while arguing that many familiar human-centred formulations of those theories may not transfer straightforwardly to AI cases without substantial revision. The overall conclusion is that the emergence of AI's own ethics would place significant pressure on current frameworks and may require substantial refinement, reconstruction, or reconceptualisation. --- ### FORGE: Forward-Only Test-Time Adaptation for Integer-Only Vision Models on Microcontrollers Authors: Muhammad Rehan, Haider Ali, Muhammad Ali Munir, Moaz Amjad Published: 2026-09-01 Categories: cs.CV, cs.AR, cs.LG arXiv: https://arxiv.org/abs/2609.01683v1 Vision models deployed on microcontrollers (MCUs) are quantized to integer-only arithmetic and run in inference-only runtimes that do not carry the machinery backpropagation needs: the standard tool for adapting a model to the distribution shift (sensor noise, blur, lighting) it meets in the field. Existing forward-only test-time adaptation (TTA) methods either run only on server- or edge-GPU-class models (not true microcontroller integer execution), or require the batch-normalization (BN) layers that integer deployment fuses away. We present a forward-only TTA method that operates on deployed, BN-folded, integer-only convolutional networks. The key observation is that fusing BN into the preceding convolution, a mandatory step for integer inference, destroys the statistics that normalization-based adaptation relies on. We restore adaptation by re-normalizing each folded convolution's per-channel output to its clean training statistics, using only forward-pass estimates. The method (i) recovers most of gradient-based TENT's accuracy gain (+20.9 vs. +24.9 points) and matches forward-only BN adaptation, while being the only method that runs on a folded integer-only model; (ii) needs to adapt only 3 of 21 layers (selected without seeing the test corruptions) to recover 93% of the benefit; (iii) survives single-sample streaming with a batch-size-scaled momentum; and (iv) generalizes across three datasets (up to 200 classes) and two architectures. We validate bit-exact int8 convolution execution and deploy on an ESP32-S3, where, measured with a Nordic PPK2 power profiler, the forward-only adaptation (a lightweight fp32 recalibration around the int8 convolutions) costs only 8.3 mJ (6.8% of inference energy) and 21.9 ms on the deployed SIMD-optimized model: forward-only adaptation is cheap on a real microcontroller. --- ### Reinforcement Learning and Rule-Based Peer-to-Peer Pricing in Residential PV-BES Communities Authors: Pablo Benalcazar, Maciej Kalka, Wilian Guamán, Jacek Kamiński Published: 2026-09-01 Categories: cs.LG, cs.CY arXiv: https://arxiv.org/abs/2609.01680v1 This paper compares rule-based and learning-based pricing mechanisms for peer-to-peer (P2P) electricity trading in residential photovoltaic communities. The rule-based benchmarks comprise bill-sharing as an ex post allocation mechanism, the mid-market rate, and supply-demand-ratio pricing. The reinforcement-learning (RL) formulation is implemented through a Deep Q-Network and evaluated under multiplier-based and learnable SDR-shaped pricing, with a fixed-parameter SDR variant as a non-learning control. Performance is assessed through community savings together with complementary financial and operational indicators. In the base PV-only configuration, the rule-based benchmarks outperform the best RL policy. With battery energy storage, evaluated for the RL policies only, community savings under the best RL policy increase from EUR 734.23 to EUR 978.52. Across the learning-based modes and in both configurations, SDR-shaped pricing outperforms the multiplier-based parameterization considered. The results indicate that rule-based pricing remains highly competitive wherever the two families are compared directly, and that storage substantially improves the learning-based outcomes under this accounting, while the distribution of benefits remains heterogeneous across households. --- ### A Survey on Self-Improving Test-Time Intelligence: Feedback-Driven Adapting, Learning, and Scaling at Inference Authors: Shuaicheng Niu, Guohao Chen, Yaofo Chen, Zhiquan Wen, Jinwu Hu, Zeshuai Deng, Deyu Chen, Shuhai Zhang, Renjie Chen, Zihao Lian, Shoukai Xu, Gang Dai et al. Published: 2026-09-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.01679v1 The ability of AI systems to improve their behavior during deployment is becoming increasingly important. As inference moves beyond the static execution of a fixed trained model, a growing body of work studies how models can refine their behavior on the fly by exploiting test-time information and additional computation. These developments have largely evolved along two directions: methods that modify the model's state using test-time signals, and methods that improve predictions through extra inference-time resources such as more sampling and tool use. However, these directions are often studied in separate communities with different terminology, making their connections harder to see. In this survey, we present feedback-driven Test-Time Intelligence (TTI) as a unified perspective for understanding such deployment-time improvement. We use this view to relate test-time adaptation, test-time learning, and test-time scaling, highlighting both their distinctions and their growing overlap in hybrid systems. This unified framework helps connect previously fragmented ideas and provides a clearer conceptual foundation for studying inference-time self-improvement. We review major methodological paradigms, representative applications, and open challenges across vision, language, multimodal learning, generative models, robotics, and healthcare. Our goal is to provide a coherent foundation and research roadmap for the study of self-improving AI systems at test time. --- ### Sim2Signal: Sim-to-Real Benchmarks for Traffic Signal Control Authors: Ferdous Al Rafi, Susrik Mukherjee, Latika Liladhar Dekate, Jennifer Yawa Lavoe, Huaiyuan Yao, Shlok Mohanty, Longchao Da, Xuesong Zhou, Hua Wei Published: 2026-09-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.01676v2 Reinforcement learning achieves strong traffic signal control performance in simulation, yet policies trained in simulators often fail once deployed in the real world, a failure known as the Sim-to-Real gap. When RL is applied to traffic signal control, this gap arises from several sources: sensing, action execution, traffic dynamics, and the control objective. Their relative impact and the reliability of existing Sim-to-Real mitigation methods remain insufficiently understood, and the field lacks a standard benchmark for systematically measuring the gap and evaluating mitigation methods. We present Sim2Signal, a benchmark that decomposes the Sim-to-Real gap into observation, action, transition, and reward gaps, corresponding to mismatches in the four components of the underlying MDP, and induces each gap in isolation under a shared protocol. We evaluate 18 mitigation methods on 2 base controllers, across 33 gap settings and 10 calibrated networks built from 5 real-world locations. We find that direct transfer consistently degrades performance across all four gap sources, but the severity of the degradation does not predict the effectiveness of mitigation. Instead, mitigation effectiveness depends strongly on the network and gap setting: outside the action gap, a method that helps in one case may fail in another. The most effective methods generally estimate what the gap changes, rather than make the policy insensitive through domain randomization or invariant representations. Our code is available at https://github.com/DaRL-LibSignal/Sim2Signal --- ### Random Forest-Informed Cellular Automaton for Large-Scale Wildfire Spread Modelling Authors: Siyu Chen, Esha Saha, Hao Wang Published: 2026-09-01 Categories: stat.AP, cs.LG arXiv: https://arxiv.org/abs/2609.01675v1 Accurate large-scale wildfire spread modelling requires models that capture both the environmental conditions associated with fire occurrence and the local dynamics of fire propagation. We propose a three-stage framework that combines a Random Forest (RF) model with a cellular automaton (CA). First, an RF model trained on the 2021 Canadian fire season estimates daily pixel-level fire-occurrence probabilities. Second, quantile gradient boosting models provide optional spread-rate priors for sensitivity analysis. Third, an RF-informed CA combines the RF probability layer with neighbourhood-driven spread on a 5 km grid. The RF model achieved AUC values of 0.725--0.795 on the 2022--2024 datasets, while the RF-informed CA achieved substantially higher spatial overlap than the evaluated CA-only baselines in the 2023 simulation. A higher-resolution simulation provides an additional qualitative assessment of local spatial errors. These results suggest that combining RF-derived probabilities with local CA spread can improve large-scale wildfire simulations under the tested conditions. --- ### CliffRank: A Dual-Branch Framework for Activity-Cliff Ranking Prediction Authors: Kewei Li, Rongying Zhang, Peiyu Yang, Zhongjian Wang, Qiuchen Zhao, Lan Huang, Fengfeng Zhou Published: 2026-09-01 Categories: cs.LG, cs.AI, q-bio.BM arXiv: https://arxiv.org/abs/2609.01673v1 Activity-cliff ranking remains difficult because local structural changes can cause large activity differences, while high-quality data that resolve the underlying mechanisms remain limited. To use available activity labels more effectively, we combine absolute-activity regression with ranking-consistency learning. CliffRank trains two parallel predictors with mean squared error, a thresholded listwise loss, and Pairwise Preference Consistency (PPC), which aligns relative ordering in the preference-probability space. On three antimicrobial peptide datasets, CliffRank with ESM2-t12 achieved the highest mean Spearman correlation of 0.5393 and mean Recall@50 of 21.4, although the leading method varied across individual datasets. On three small-molecule datasets, CliffRank with PNA, where PPC was activated after 120 epochs, achieved the highest mean Spearman correlation of 0.6890, while its mean Recall@50 of 30.4 matched that of ACANet-PNA. The PPC results also define its practical limits. Asymmetric initialization improved the MolCLR-GIN averages but did not improve every target. For PNA without pretrained weights, delayed PPC improved selected metrics, but no schedule was best for both mean Spearman correlation and mean Recall@50. Future work should evaluate more targets and antimicrobial peptide systems, develop adaptive PPC schedules, and incorporate protein or membrane context when available. --- ### Ranked by the Matcher: A Reproducibility Audit of Knowledge Graph Extraction from Threat Reports Authors: Safayat Bin Hakim, Houbing Herbert Song Published: 2026-09-01 Categories: cs.CR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.01671v2 Security teams and researchers choose knowledge-graph extraction tooling for threat reports on the strength of published triple-F1 scores, yet those scores depend on how predicted triples are matched to gold annotations. We could reimplement the stated matching rule for only five of twelve inspected systems. Re-scoring ten system outputs on shared documents under eight protocols reverses eleven of forty-five pairwise orderings; one fixed prediction set spans 0.16--0.70 F1. On an external, human-adjudicated set, no mechanical matcher---lexical, embedding, or entailment---agrees with the reviewers more than seven times in ten; an LLM judge agrees far more often. To separate component effects from matcher rewards, we build CTIForge, whose deterministic validation layer can vary while extraction is held byte-identical. Across seven tested deployment configurations, no hosted backbone loses precision under validation and every offline one does. Backbone, decoding, and prompting covary across those configurations; on the one backbone we could serve both ways, serving alone reproduces the split. It coincides with a roughly 2.8-fold increase in actions explicitly disputing entity type, consistent with hand-written rules encoding the conventions of the extractor against which they were developed. We release the pipeline, protocol suite, and per-triple audit records. --- ### Private Computation Space: Experience with Trusted Multi-Cluster Federated Learning for Agriculture Authors: Shuangyu Lei, Muhammad Salman Abid, Jacob Belding, Sam Mosher, Manushi B. Trivedi, Shivranjani Baruah, Liam Wickes-Do, Andrew Anderson, Braulio Dumba, Alyssa Whitcraft, Ritvik Sahajpal, Sijin Li et al. Published: 2026-09-01 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2609.01667v1 Artificial Intelligence has shown to help improve agricultural practices, yet adoption remains limited: 69% of U.S. farmers have privacy concerns with sharing their data, and these concerns must be addressed before adoption is widespread. While Federated Learning has been demonstrated to protect privacy at scale for other sectors, deploying a system for agriculture comes with its own set of challenges; the problem necessitates a system that can protect farmer data and identities while preserving model utility, runs on commodity hardware, and is resilient to fragile rural infrastructure. To address these concerns, we introduce the Private Computation Space (PCS), a deployed, open-source Machine Learning system to provision and process farmer data securely. We design a system tailored to an agricultural setting, with multi-cluster orchestration for reliability in rural areas with asynchronous Federated Learning (FL), Differential Privacy (DP), and Trusted Execution Environments (TEEs), to allow farms to participate in the framework while keeping their data private. We evaluate the system on two deployed workloads: monitoring nitrogen with living plant sensors in NY for six months and predicting evapotranspiration from weather stations in CA for ten months. Our evaluation finds a Dice Similarity Coefficient (DSC) of 0.71 and $R^2$ accuracy of 0.84 for the respective workloads, improving the worst single-site model accuracy by 22.4% and 9.1%, respectively, while preserving privacy. --- ### Uncovering Understanding-Generation Synergy in Native Unified Multimodal Models: From Representation, Task to System Authors: Penghao Wu, Haiwen Diao, Weichen Fan, Lewei Lu, Dahua Lin, Ziwei Liu Published: 2026-09-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.01607v1 While unified multimodal models (UMMs) jointly perform visual understanding and generation within a single model, functional unification does not guarantee learning synergy: the two objectives may reinforce each other, compete for capacity, or merely coexist. We investigate their relationship at the representation, task, and system levels in a controlled, structurally native setting without pretrained vision priors. At the representation level, we find that each objective provides useful signal to the other: generation enriches the visual features learned for understanding, while understanding strengthens vision--language alignment for generation. However, when both objectives are forced through the same computation path, one tends to dominate. A task-decoupled architecture that specializes conflicting visual computation while preserving semantic interaction avoids this asymmetric degradation. At the task level, through three case studies, we find positive bidirectional transfer when understanding and generation tasks rely on shared knowledge. At the system level, we show that an end-to-end UMM outperforms a matched planner--executor pipeline on complex tasks that explicitly require both image understanding and generation. Together, these results show that the value of UMMs extends beyond a unified interface: appropriate specialization, shared task knowledge, and end-to-end optimization can turn coexistence into synergy. --- ### Beyond Scores: Understanding LLM-as-a-Judge Mechanisms in Summarization Evaluation Authors: Himil Vasava, Ming Jiang Published: 2026-09-01 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.01604v1 LLM-based evaluators of natural language generation (NLG) quality are widely deployed as scoring tools and as automated training signals, yet the internal procedure by which they assign a rating remains poorly understood. We investigate this procedure mechanistically through an eight-attack perturbation taxonomy across the Readability and Adequacy dimensions of NLG quality, a generation pipeline that produces paired clean and corrupt summaries with controlled error intensity and explicit token-level modification maps, and a four-experiment battery of causal tracing, logit-lens vocabulary projection, and attention-head knockout applied to Themis (Llama-3-8B) and Prometheus (Mistral-7B). Both evaluators implement a structured, coherent evaluation pipeline operating in two stages: below layer 15, attention performs local error comparison and routes the result to the final input position; above it, the MLP cascade integrates the signal and writes the rating, with the decision crystallizing in the residual stream at a sharp late layer (L = 26 on Themis, L = 25 on Prometheus). Furthermore, a base-model control at the same scale (Llama-3-8B) reproduces the routing architecture and crystallization but not the stage separation, isolating the two mechanisms that fine-tuning specifically installs, suppression of below-L15 MLP contribution at the last position and a two-layer advance of the crystallization depth, indicating that fine-tuning sculpts an existing substrate rather than building the pipeline from scratch. We release the source code and data at https://github.com/himil-v/judge-mech --- ### Efficient SWE Agent Benchmarking via Trajectory-Aware Evaluation Authors: Kefeng Duan, Dewu Zheng, Yanlin Wang, Xiwen Wang, Ensheng Shi, Xilin Liu, Yuchi Ma, Jiachi Chen, Mingwei Liu, Zibin Zheng Published: 2026-09-01 Categories: cs.SE, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.01603v1 Evaluating software engineering agents on realistic benchmarks is costly, since each task may require multi-step code exploration, modification, and test execution. Existing efficient evaluation methods select representative subsets to estimate full-benchmark performance, but are largely result-only: they fit historical pass/fail response matrices or static task semantics, discarding how agents solve problems. We propose PTA-IRT, a Privileged Trajectory-Aware Item Response Theory framework that fuses process and outcome signals. Historical execution trajectories supply process-level evidence beyond pass/fail, such as explored context, attempted edits, and solving paths, which PTA-IRT uses as privileged information for calibration subset selection and ability estimation. Under low calibration budgets, PTA-IRT consistently outperforms prior IRT baselines on score and ranking recovery across four SWE benchmarks. Code and data are publicly available at https://github.com/DeepSoftwareAnalytics/PTA-IRT. --- ### Adaptive Critical Token-Aware Retrieval for Repository-Level Code Generation Authors: Kefeng Duan, Dewu Zheng, Yanlin Wang, Terry Yue Zhuo, Mingwei Liu, Jianxing Yu, Jiachi Chen, Ensheng Shi, Xilin Liu, Yuchi Ma, Zibin Zheng Published: 2026-09-01 Categories: cs.SE, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.01601v1 The repository-level code generation task requires synthesizing code that satisfies task requirements while remaining consistent with the target repository context. Since real-world repositories often exceed the input length limits of LLMs, existing approaches commonly adopt retrieval-augmented generation (RAG) to provide repository-specific context. Despite improving repository-context retrieval, existing methods typically provide context as task-level support, without explicitly identifying the critical tokens that require fine-grained repository context during generation. During the autoregressive generation process of LLMs, errors often concentrate at a small number of decisive positions: once such tokens are generated incorrectly, subsequent code may follow an incorrect semantic path and eventually lead to functional failure. We refer to these positions as "critical tokens". In this paper, we propose ACToR, an adaptive critical token-aware retrieval framework for repository-level code generation. ACToR identifies critical tokens during generation and triggers targeted retrieval on demand to provide repository context at these decisive positions. In addition, we design a position-aware weighting method for dense retrievers to prioritize context that is more informative for generation. We evaluate ACToR on two representative repository-level benchmarks, RepoExec and CoderEval. Experimental results show that ACToR consistently outperforms state-of-the-art methods, achieving relative improvements of 8.4% on RepoExec and 15.4% on CoderEval. Beyond performance gains, we systematically quantify the impact of critical tokens, revealing their central role in major generation failures and highlighting the necessity of targeted retrieval strategies. We provide the code and data at https://github.com/DeepSoftwareAnalytics/ACToR. --- ### CordisBench: Can Language Models Reason About Component Lifecycles in Dynamic Agent Harnesses? Authors: Damien Sileo, Dimitri Kachler Published: 2026-09-01 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.01600v1 Dynamic agent harnesses let language models change the software that shapes their own execution. This flexibility brings a new reasoning burden: a local plugin change can propagate through dependencies and cleanup. We introduce CordisBench, a 1,200-question benchmark of this lifecycle reasoning. It combines a controlled formal setting with programs executed against Cordis, a runtime that manages component dependencies and cleanup, and asks models to identify affected components, predict state after a specified teardown order, determine which conditions hold under all or some orders, and choose reconfigurations that succeed when executed. Across these tasks, we evaluate three efficiency-oriented models at low reasoning effort with 2, 4, 8, 16, 24, or 32 relevant interactions, using deterministic task-specific scoring. Models usually handle small systems well but grow less reliable as more interactions become relevant, especially when predicting final state and when reasoning across teardown orders. Additional inference effort recovers marked gains for some models. The cost is nontrivial: on our 16-interaction subset, GPT-5.6 Luna uses nearly 3,000 reasoning tokens per question at medium effort. For these controlled instances, that cost is avoidable: an independent finite reference semantics agrees with Cordis execution on every observation and action outcome used for scoring across all 528 executable questions. --- ### UI-VISA: U-Net Initialized Vascular Image Segmentation Architecture Authors: Asees Kaur, Suzanne S. Sindi, Erica M. Rutter Published: 2026-09-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.01598v1 Accurate segmentation of vascular structures in digital subtraction angiography (DSA) images remains challenging due to the thin, elongated, and branching nature of blood vessels. Pixel-wise deep learning approaches such as U-Net achieve strong general-purpose segmentation performance but often produce fragmented or discontinuous predictions in fine vascular regions, since they do not explicitly enforce structural connectivity. Region growing algorithms preserve spatial context and topological continuity, but are highly sensitive to seed point initialization and can be computationally expensive. We propose UI-VISA (U-Net Initialized Vascular Image Segmentation Architecture), a hybrid pipeline that combines the complementary strengths of both approaches. UI-VISA uses U-Net's foreground predictions as informed seed points for a CNN-guided region growing algorithm, which then iteratively refines the segmentation by enforcing local connectivity and recovering fine vessel details that U-Net alone tends to miss or over-predict. We evaluate UI-VISA against standalone U-Net and a prior region-growing-based method (VISA) using 5-fold cross-validation on 26 DSA images. UI-VISA achieves the highest mean Dice and clDice scores across folds, and a paired Wilcoxon signed-rank test shows the improvement in clDice is statistically significant ($p=0.023$), consistent with the method's design goal of preserving vascular connectivity, while the improvement in Dice does not reach significance ($p=0.104$). --- ### The Rise of Verbal Reinforcement Learning Authors: Kshitij Tayal, Arun Sharma, Genta Indra Winata, Anirban Das, Sambit Sahu Published: 2026-09-01 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.01597v1 Natural language is emerging as a primary feedback channel for improving language agents, capable of conveying intent, preferences, and causal structure in forms interpretable by both humans and modern language models. We call this paradigm Verbal Reinforcement Learning (VRL) and offer the first unified account of it. We organize the field around a single axis, \textit{when} verbal feedback takes effect in an agent's lifecycle and \textit{what} it modifies, yielding three pillars: (1) \textbf{Language as Grounding Signal}, where language defines the task itself by specifying goals, states, and reward structures; (2) \textbf{Language as Deliberative Feedback}, where natural language guides reasoning at test time without the need to update model parameters; (3) \textbf{Language as Learning Signal}, where language-based feedback shapes model parameters through training. Within each pillar, we synthesize representative work, distinguish key subcategories of approaches, and outline the distinct role language plays in shaping agent behavior. Together, this taxonomy shows how verbal reinforcement is reshaping agent development, while also defining the challenges and opportunities for building more capable and aligned agents. --- ### Facet-0: A Robotic Foundation Model for Contact-Rich Precise Manipulation Authors: Haoyuan Deng, Haichao Liu, Wenkai Guo, Yuan Ling, Zaijia Yang, Yuanjiang Xue, Haosheng Sun, Liangzi Wang, Ziwei Wang Published: 2026-09-01 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2609.01596v1 Real-world robotic assembly at sub-millimeter tolerances demands spatial precision, compliant interaction, and robustness to contact failures. We present Facet-0, a robotic foundation model that predicts and values the contact consequences of its actions. Facet-0 unifies multimodal representation learning and reinforcement learning (RL) post-training around a joint action-wrench proposal: a causal wrench history is aligned with vision-language semantics and kinematic state, and flow matching generates each action chunk together with the future wrist-wrench profile it is expected to induce. Deployment rollouts train a distributional Action-Wrench Critic to distinguish motions with similar task progress but different contact outcomes, while phase-aware rewards and contact-selective credit concentrate policy improvement on decisive interactions. To accommodate part-specific dynamics, a lightweight bounded actor reuses the frozen representation for on-robot adaptation; RL remains defined over executable Cartesian actions, while an auxiliary wrench head preserves predictive, non-commanded action-contact coupling. Trained on ManuFacet-1K, a 1,000-hour force-synchronized corpus spanning three embodiments and multiple manufacturing cells, the bounded task-adapted system reaches 82% mean success on five sub-millimeter computer-assembly tasks, compared with 15% for the strongest baseline, with 0.5 mm placement accuracy and 50 ms command latency. --- ### Mechanism Design for Alignment and Control Authors: Dirk Bergemann, Andrew Koh, Stephen Morris Published: 2026-09-01 Categories: econ.TH, cs.AI, cs.GT arXiv: https://arxiv.org/abs/2609.01595v1 We develop a framework for mechanism design with AI agents whose alignment (preferences) and capabilities (feasible actions and information) are unknown. We want such agents to act on our behalf so mechanisms must incentivize both honesty and obedience. A one-sided imitation structure---capabilities can be concealed but not counterfeited---yields a revelation principle, a characterization of implementable policies via nested cyclical monotonicity, and conditions under which eliciting higher-order beliefs can discipline multiple agents. We apply our framework to stylized examples of (i) sandbagging in which a more capable agent pretends to be less capable; (ii) an alignment--interpretability trade-off, where the two are substitutes in the instrument but complements in value; (iii) discipline via peer scoring; (iv) coupling rewards to induce competition among multiple agents; and (v) scalable oversight and reward shaping. --- ### StudentSim: Training LLM-based Student Simulators Authors: Ke Yang, Chenglong Wang, Michel Galley, Chandan Singh, Jeevana Priya Inala, ChengXiang Zhai, Jianfeng Gao Published: 2026-09-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.01591v1 AI tutors are most useful when they adapt to each student's strengths, weaknesses, and preferred guidance, but evidence about which guidance works for which student is sparse, slow, and costly to collect from real learners. Student simulators can provide this signal as a proxy, yet existing approaches are limited: state-tracking models fit student behavior but struggle to process explanations or corrections, while LLM role-play follows guidance fluently but does not reliably match the competence of the student being imitated. We present StudentSim, a training framework that turns sparse per-student data into individualized simulators through pooled training followed by per-student specialization. The resulting simulators both mirror a student's own responses and update them under tutor guidance. We also introduce StudentSimEval, a standardized protocol covering 60 students across chess, second-language English writing, and mathematics, using public learner datasets with de-identified records shared for research. StudentSimEval measures behavioral fidelity (F), or how well a simulator matches a student's responses, and guidance responsiveness (R), or how readily it updates under tutor guidance, with all methods fit and evaluated on the same records. Across all three domains, StudentSim outperforms GPT-5.4 on both metrics. In chess, StudentSim reaches F=0.51 and R=0.91, compared with 0.23 and 0.72 for GPT-5.4 and 0.45 and 0.27 for Maia2. As a proof of concept, using StudentSim as a reward model for tutor reinforcement learning produces a chess tutor that expert humans rate as more accurate, better-guided, and more personalized than a no-RL baseline and a tutor trained against a GPT-5.4 simulator reward. Code is available at https://github.com/microsoft/StudentSim. --- ### Designing Proactive Thought Partners for Writing Authors: Chao Zhang, Abe Davis, Chih-Wei Chen, Chin-Chia Hsu Published: 2026-09-01 Categories: cs.HC, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.01588v1 Writing involves diverse cognitive activities, from ideation to revision, and writers' needs vary across individuals and moments. Proactive AI promises to provide the right support at the right time, yet existing proactive tools largely focus on generic textual assistance, such as autocomplete. This paper studies the design space of proactive thought partners: AI agents that proactively offer customizable, higher-level cognitive support during writing. We instantiated this concept in a technology probe and deployed it with 16 participants for one week. The probe allows users to create partners by configuring their roles and proactivity. As users write, relevant partners take the initiative at appropriate moments to offer suggestions. Our findings show that participants configured proactive support through prospective planning, used suggestions for both idea generation and self-monitoring, and valued lightweight visual representations alongside non-directive rhetorical framing for non-intrusive interventions. We derive implications for designing proactive writing assistants around customization, timing, engagement, and representation. --- ### The Structure of Quantization Damage in LLMs: Why the Next Bit Should Be Spent Globally Authors: Jundong Hu, Shekar Ramachandran Published: 2026-09-01 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2609.01587v1 Post-training quantization (PTQ) is widely used to reduce the cost of serving large language models (LLMs), but its accuracy cost is uneven and is often tuned per model. We study where quantization damage occurs and how to allocate a small additional precision budget. Using causal mixed-precision intervention as ground truth (raise each layer to 8-bit in turn and measure the accuracy it recovers) across 9 open-weight models in 4 architecture families, we test 3 intuitive hypotheses: that quantization damage lives in task circuits, where the model computes, or in weight statistics. None of them predicts which layers benefit from restored precision. Recovery is instead diffuse: for 8 of 9 models, recovering 75% of the gap takes roughly half the layers; the lone exception, Qwen3-8B, is sharply concentrated. At a matched precision budget, spending it globally on finer quantization granularity beats locally repairing the most recoverable layers for all 8 group-128-compatible models (all but OpenLLaMA, whose width rules out group-128), by 21-52 points, including the concentrated Qwen3-8B. We report 2 secondary findings: the residual is budget-limited (8-bit is near-lossless in our evaluation across RTN, GPTQ, and AWQ), and the location of peak recovery correlates with architecture within a family, though not across families. Within this budget setting, global granularity is a better default than selectively protecting critical layers. More broadly, cheap signals that correlate with quantization damage do not necessarily identify where restoring precision improves accuracy; this must be tested with causal intervention. --- ### A Benchmark for Vehicle Attribute Classification in Cross-Domain Surveillance Scenarios Authors: Sergio M. Silva, Otavio T. Remer, Gabriel E. Lima, Lucas Wojcik, Rayson Laroca, David Menotti Published: 2026-09-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.01584v1 Vehicle attribute analysis is a key component of Intelligent Transportation Systems (ITS), supporting applications such as vehicle identification, traffic monitoring, and forensic investigation. However, models trained under controlled conditions often degrade in real surveillance scenarios due to changes in viewpoint, occlusion, illumination, and sensor characteristics. This paper introduces Unconstrained Vehicle Identification Benchmark (UVIB), a benchmark for evaluating three operational vehicle-analysis tasks: front/rear orientation, occlusion-related suitability for Vehicle Make and Model Recognition (VMMR), and color clarity. The benchmark contains 84,835 vehicle images from seven public Brazilian datasets, grouped into surveillance and general acquisition domains, with unified binary annotations that were not jointly available in the original sources. Four representative architectures, EfficientNetV2-S, ResNet-50, ViT/B-16, and YOLO11s-cls, are evaluated under mixed-domain, cross-domain, and cross-dataset protocols. The results show that domain shift has a stronger impact than architecture choice, with substantial degradation in cross-domain settings, especially for VMMR suitability and color clarity. While orientation generalizes more reliably, VMMR suitability remains affected by class imbalance and ambiguous occlusions, and color clarity is highly sensitive to illumination and sensor modality. These findings highlight the need for benchmarks and evaluation protocols that explicitly measure operational robustness beyond standard in-domain accuracy. The proposed benchmark is publicly available at https://github.com/UFPR-IPASP-PR/uvib-vehicle-attributes/. --- ### SpatialGuard: Harness-Guided Verifiable Spatial Reasoning for Text-to-Image Generation Authors: Ziyun Qian, Zizhi Chen, Yizhou Liu, Mingyang Sun, Dingkang Yang, Lihua Zhang Published: 2026-09-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.01582v1 Complex 3D spatial text to image generation requires models to convert natural language into stable visual geometry, not merely semantic appearance. Existing prompt-driven or layout-conditioned methods improve controllability, but often lack an optimizable and verifiable spatial intermediary before visual sampling. As a result, object relations, occlusion, visibility, and camera constraints can decay during multi-round generation. This paper presents SpatialGuard, a structured layout-guided framework for complex 3D spatial text-to-image generation. SpatialGuard parses prompts into image synthesis-oriented 3D layouts through a Spatial Layout Architect, realizes them as visual conditions and candidate images through a Visual Realizer, and uses a Visual Alignment Critic to validate consistency among prompt, layout, and image. To keep constraints stable across iterations, SpatialGuard introduces a Layout Harness that organizes rule constraints, tool invocation, shared knowledge, and feedback loops around the editable layout state. This design turns complex spatial generation from implicit prompt following into a verifiable process of planning, realization, validation, and repair. Comprehensive experiments show that SpatialGuard achieves state-of-the-art performance in complex 3D spatial layout generation and improves spatial faithfulness over existing text-to-image and layout control baselines. --- ### Simultaneous Pointwise Majorization for Mixed Tail Processes with Applications in Gaussian Chaos and Ergodic Diffusions Authors: Haichen Hu, David Simchi-Levi Published: 2026-09-01 Categories: math.PR, math.ST, stat.ML arXiv: https://arxiv.org/abs/2609.01576v2 Classical chaining controls an indexed stochastic process through a single worst-case bound and can therefore obscure substantial variation across the index set. We develop the first simultaneous pointwise majorization theory for Banach-valued processes with finite-metric mixed-tail increments. Suppose that an anchored process $(Z_t)_{t\in T}$ satisfies, for some integer $m\ge1$, pseudo-metrics $d_1,\ldots,d_m$, and orders $α_1,\ldots,α_m>0$, \begin{align*} \mathbb{P}\{\|Z_t-Z_s\|>\sum_{j=1}^m u^{1/α_j}d_j(s,t)\}\le 2e^{-u},s,t\in T. \end{align*} For ambient priors $μ_1,\ldots,μ_m$, let $v_j(t):=d_j(t,t_0), Φ_j(t):=\int_0^{4v_j(t)}(\log\frac{1}{μ_j(B_{d_j}(t,r))})^{1/α_j}dr$. We prove that, $\forall δ\in(0,1)$, with probability at least $1-δ$, simultaneously for all $t\in T$, \begin{align*} \|Z_t\|\le C_{m,\boldsymbolα}\sum_{j=1}^m\{Φ_j(t)+v_j(t)(\log(e/δ))^{1/α_j}\}. \end{align*} Here $\boldsymbolα:=(α_1,\ldots,α_m)$ and $C_{m,\boldsymbolα}$ depend only on $m$ and these tail orders. The result subsumes single-metric sub-Weibull processes of every positive order as the case $m=1$. In the Gaussian setting, it sharpens the pointwise upper bound of \citet{xu2026} by eliminating the logarithmic terms generated by dyadic peeling. The proof retains the index-wise costs of measure-generated admissible chains and synchronizes the regimes through a nested common refinement. Finally, we apply our theorems to stationary diffusion empirical processes and decoupled Gaussian chaos to obtain simultaneous pointwise envelope bounds, which can further be applied to other statistics problems. --- ### Pointwise Majorization for sub-Weibull and Mixed Tail Processes with Applications in Quadratic Chaos and Ergodic Diffusions Authors: Haichen Hu, David Simchi-Levi Published: 2026-09-01 Categories: math.PR, math.ST, stat.ML arXiv: https://arxiv.org/abs/2609.01576v1 Classical chaining controls an indexed stochastic process through a single worst-case bound, which can obscure substantial variation across the index set. We establish the first simultaneous pointwise majorization theory for Banach-valued processes with sub-Weibull or two-metric mixed-tail increments. For an anchored sub-Weibull process on a separable index space, write $v(t):=d(t,t_0)$. Given a reference measure $μ$, the envelope at $t$ is governed by the pointwise Fernique-Talagrand functional of order $α$, $Φ_{μ,d}^{(α)}(t):=\int_0^{4v(t)}(\log\frac{1}{μ(B_d(t,r))})^{1/α}dr$. $\forall δ\in(0,1)$, we obtain that $$ \mathbb{P}(\|Z_t\|\lesssim\{Φ_{μ,d}^{(α)}(t)+v(t)(\log(e/δ))^{1/α}\},\forall t)\ge 1-δ. $$ Our bound is determined by the pointwise complexity $Φ_{μ,d}^{(α)}$ rather than a global quantity. The result holds for every $α>0$ and does not involve dyadic logarithmic terms from peeling. For mixed tail processes, with fixed measures $μ_1,μ_2$ and $v_j(t):=d_j(t,t_0)$, $Φ_j(t):=\int_0^{4v_j(t)}(\log\frac{1}{μ_j(B_{d_j}(t,r))})^{1/α_j}dr, j=1,2$, for any $δ\in(0,1)$, we show that $$\mathbb{P}(\|Z_t\|\lesssim\sum_{j=1}^2\{Φ_j(t)+v_j(t)(\log\frac{e}δ)^{1/α_j}\},\forall t)\ge 1-δ.$$ Although the two regimes are coupled in the mixed tail condition, each retains its own pseudo-metric, reference measure, pointwise Fernique-Talagrand functional, and tail exponent. The proof tracks the index-wise costs of measure-generated admissible chains and synchronizes them through a nested common refinement. For applications, we derive matrix-specific bounds for centered quadratic chaos under pseudo-metrics induced by the operator and Frobenius norms, and observable-specific finite-time bounds for diffusion empirical processes. --- ### Closing Cost-Quality Gap in Document VLMs: Difficulty-Aware Data Curation and Quality-Adjusted Deployment Economics Authors: Maksim Evdokimov, Matvey Ivanov, Dmitrii Tsiupin, Olga Tsymboi, Anatolii Potapov, Aleksandr Ivanov Published: 2026-09-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.01575v1 Extracting structured fields from hundreds of millions of documents annually remains costly in regulated industries: bespoke OCR cascades cover only a fraction of workflows, privacy rules preclude external models, and existing open-source VLMs that clear quality thresholds cost more to serve than human annotation. We present a deployed document-understanding system built on a Mixture-of-Experts VLM (35B total, 3B active), fine-tuned on in-house production data mixed with open-domain documents curated by a Difficulty-Aware pipeline for layout diversity, fact-extractability, and cross-model consistency. Fitting on a single H100 and serving heterogeneous workflows via prompting, the model leads all deployable (non-reasoning) baselines up to an order of magnitude larger. A quality-adjusted cost analysis, with confirmation and correction costs calibrated from production telemetry, shows it reduces expected costs by over 80% against the human baseline and by more than 50% against the best competing open-source model, while larger baselines remain economically unviable. --- ### Scaling Near-Optimal SFT-RL Annotation Budget Allocation from Small to Large LLMs Authors: Jingtan Wang, Arun Verma, Xiaoqiang Lin, Zhengyuan Liu, Nancy F. Chen, Daniela Rus, Bryan Kian Hsiang Low Published: 2026-09-01 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.01573v1 How to divide a fixed annotation budget between supervised fine-tuning (SFT) and reinforcement learning (RL) during LLM post-training remains an open problem. Existing work characterizes only broad trends (e.g., SFT dominates in low-data regimes), lacks a principled allocation framework, and does not examine whether the optimal ratio transfers across model sizes. We frame this problem in terms of near-optimality: rather than seeking a single optimal SFT-RL ratio, we characterize the near-optimal region, the set of allocations within a specified tolerance of peak performance. Empirically, this region is wide even for small tolerances (2-10%), widens with model scale, and transfers reliably from small proxy models to large target models. This yields a practical strategy: small proxy-model experiments suffice to identify a transferable near-optimal region, eliminating the need for exhaustive large-scale search. Our results hold consistently across tasks, model families, and both preference-based off-policy and reward-supervision on-policy RL methods. We further analyze how the asymmetry in annotation costs between SFT and RL data shifts the near-optimal region. --- ### From Production Traffic to Post-Training: Building a Self-Hosted LLM That Covers the Corporate Request Mix Authors: Olga Tsymboi, Dmitrii Stoianov, Ramil Latypov, Danil Taranets, Daniil Dryabin, Mikhail Gashkov, Viktor Zelenkovskiy, Aleksandr Fida, Gleb Alektorov, Nikita Gulyakov, Arthur Babkin, Aleksandr Medvedev et al. Published: 2026-09-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.01572v1 Data-residency constraints force enterprises to self-host LLMs, but continuous adoption of newer models without decommissioning their predecessors expands the serving fleet, fragmenting a finite GPU pool. We consolidate traffic from over 200 internal applications onto a single model by closing quality gaps identified through production error analysis along three axes: instruction following, function-calling, and internal task distribution. Quality is tracked by offline benchmarks stratified to production traffic and scored by deterministic verifiers or calibrated LLM judges. Rather than optimising all objectives jointly, which introduces cross-domain reward interference, we train a separate GRPO expert per axis and merge them via two-stage SLERP. Each expert's reward exposes a distinct failure mode, namely semantic collapse, over-calling, and verbosity hacking, each requiring a domain-specific fix. In non-reasoning mode the recipe surpasses a ${\sim}7\times$ larger by total parameters baseline on the in-house Arena with 69.6 to 65.8, instruction following with 0.85 to 0.83, and function-calling with 0.79 to 0.77, while lifting general dialogue benchmarks. The model absorbs 50% of platform traffic, 116M requests per month, at a fraction of the serving cost. --- ### Selective Agent Guidance via Entropy: Learning Autonomous Policies from Imperfect VLM Teachers Authors: Giovanni Bonetta, Matteo Merler, Davide Zago, Rossella Cancelliere, Bernardo Magnini Published: 2026-09-01 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.01567v2 Vision-Language Models (VLMs) provide useful priors for interactive decision-making, but using them directly as policies is expensive and brittle: they must be queried at every step, do not improve from environment interaction, and can repeat systematic errors. We study how to learn a cheap autonomous policy from an online, expensive, and imperfect but informative VLM teacher. We propose SAGE (Selective Agent Guidance via Entropy), a framework that queries a VLM only when the learner is uncertain, executes the suggested action during training, and distills guidance into a lightweight Reinforcement Learning (RL) policy. Because VLM advice is not always reliable, SAGE can weight teacher-action distillation using environment-derived advantages rather than treating all suggestions as equally useful. Across sparse-reward visual reasoning and navigation tasks, SAGE learns policies that act without VLM guidance at evaluation time and improves over unguided RL in several environments, including settings where the learned policy exceeds its VLM teacher. The results show that selective guidance is most beneficial when the VLM can help the agent discover high-reward trajectories, and less useful when unguided exploration already succeeds or teacher actions do not lead to informative experience. SAGE also reduces VLM usage by prompting the teacher only on a fraction of training steps and requiring no VLM calls at deployment. Overall, our results suggest that VLMs don't need to be used as fixed policies to be useful; they can instead act as temporary, imperfect sources of guidance whose value is tested and internalized through interaction. --- ### Selective Agent Guidance via Entropy: Learning Autonomous Policies from Imperfect VLM Teachers Authors: Matteo Merler, Giovanni Bonetta, Davide Zago, Rossella Cancelliere, Bernardo Magnini Published: 2026-09-01 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.01567v1 Vision-Language Models (VLMs) provide useful priors for interactive decision-making, but using them directly as policies is expensive and brittle: they must be queried at every step, do not improve from environment interaction, and can repeat systematic errors. We study how to learn a cheap autonomous policy from an online, expensive, and imperfect but informative VLM teacher. We propose SAGE (Selective Agent Guidance via Entropy), a framework that queries a VLM only when the learner is uncertain, executes the suggested action during training, and distills guidance into a lightweight Reinforcement Learning (RL) policy. Because VLM advice is not always reliable, SAGE can weight teacher-action distillation using environment-derived advantages rather than treating all suggestions as equally useful. Across sparse-reward visual reasoning and navigation tasks, SAGE learns policies that act without VLM guidance at evaluation time and improves over unguided RL in several environments, including settings where the learned policy exceeds its VLM teacher. The results show that selective guidance is most beneficial when the VLM can help the agent discover high-reward trajectories, and less useful when unguided exploration already succeeds or teacher actions do not lead to informative experience. SAGE also reduces VLM usage by prompting the teacher only on a fraction of training steps and requiring no VLM calls at deployment. Overall, our results suggest that VLMs don't need to be used as fixed policies to be useful; they can instead act as temporary, imperfect sources of guidance whose value is tested and internalized through interaction. --- ### From Confusion to Clarity: Confusion-Aware Retrieval and Knowledge Injection for Text Classification Authors: Manish Gupta, Chaitanya Giri, Jayasimha Talur Published: 2026-09-01 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.01564v1 Large language models (LLMs) struggle to classify text into taxonomies with many semantically similar labels, as the distinctions are domain-specific and not captured by pre-training. To handle large label spaces, a common approach retrieves top-$K$ candidate labels by embedding similarity and prompt the LLM to choose among them. However, top-$K$ retrieval reduces the number of candidates but does not help the model tell similar ones apart. When two similar labels both appear as candidates, the model lacks the signal to choose correctly between them. We propose a framework that (1) identifies which label pairs the model struggles to distinguish, (2) expands the candidate set to include confusable labels, and (3) generates targeted rules to differentiate between similar candidates. The framework requires no fine-tuning, and the generated rules transfer to smaller, cheaper models. On three benchmarks (WOS, Flipkart, LEDGAR), our approach improves Macro F1 by up to 10.0pp over retrieval baselines, with smaller models (2B--20B) gaining up to 11.5pp via cross-model transfer. --- ### A systematic Approach to constructing a Chance-and-Risk Matrix for Semiconductor Supply Chains Authors: Ema Salkić, Alexander Fichtl, Philipp Ulrich, Hans Ehm, Marta Bonik, Georg Groh Published: 2026-09-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.01563v1 Semiconductor supply chains face escalating risks from geopolitical tensions, geographic concentration, and rapid technological shifts, yet no scalable system continuously extracts, structures, and prioritizes risk intelligence from public corporate disclosures. We present an end-to-end pipeline that retrieves corporate documents for semiconductor companies and uses large language models (LLMs) to extract the risks and opportunities they describe. It organizes these into a knowledge graph linking each item to its category, sources, and related events, then merges duplicates and ranks them with a three-layer mechanism combining an algorithmic formula, an LLM relevance adjustment, and expert validation. Applied to five companies across the value chain, the pipeline produces 76,207 scored items, of which an independent check finds 92.6% valid. The automated rankings match expert judgment at an average Spearman correlation of 0.55 for risks and 0.72 for opportunities, and the resulting matrices identify trade restrictions as the dominant cross-company risk. --- ### H3-World: Turning Language Understanding into World Control Authors: Danze Chen, Zeqing Wang, Ziyue Lin, Xingyi Yang, Yeying Jin Published: 2026-09-01 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.01560v1 We present H3-World, an efficient framework that turns the 33B MiniMax-H3 video generator into an interactive world model. Our key finding is that, as large video generators become more capable, language is emerging as a natural interface for control. MiniMax-H3, for example, already supports zero-shot control of character behavior and camera motion through natural-language instructions. Building on this, H3-World turns this coarse language interface into precise, temporally grounded world control, without introducing dedicated action modules. Specifically, we represent each action as a structured combination of character and camera instructions, and align them with the corresponding temporal video latents. To make the control temporally precise, we further introduce temporal attention routing, which restricts each instruction to its intended time interval and reduces control leakage across actions. Importantly, H3-World directly reuses the semantic representations learned during large-scale video pretraining and requires only lightweight adaptation. With only 8,000 gameplay samples, 10,000 LoRA optimization steps, and 0.199% trainable parameters, H3-World achieves effective character and camera control while preserving strong generation quality. It also generalizes to unseen scenarios. These results show that the control capabilities emerging in large video generators can be efficiently transformed into interactive world control. --- ### Gradient-Update Mismatch: Rethinking Conflict-Free Training of Physics-Informed Neural Networks Authors: Jing Xiao, Xinhai Chen, Qinglin Wang, Menghan Jia, Zhiquan Lai, Dongsheng Li, Jie Liu, Tiejun Li Published: 2026-09-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.01558v1 Training Physics-Informed Neural Networks (PINNs) requires jointly optimizing physics residual and initial/boundary condition loss terms, which often induce conflicting gradients. Gradient surgery methods mitigate this issue by constructing directions from loss-specific gradients to reduce conflict before optimizer transformation. However, even when the constructed direction is conflict-free, this property may not be preserved after optimizer transformation. Let $a_t$ denote the direction constructed by gradient surgery, $u_t$ the optimizer proposal, and $\mathcal{C}_t$ the conflict-free cone induced by the loss-specific gradients. We show that modern optimizers can transform $a_t$ through mechanisms such as historical state, adaptive scaling, preconditioning, or decoupled weight decay, so $a_t \in \mathcal{C}_t$ does not generally imply $u_t \in \mathcal{C}_t$. We refer to this optimizer-induced discrepancy in conflict-freeness between $a_t$ and $u_t$ as Gradient-Update Mismatch (GUM). Accordingly, we propose Gradient-Update Alignment (GUA), which projects $u_t$ onto $\mathcal{C}_t$ to obtain the aligned update $p_t$ and applies $p_t$ to the parameters. When the optimizer maintains internal state, GUA further adjusts this state toward targets reconstructed from the applied update. We conduct extensive experiments and find that GUM is widespread across momentum, adaptive, and curvature-based optimizers, with conflict rates reaching up to 86.3%. Across all PINN settings, GUA achieves conflict-free applied updates and consistently improves various gradient surgery methods, reducing the relative $L_2$ error by up to 98.2% in individual settings. Data and code are available at https://github.com/JingXiao10/GUA. --- ### Retrieved but not ranked: surface-form bias in structural retrieval, from mathematics to agent trajectories Authors: Nabira Rashid, Manolis Kellis Published: 2026-09-01 Categories: cs.LG, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2609.01556v1 We evaluate embedding retrieval where surface form and meaning are pulled apart on purpose: retrieving items that share underlying structure but not wording, in two unrelated domains under one protocol, competition mathematics (MathNet-Retrieve; 500 queries, 117,088-item corpus) and embodied-agent trajectories (ALFWorld-derived; 118 queries, 336 trajectories). In mathematics the failure is complete: strict Hit@1 at the heaviest disguise tier is 0.0% for both production embedders (bootstrap 95% CI (0.0, 0.0)) while the correct item sits in the top 10 nearly always, and in 95.2 to 99.8% of misses the winner is more lexically similar to the query than the correct answer. In trajectories, where surface variation is incidental, the same models land at or near hypergeometric chance when gold must involve a different object, and below chance for all three embedders once gold must differ in object and receptacle: retrieval anchors on literal tokens, not task structure. A lexical reranker control hurts in mathematics and helps in trajectories (closing 26 to 36% of the gap, CIs excluding zero); its sign reveals whether a benchmark's surface variation is adversarial or incidental. An LLM reranker recovers 5 to 63% of the gap in mathematics and 43 to 76% in trajectories; direction replicates across three judges (all 21 cells positive), but effect sizes, tier profiles, and the outlier judge change with domain (paired differences excluding zero everywhere). Mathematics gains concentrate on well-known competitions (+19.8 points, CI (+6.7, +33.2), one of six cells), so part of the recovery is memorization. In a paired downstream experiment (210 queries, graders at 96 to 99% agreement), oracle retrieval was indistinguishable from adversarially bad retrieval (McNemar p = 0.678); the solver's 69.5% zero-shot accuracy is largely a truncation proxy (97 to 100% on finished answers), leaving no headroom. --- ### BS: Take the Hint - Interactive Multitracer PET/CT Lesion Segmentation with a Scribble-Conditioned ResEnc U-Net Authors: Marven Sherif, Amgad Elmasry, Youssef Ghazal, Ayman Elghotni Published: 2026-09-01 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.01554v1 Automated lesion segmentation in whole-body PET/CT is complicated by the variety of physiological tracer uptake patterns and by the differing appearance of lesions across tracers. The autoPET/CT V challenge addresses this by making segmentation interactive: user scribbles marking foreground and background are supplied alongside the image, and the algorithm is expected to exploit them. We present our submission, a scribble-conditioned residual encoder U-Net operating on four input channels: CT, PET, and a sparse scribble map for each of foreground and background. The network is initialised from the autoPET-III winning weights and extended from two to four input channels, with the two scribble channels zero-initialised so that the pretrained representation is preserved exactly at initialisation. Every model is fine-tuned per fold from the corresponding autoPET-III fold checkpoint, so that no validation case is seen during pretraining. PET intensities are normalised against a per-scan aorta blood-pool reference derived from a CT segmentation, which removes tracer- and centre-specific scaling without requiring lesion labels. At inference the five fold models are ensembled by averaging their softmax outputs per sliding-window patch, before Gaussian-weighted stitching. On the challenge's five-fold split, with each fold evaluated on its own validation cases, mean Dice is 0.554 and mean lesion-level F1 is 0.528 without scribbles, rising to 0.751 and 0.733 after five correction rounds. About 85% of that gain follows the first scribble, and the spread between fold models narrows five-fold over the same rounds, so interaction largely compensates for how well or badly a given model segments unaided. --- ### Can LLMs Discover Scientific Laws in Real and Parallel Worlds? Authors: Yiming Huang, Ziche Liu, Zhuohang Wu, Yiqian Wang, Junxia Cui, Xinkai Zou, Linjun Mao, Nan Huang, Naicheng Yu, Kaijie Zhu, Yue Ma, Kun Zhou et al. Published: 2026-09-01 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.01552v1 Scientific equation discovery has long been central to scientific progress, proceeding through iterative cycles of hypothesis generation, observational testing, and refinement under scientific constraints. As LLM capabilities advance and their role in AI for Science expands, it remains an open problem whether they can genuinely discover scientific laws and how this ability should be evaluated. Existing evaluations, however, often either simplify discovery through synthetic settings or reuse published targets that may already be familiar to LLMs. We therefore introduce SCILAWS-BENCH, a benchmark for scientific law discovery built from published research and real scientific data. It comprises 118 problems drawn from 381 scientific papers, covering 291 candidate laws and roughly 8M real data points across six scientific disciplines. Each problem is instantiated in two complementary settings: (1) SCILAWS-REAL asks models to propose laws from fixed real observations and evaluates held-out predictive fit and scientific validity derived from the source literature, and (2) SCILAWS-PARALLEL asks models to actively query residual-calibrated worlds and recover synthesized hidden laws derived from published forms. This two-setting task design preserves each problem's scientific context while separately evaluating fixed-record law discovery and active recovery of a newly synthesized hidden law. We find that predictive fit can diverge from scientific validity, memorization shapes whether models reproduce or move beyond published formulas, and our best-of-N study reveals a selection bottleneck. Our work provides a paper-grounded benchmark and new empirical perspectives for evaluating AI for scientific discovery. Project page: https://yiyihum.github.io/SciLaws-Bench --- ### What, Where, and How: Probing Spatiotemporal Representations in Video Foundation Models Authors: Sharon S. Musa, Fereshteh Forghani, Harrish Thasarathan, Sonia Joseph, Matthew Kowal, Konstantinos G. Derpanis Published: 2026-09-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.01551v1 Self-supervised video foundation models learn rich spatiotemporal representations, yet it remains unclear what visual concepts these representations encode, where they emerge across transformer layers, and how they are geometrically organized. In this work, we tackle these three questions through a systematic layer-wise analysis of V-JEPA 2 and VideoMAE-v2. We leverage lightweight probes trained to discover three temporally grounded properties: (i) camera motion understanding, (ii) intuitive physics, and (iii) anomaly detection. Both models encode camera motion, with best results ($>90$ ROC AUC) emerging at 60-70% of network depth, and achieve moderate anomaly detection performance ($>60$ ROC AUC), but remain near chance on intuitive-physics tasks, suggesting a limited encoding of deeper physical reasoning. Beyond classification, we find that temporal features from individual videos form smooth low-dimensional trajectories in representation space, suggesting that camera motion is not only linearly decodable but also geometrically organized. Based on these results, we apply geometry-aware spline-based steering in the model's latent representations to interpolate camera motion, yielding steered videos with smoother trajectories and more coherent temporal progression than linear interpolation. --- ### Linear Reusable Neural Bases Architecture for Network Compression Authors: Binshuai Wang, Peng Wei, Mahyar Ghazanfari Published: 2026-09-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.01550v3 Memory constraints remain a critical bottleneck in the deployment of large-scale AI models. Parameter sharing across network depth reduces model storage, but repeatedly applying an identical transformation limits flexibility across layers. Inspired by time--memory trade-offs in classical algorithms, we introduce the Linear Reusable Neural Bases (LRNB) architecture, an RNN-based framework that improves parameter efficiency through parameter reuse at the \textit{neuron level}. Each feedforward residual module is represented as a linear combination of shared neural bases, with depth-specific learnable coefficients and optional shifts providing flexibility across layers. This formulation reduces parameter redundancy across depth and enables the construction of wider and deeper networks within a fixed parameter budget. We further provide a geometric interpretation of the neural bases from a vector-field perspective and extend the framework to linear projection modules. Experiments demonstrate that the LRNB architecture achieves comparable or lower final training loss than independently parameterized baselines while using fewer parameters and maintaining stable training dynamics. These findings support neuron-level reuse as a practical approach to parameter-efficient network design. --- ### A Mathematical Theory of Reusable Neural Bases for Network Compression Authors: Binshuai Wang Published: 2026-09-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.01550v1 As large AI models become increasingly prevalent across a wide range of applications, memory cost has become a critical bottleneck in both training and inference. To mitigate this issue, we introduce the Linear Reusable Neural Bases Architecture (LRNBA), a novel framework aimed at improving parameter efficiency and reducing memory cost. Inspired by recurrent neural network (RNN) designs, the core idea of our approach is to represent each network block as a linear combination of a shared set of neural bases, thereby enjoying highly network compression rate while maintaining stable training. The proposed architecture allows for the construction of significantly wider and deeper networks under the same parameter budget. Extensive experiments demonstrate that our model achieves comparable or even faster convergence and lower loss than classical architectures, while maintaining stable training dynamics. --- ### NashDreamer: Model-Based Reinforcement Learning for Zero-Sum Imperfect-Information Games Authors: Tomáš Holeček, Viliam Lisý Published: 2026-09-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.01549v1 Model-based reinforcement learning (MBRL) has achieved remarkable results in single-agent domains, yet its extension to competitive imperfect information games (IIGs) remains underexplored. In multi-agent settings, opponent-induced non-stationarity complicates the learning process, and decentralized model learning faces severe identifiability barriers, which we argue make centralized model learning a mathematical necessity. Building on this analysis, we propose NashDreamer, a principled MBRL framework for two-player zero-sum IIGs. It introduces a centralized Multi-Agent Recurrent State-Space Model (MARSSM) that decouples environment dynamics from the effect of players' strategies on their individual observations. NashDreamer is designed to use arbitrary policy gradient algorithms and inherits their convergence guarantees towards Nash equilibria under an idealized model. Empirical evaluations across four benchmark games demonstrate that NashDreamer substantially improves sample efficiency over model-free baselines early in the training. Finally, we theoretically analyze the architecture's optimization landscape, identifying the vulnerability of the Dreamer family of algorithms to posterior collapse in stochastic environments. We leave it as an open challenge. --- ### SDARE-Bench: Evaluating Large Language Models on Conversational Stigma Detection and Response in Dyadic and Group Dialogue Authors: Stephanie Fong, Yiwen Jiang, Zimu Wang, Hongxi Yang, Yaling Shen, Hiu Weh Naomi Chow, Heung Ying Lai, Xiangyu Zhao, Qingyang Xu, Zhongxing Xu, Jiahe Liu, Guilherme C. Oliveira et al. Published: 2026-09-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.01548v1 Large Language Models (LLMs) are increasingly used in advice seeking and decision making that may affect social judgements. Despite stigma's profound effects on people and communities, benchmarks remain scarce. Existing general-domain evaluations typically rely on static prompts and fixed-format tasks, overlooking conversational contexts and audience effects in everyday communication. To address these gaps, we introduce SDARE-Bench, the first scenario-based benchmark evaluating both stigma detection and open-ended response generation in LLMs, comprising 1,138 dyadic queries and 1,388 group dialogue. Empirical results across 8 LLMs consistently demonstrate poor identification of stigma components, especially in group dialogues. In open-ended response generation, stigma expression was substantially higher in group settings than in dyadic, with weaker resistance to stigma and more unrealistic advice. Responses were evaluated using a classifier trained on 1,392 human annotated responses. In constructed group pressure settings, stigma expression rates further increased to a striking average of 97.5%. Our findings identify stigma response as a recurring LLM safety vulnerability, especially in socially complex conversational contexts. --- ### Variable Selection for Feature-Based Newsvendor Authors: Zhaoliang Yuan, Jie Wang Published: 2026-09-01 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.01544v1 Feature-based newsvendor models use observable covariates to tailor inventory decisions, aiming to balance holding and shortage costs under demand uncertainty. However, high-dimensional feature sets often hinder interpretability and inflate data collection and implementation costs. This paper studies variable selection for the feature-based newsvendor problem under a hard cardinality constraint on the number of selected features. We formulate the resulting $\ell_0$-constrained empirical newsvendor problem with $\ell_2$-regularization, establish its computational hardness, and develop a mixed-integer second-order cone programming reformulation that strengthens the standard Big-$M$ formulation. To enable scalability beyond exact optimization, we develop a randomized-rounding algorithm with a bi-criteria guarantee and a greedy heuristic. Statistically, we provide theoretical analysis of the resulting sparse policy estimator, including finite-sample estimation error, out-of-sample risk bounds, and support recovery guarantees. Extensive experiments on both synthetic and real data illustrate the computational and statistical trade-offs among various baselines. Our results demonstrate that the proposed variable selection framework achieves competitive out-of-sample operational costs while using substantially fewer covariates. --- ### Quantum Sparse Autoencoders for Q-Matrix Estimation in Cognitive Diagnosis Authors: Arif Hassan Zidan, Yi Pan, Bowen Guo, Xiang Li, Yu Bao, Yingfeng Wang, Tianming Liu, Wei Zhang Published: 2026-09-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.01537v1 Q-matrices play a central role in cognitive diagnosis within educational data mining (EDM), specifying which latent skills each assessment item requires. Data-driven Q-matrix estimation remains challenging when assessments involve many correlated skills and when real response patterns depart from idealized generative assumptions. We introduce a novel quantum sparse autoencoder (QSAE) for Q-matrix estimation, which, to the best of our knowledge, is the first application of quantum machine learning (QML) to cognitive diagnosis. Overall, the QSAE embeds each student's binary response vector into a quantum circuit using an encoder, compresses it into a sparse latent representation, and maps that representation to the Q-matrix. We benchmark the QSAE against a classical autoencoder (CAE) across 60 simulated datasets and 9 real-world assessment datasets. The results reveal complementary strengths. Although the CAE partially achieves higher average accuracy under several simulation conditions, the QSAE is substantially more stable across replications, exhibiting lower variance in 49 of the 60 conditions. Moreover, on real assessment data, the QSAE outperforms the CAE on 6 of the 9 datasets. These findings suggest that the principal advancement of QML in this setting is not universal accuracy improvement, but enhanced robustness and capability to explore latent-structure complexity in real datasets. --- ### Can LLMs Design Video Coding Tools? A Case Study on Planar Mode Authors: Yingwen Zhang, Meng Wang, Liqiang He, Shiqi Wang Published: 2026-09-01 Categories: cs.MM, cs.AI arXiv: https://arxiv.org/abs/2609.01535v1 This paper explores whether large language models (LLMs) can design video coding tools, a highly challenging task due to the intricate algorithmic coupling of tool modifications. In particular, we present an empirical case study on the Planar mode, a long-standing intra prediction tool in video coding standards. Our experiments operate within a generation-and-evaluation loop, with the LLM generating new Planar predictors, encoder trials evaluating their coding performance, and the LLM re-generating refined implementations based on the evaluation feedback. We first examine directly replacing the default Planar mode in the Fraunhofer Versatile Video Encoder (VVenC) under its faster preset. Experimental results demonstrate that the LLM-generated mode can outperform the conventional Planar mode on this lightweight toolset, achieving 0.18% bitrate savings with 0.4% complexity overhead on the standard benchmark. We further extend our evaluation to the Enhanced Compression Model (ECM). Leveraging newly introduced directional Planar modes, we investigate two integration strategies: directly replacing them, and introducing the LLM-generated predictor as an additional prediction mode with new syntax elements. The empirical results suggest that both strategies can yield coding gains under a constrained low-resolution setting. Overall, this study offers preliminary evidence and practical insights, highlighting both the potential and open challenges of LLM-based coding tool design. --- ### Knowledge Distillation During Mid-Training Favors Reasoning over Factual Recall Authors: Jacqueline He, Howard Yen, Shuyue Stella Li, Margaret Li, Hanqing Zeng, Yinglong Xia, Benyu Zhang, Zhuokai Zhao, Qiang Zhang, Pang Wei Koh, Luke Zettlemoyer, Wen-tau Yih Published: 2026-09-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.01532v2 Logit-based knowledge distillation (KD) is used to train smaller language models (LMs) via supervision from stronger teachers, but whether its benefits are consistent across training stages remains unclear. Through controlled experiments, we find that forward Kullback-Leibler (KL) distillation--the standard KD formulation--with post-trained teachers behaves fundamentally differently during mid-training, an intermediate phase of self-supervised learning on curated corpora. Surprisingly, while forward KD simultaneously improves reasoning and factual recall during pre-training relative to standard next-token prediction (NTP), it instead slows factual recall acquisition during mid-training despite continued reasoning gains. We trace this stage dependence to an asymmetry in teacher confidence across data domains and the student's evolving knowledge state: teachers are more confident on procedural than knowledge-intensive data, while students acquire low-entropy factual knowledge earlier in training. To mitigate this imbalance, we propose Switch Distillation, a simple mid-training objective that distills on tokens where the teacher is confident, using teacher predictive entropy as a lightweight routing signal, and otherwise falls back to cross-entropy. Switch Distillation consistently outperforms existing distillation objectives across teacher sizes. Relative to standard NTP, it achieves 1.61-1.71x the reasoning performance and 1.13-1.19x the knowledge and commonsense performance while preserving 96.7-96.8% of factual recall. Crucially, these benefits persist after post-training: Switch Distillation closes the factual recall gap while maintaining 1.25-1.32x and 1.13-1.20x gains in reasoning and knowledge and commonsense, respectively. --- ### Knowledge Distillation During Mid-Training Favors Reasoning over Factual Recall Authors: Jacqueline He, Howard Yen, Shuyue Stella Li, Margaret Li, Hanqing Zeng, Yinglong Xia, Benyu Zhang, Zhuokai Zhao, Qiang Zhang, Pang Wei Koh, Luke Zettlemoyer, Wen-tau Yih Published: 2026-09-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.01532v1 Logit-based knowledge distillation (KD) is used to train smaller language models (LMs) via supervision from stronger teachers, but whether its benefits are consistent across training stages remains unclear. Through controlled experiments, we find that forward Kullback-Leibler (KL) distillation--the standard KD formulation--with post-trained teachers behaves fundamentally differently during mid-training, an intermediate phase of self-supervised learning on curated corpora. Surprisingly, while forward KD simultaneously improves reasoning and factual recall during pre-training relative to standard next-token prediction (NTP), it instead slows factual recall acquisition during mid-training despite continued reasoning gains. We trace this stage dependence to an asymmetry in teacher confidence across data domains and the student's evolving knowledge state: teachers are more confident on procedural than knowledge-intensive data, while students acquire low-entropy factual knowledge earlier in training. To mitigate this imbalance, we propose Switch Distillation, a simple mid-training objective that distills on tokens where the teacher is confident, using teacher predictive entropy as a lightweight routing signal, and otherwise falls back to cross-entropy. Switch Distillation consistently outperforms existing distillation objectives across teacher sizes. Relative to standard NTP, it achieves 1.61-1.71x the reasoning performance and 1.13-1.19x the knowledge and commonsense performance while preserving 96.7-96.8% of factual recall. Crucially, these benefits persist after post-training: Switch Distillation closes the factual recall gap while maintaining 1.25-1.32x and 1.13-1.20x gains in reasoning and knowledge and commonsense, respectively. --- ### Revisiting Cross-View Completion: Self-Supervised Pre-Training via Reconstruction Error Comparison Authors: Thibaut Loiseau, Guillaume Bourmaud, Vincent Lepetit Published: 2026-09-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.01530v1 Self-supervised pre-training via cross-view completion learns strong features for 3D vision from co-visible regions of image pairs. However, the reference view provides little information for reconstructing non-co-visible patches, implicitly yielding a monocular training signal in these regions. We introduce Gekko, which turns this limitation into a useful signal. The relative improvement of the cross-view reconstruction error over a masked-autoencoder error is a self-supervised proxy for co-visibility: large improvements indicate co-visible regions, negligible ones non-co-visible areas. Gekko is a network, trained from scratch, that jointly performs cross-view completion, masked autoencoding, and per-pixel prediction of this relative improvement, providing an additional binocular signal for all masked regions without any ground-truth 3D annotation. Under identical architectures and training data, Gekko consistently outperforms CroCo on zero-shot correspondence estimation, relative pose estimation, and pointmap regression, with up to 6 times higher accuracy at the strictest relative-pose threshold and a 22% drop in end-point error on ETH3D. The extra channel it learns is itself a strong co-visibility detector on unseen scenes, and Gekko's frozen features outperform released cross-view backbones of comparable or larger size. It can also be trained directly from raw videos with a simple stride-based curriculum, removing the cumbersome 3D preprocessing prior methods require while matching models trained on curated data. Code and pre-trained models are publicly available. --- ### Sierpiński--Knopp Wasserstein Distance for Persistence Diagrams and Applications to 2-Wasserstein Approximation Authors: Sebastien Tchitchek, Julien Tierny Published: 2026-09-01 Categories: cs.CG, cs.LG arXiv: https://arxiv.org/abs/2609.01528v1 This paper introduces the Sierpiński-Knopp (SK) Wasserstein distance, a fast metric between persistence diagrams. The SK-Wasserstein distance, denoted $d_{\mathrm{SK}}$, maps diagram points and their diagonal projections to the unit interval via the Sierpiński-Knopp space-filling curve on the upper diagonal triangle. The encoded point sets are then efficiently matched via one-dimensional optimal assignment, in \(O(N\log N)\) steps, yielding an explicit diagonal-aware point assignment between the two input persistence diagrams. We show that the SK-Wasserstein distance controls the classical \(2\)-Wasserstein distance between diagrams, admits an explicit isometric embedding into a Hilbert space, and induces a positive-definite Gaussian kernel, making the resulting geometry directly compatible with Euclidean and kernel-based learning methods. A tighter surrogate dissimilarity, noted \(W_Γ\), is also introduced based on the point assignments along the curve. Experiments on 12 scientific collections comprising 227 diagrams show median per-collection speedup of \(d_{\mathrm{SK}}\) over state-of-the-art approximations of \(W_2\) is \(626\times\), while the aggregate speedup over the full benchmark is \(2100\times\). Average-linkage partitions obtained from \(d_{\mathrm{SK}}\) and \(W_Γ\) each exactly match the corresponding \(W_2\) partition on 8 of the 12 collections. Hilbert \(k\)-means and Gaussian spectral clustering, both based on \(d_{\mathrm{SK}}\), achieve mean adjusted Rand indices (ARI) of \(0.756\) and \(0.800\), respectively, with respect to the benchmark reference partitions, compared to \(0.750\) obtained by average linkage on \(W_2\). The Gaussian \(d_{\mathrm{SK}}\) kernel supports other kernel-based analysis tasks, as illustrated by its use for contiguous segmentation of ordered diagram collections in our experiments. ---