# Research Papers — 2026-07 · bundle page 124 of 222 > Full-text markdown of papers #6151–6200 of 11075 published in 2026-07, newest first. AI-readable mirrors of the full 11075-paper 2026-07 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-07/page/124 [← Previous bundle](https://aipod.dev/papers/archive/2026-07/page/123.md) [Next bundle →](https://aipod.dev/papers/archive/2026-07/page/125.md) ### QFireNet: A Quantum-Enhanced U-Net for Wildfire Segmentation from Sentinel-2 Imagery Authors: Jaiman Munshi, Tanvi Tewary, Sawyer Bloom, Aidan Chu, Chetan Maviti, Kyon Winston-Bey, Harshit Badjatia, Farhan Kittur, Vardhan Madhavarapu, Varun Kota, Joshua Kwon, Nazia Rangwala-Vohra et al. Published: 2026-07-14 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2607.14160v1 Wildfire detection from satellite imagery is a semantic image segmentation problem that has proven to be difficult due to challenges such as class imbalance, feature complexity, and atmospheric interference. In this paper, we build on the foundational U-Net image segmentation model to develop a quantum-hybrid solution in hopes of more effectively modeling the high-dimensional spectral feature space of the Sen2Fire dataset. We inject a variational quantum circuit in the bottleneck portion of U-Net, specifically the QuFeX and QB-Net ansatzes. We test a classical Feature Pyramid Network (FPN) for further comparative analysis of the model, and we also explore classical improvements to the U-Net model and its training process, including a compression of parameters, alternative loss functions, and uniform mixing of input data. Our primary finding is that under matched conditions, both QB-Net (with an $F_1$ score of 31.18) and QuFeX ($F_1 = 30.79$) outperformed the classical U-Net baseline results ($F_1 = 28.71$). Additionally, the classical FPN achieved a comparable score of 31.13. A crucial finding was that data mixing removed a significant domain shift between the geographically-separated train and test sets, which boosted the classical FPN $F_1$ score to 39.76. We validate the architecture's robustness and generalizability to the wildfire detection problem via cross-dataset transfer on the California Burned Areas (CaBuAr) dataset. Overall, we find that quantum machine learning has potential to provide an advantage in the problem of wildfire image segmentation, and further experiments will continue to validate and expand upon this finding. --- ### MemoHarness: Agent Harnesses That Learn from Experience Authors: Yue Huang, Wenjie Wang, Han Bao, Yuchen Ma, Xiaonan Luo, Yi Nian, Haomin Zhuang, Zheyuan Liu, Yue Zhao, Xiangliang Zhang Published: 2026-07-14 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2607.14159v1 An agent harness is the external control layer that turns a base LLM into an executable agent by managing context, tools, orchestration, memory, decoding, and output handling. While harness design strongly affects agent behavior, most automatic improvement methods optimize narrower artifacts such as prompts, pipelines, or workflows, and deployed agents usually reuse a single global harness for all cases. We introduce MemoHarness, an adaptive harness optimization framework that learns from its own executions. MemoHarness decomposes the harness into six editable control dimensions, stores per-case diagnoses and distilled global patterns in a dual-layer experience bank, and adapts the learned harness to each test case using retrieved experience without test-time labels, feedback, or additional search. In our evaluation across shell-agent, code-generation, and analytical-reasoning benchmarks, MemoHarness improves over the fixed harnesses we compare against and shows selective transfer to unseen suites and base models. Its additional context can also remain cost-competitive when much of the retrieved experience is cacheable. These results provide evidence that execution experience is a practical substrate for building agent harnesses that are more adaptive than a single static configuration, while leaving broader claims about statistical robustness and component attribution to future work. --- ### Orchestrating Power Grid Studies with Multi-Agent AI and MCP Servers Authors: Jérôme Picault, Clément Goubet Published: 2026-07-14 Categories: cs.AI, cs.CL, cs.MA, cs.SE arXiv: https://arxiv.org/abs/2607.14158v1 This position paper explores how Agentic AI and Model Context Protocol (MCP) can support power-grid studies in a Transmission System Operator (TSO) context. We focus on integrating Large Language Models with numerical simulation tools, structured workflows, and human supervision. We identify key industrial requirements for agent assisted grid studies and introduce pypowsybl-mcp, an MCP-based interface exposing selected capabilities of our simulation tool, pypowsybl to AI agents. This first step provides a testbed to study how agents can setup simulations, execute analyses, retrieve results, and interact with power-system simulators through standardized tool calls. We also discuss principles for human-in-the-loop, multi-agent workflows and outline an evaluation strategy combining technical metrics and practitioner feedback. The paper positions MCP-based tool integration as a step toward more interactive, auditable, and scalable grid-study environments. --- ### Certified Domain Consistency for Multi-Domain Retrieval: Label-Free Per-Domain Contamination Control with Conformal Risk Guarantees Authors: Jayakumar Manoharan Published: 2026-07-14 Categories: cs.LG, cs.AI, cs.IR, eess.SY arXiv: https://arxiv.org/abs/2607.14157v1 Retrieval over corpora that mix several domains often returns relevant but wrong-domain evidence that ranking metrics miss and that conformal risk control bounds only marginally, under-covering the worst domains. This work introduces C3R, a drop-in control layer that, from an inferred domain posterior and no query-time label, certifies a per-domain contamination budget where feasible and otherwise abstains rather than silently violating; on the hardest domains it guarantees a reduction, not a tight bound. The core is a two-split scheme built on risk-controlling prediction sets, whose finite-sample transfer bound crosses from the inferred to the true domain with fully estimable slack, supports heterogeneous budgets, and inverts for deployment. Population validity rests on this bound and a controlled simulation; across a thousand resampled calibrations the certificate never violates (a stability result) while marginal control violates the most-contaminated domain in every draw, and soft demotion retains more recall than the strongest calibrated cascade at equal certified contamination. The method replicates across open testbeds including an independent one from public federal regulations, and an LLM-judged downstream probe indicates wrong-authority grounding rises with contamination and falls under control. The layer is frozen-stack and reranker-agnostic. --- ### "Trust Junk" Leads to Unjustified Support for Highly Discriminatory Predictive Models Authors: Michael Correll, Lucy Havens, Mahsan Nourani Published: 2026-07-14 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2607.14152v1 The persuasive power of data visualizations can go awry: for instance, in an explainable AI (XAI) context, visualizations can produce over-trust of predictive models. In this paper, we use a crowdsourced study to show that providing accurate (but superfluous or irrelevant) data in a model explanation can, in fact, result in unjustified trust and other positive beliefs about a model, even when the model is patently discriminatory and unfair. Our results suggest that XAI designers and developers need to consider the implicit or explicit rhetorics of their work, and beware of the potential of visualizations to imbue models with unearned trust. --- ### Enhancing Small Language Models Reasoning through Knowledge Graph Grounding Authors: Dimitrios Kelesis, Konstantinos Bougiatiotis, Georgios Paliouras Published: 2026-07-14 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.14149v1 Although large language models (LLMs) have set benchmarks for zero-shot reasoning, their deployment remains cost-prohibitive and environmentally taxing. Small Language Models (SLMs) offer a sustainable alternative, but prone to errors, on tasks requiring complex, multi-hop logical grounding. We investigate a neuro-symbolic agentic framework to enhance the reasoning capabilities of SLMs, specifically Gemma 3 (1B, 4B) and Llama 3.2 (3B), using the CLUTRR kinship benchmark. Our approach transforms the SLM into a minimalist agent utilizing two specialized tool calls: extract_facts for symbolic triplet extraction and get_hint for expert reasoning via a Relational Graph Convolutional Network (RGCN). We evaluate these models across two configurations, both in an Oracle scenario with ground-truth triplets and a Realistic scenario relying on self-extracted knowledge. Our results reveal that while RGCN-derived hints provide a 1.5 - 2x performance gain over story-only baselines, the system is constrained by the extraction bottleneck and sequential deductive fragility, where early extraction errors compound over multi-hop chains. Furthermore, we identify a "distraction effect" in specific architectures where noisy, self-generated facts degrade performance despite the presence of expert hints. This work characterizes the challenges of symbolic grounding in low-resource agentic systems and provides a roadmap for iterative verification in neuro-symbolic agentic pipelines. --- ### ITGPT: A Transformer Based Architecture for the Generation of Dance Dance Revolution and In the Groove Charts Authors: Miguel O'Malley Published: 2026-07-14 Categories: cs.SD, cs.LG arXiv: https://arxiv.org/abs/2607.14148v1 Dance Dance Revolution and In the Groove are rhythm games consisting of songs and accompanying choreography, referred to as charts. Players press arrows on a device referred to as a dance pad in time with steps determined by the song's chart. The process of manual chart generation is timestaking and difficult, motivating interest in automation. We propose ITGPT, a new transformer based architecture for the generation of DDR/ITG charts, and demonstrate significant improvements to generation accuracy and computational cost in comparison to predecessor work. --- ### Breaking Refusal in the First Half: A Mechanistic Study of the Prefill Jailbreak Authors: Alex Kwon Published: 2026-07-14 Categories: cs.CL, cs.AI, cs.CR, cs.LG arXiv: https://arxiv.org/abs/2607.14147v1 Aligned language models refuse harmful requests, but a one-line prefill ("Sure, here is") strips the refusal. We ask where and how it fails. The harm representation stays intact: on the prompts the attack flips to compliance, a linear probe reads harm as high as on the refused ones (0.91-0.98), while behavioral refusal drops to chance. This holds across four models and three families (1.5-3.8B, and at 14B). Refusal is therefore a shallow, response-site computation. We localize it to an early window: a dose-matched position control shows the first half of the response suffices to break refusal, while the second half is nearly inert. Three causal probes converge on that window. Restoring the harm direction there partially re-engages refusal. Injecting the model's own refuse-state reverses the jailbreak (74%, held-out). And knocking out the early response's attention to the prefill, but not an equal attention mass elsewhere, selectively collapses the harmful continuation. A base-model control identifies the mechanism: the same knockout collapses the continuation prefill-specifically even in a non-safety-tuned base model (64% to 25% harmful content vs a matched control's 64%, replicated at 7B). So the prefill's grip is generic autoregressive conditioning, not safety-specific suppression, and "refusal restoration" is a model-dependent fallback. The dominant mechanism is passive. A small safety-specific attractor remains on top (logit-trace concentration 0.24 vs 0.03), whose active-vs-passive character we size but do not fully separate. No single direction or component is a clean handle either: the decision is decodable but distributed, and refusal tracks harm rather than scary surface. The consequence is structural: a monitor reading the untouched prompt-side representation is immune by construction, but only to response-site attacks. The mechanism is diffuse; the failure surface is local. --- ### ToolAnchor: Anchoring Counterfactual Context to Boost Agentic Tool-use Capability Authors: Weiting Liu, Jieyi Bi, Wanqi Zhou, Jianfeng Feng, Yining Ma, Ai Han, Wenlian Lu Published: 2026-07-14 Categories: cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2607.14145v1 Tool-augmented large language model agents excel at long-horizon tasks, yet they are typically post-trained on fixed toolsets. When tasks demand new tools, these agents struggle to incorporate them effectively, and retraining from scratch is often impractical. We identify the core obstacle in such toolset expansion problem as behavioral inertia: the tendency of agents to fall back on familiar tools and established reasoning patterns despite having access to new ones. We demonstrate that injecting counterfactual anchor contexts at critical decision points can break this inertia, recovering failed trajectories by eliciting suppressed agent capabilities. To scale this insight, we propose ToolAnchor, a framework that uses teacher models to hypothesize these counterfactual contexts, verifies them via student rollouts, and internalizes the successful interventions through agentic post-training. Extensive evaluations across general AI assistant (GAIA), textual search (BrowseComp), and visual search (VDR-Bench) tasks demonstrate that ToolAnchor consistently exhibits competitive performance under expanded toolsets. Our work bridges the gap between static post-training and dynamic adaptation, charting a new path for scalable agentic reinforcement learning. --- ### The Capability Convergence Hypothesis: Capability from Access Structure, Not Scale Authors: Wenhui Chen, Jianlin Chen, Ziyao Lin, Chi Man Vong Published: 2026-07-14 Categories: cs.AI, cs.IT arXiv: https://arxiv.org/abs/2607.14144v2 The Platonic Representation Hypothesis (PRH) holds that as models scale, representations of heterogeneous networks converge toward a shared model of reality. We propose its sequel and boundary, the Capability Convergence Hypothesis (CCH): under a fixed per-token inference budget, representational convergence does not entail capability convergence. Capability instead converges toward a class, the access-complete hybrid: any architecture holding both a compressive O(1)-state channel and a scalable verbatim-index channel. We anchor it on a witness task, the Newton's-apple problem in an infinite stream, and name three resource walls: a Shannon wall barring any o(Nb)-state architecture, a horizon wall barring any fixed window, and a circuit wall barring fixed-depth attention-only composition (conditional on TC0 != NC1). Under an explicit separability assumption a hybrid crosses all three by paying each wall's price, so capability is strictly super-additive under composition. We separate what we prove from what we conjecture: the access-completeness principle rests on information-theoretic lower bounds and pre-registered experiments, while the field-level convergence trend is an economics-motivated conjecture. We report the first pre-registered small-scale tests under criteria frozen before the data: the predicted scissors gap is measured (exact-retrieval error 0.994 vs. 0.000 once a 64-scalar state gains one global-attention layer), the state-tracking bifurcation lands at the registered boundary, and a conjunction witness shows an irreducibly two-channel solution; one prediction failed with its direction reversed and is reported as such. Representational convergence is given freely by scale; capability convergence must be purchased by access structure. --- ### Delving into the Temporal Challenges of Unified Video Protection Against Image-to-Video and Fine-Tuning-based Customization Authors: Yuxin Huang, Ziming Hong, Mingming Gong, Wanyu Wang, Jing Zhang, Tongliang Liu Published: 2026-07-14 Categories: cs.CV, cs.CR, cs.LG arXiv: https://arxiv.org/abs/2607.13336v1 Recent diffusion-based video generation models have enabled high-quality personalized video customization through both tuning-based pipelines, which fine-tune a video diffusion model, and reference-based pipelines such as image-to-video generation. However, these capabilities raise serious concerns about personal privacy, identity ownership and intellectual property protection. Existing anti-customization works focus on protecting images, while protection for videos against both reference- and tuning-based customization remains largely underexplored. Protecting videos in this setting raises three challenges: (i) Image-level perturbations, optimized frame by frame, cannot survive temporal compression by 3D video VAE. (ii) A video-level perturbation optimized on a single video is vulnerable to temporal editing and fails to protect unseen videos. (iii) Temporally inconsistent perturbations are not robust to temporal attacks. To address these challenges, we propose Temporally Consistent Universal Adversarial Perturbations (TC-UAP), the first protection method against both reference- and tuning-based video customization. TC-UAP optimizes an identity-level multi-frame UAP over sliding windows from multiple videos, accounting for local temporal dependencies induced by temporal compression in video VAE and enabling a single perturbation to protect unseen videos of varying lengths. Moreover, we introduce intrinsic temporal modeling and an extrinsic surrogate temporal-attack loss, which make the perturbation temporally consistent and robust to unseen temporal attacks. Empirically, quantitative and qualitative results show that TC-UAP achieves the strongest identity protection compared with existing methods under both reference- and tuning-based video customization, and remains robust under multiple unseen temporal attacks. --- ### Agora: Collective and Permissionless Internet-Scale Pretraining of Large Language Models Authors: Gil Avraham, Violetta Shevchenko, Hadi Mohaghegh Dolatabadi, Karol Pajak, James Snewin, Harry Xi, Rodney O'Donnell, Thalaiyasingam Ajanthan, Sameera Ramasinghe, Chamin Hewa Koneputugodage, Shamane Siriwardhana, Alexander Long Published: 2026-07-14 Categories: cs.LG, cs.DC arXiv: https://arxiv.org/abs/2607.13332v1 Training large language models at the multi-billion to trillion parameter scale is confined to datacenters, where data-parallel (DP) and model-parallel (MP) techniques presume homogeneous accelerators, high-speed interconnects, and a single orchestrating entity. Frontier model development is thereby concentrated among the few groups able to assemble such clusters. Meanwhile, an enormous pool of compute remains unusable for training: consumer and professional GPUs that are heterogeneous, preemptible, individually owned, and connected only by the internet. We present Agora, a system that makes efficient use of this compute. Agora combines bandwidth-efficient pipeline-parallel model sharding over internet-grade links with multi-party, fault-tolerant collective operations. Each participant holds only one stage of the model, and no single party ever possesses the full weights. We term this setup Protocol Learning: it enables collectively trained, collectively owned models, opening a path to open-source frontier training with economic sustainability. This report presents the outcome of a research effort spanning communication-efficient parallelism, asynchronous optimization, and fault-tolerant systems design. It culminates in the first demonstration of its kind: Pluralis-8B, an open, permissionless pretraining run of an 8.6B-parameter model on 500B tokens of FineWeb-Edu. The model was trained over 40 days by 330 contributor nodes, predominantly consumer GPUs on internet connections, joining and leaving throughout. The run sustained ~170k tokens/s and 4.2 tokens per TFLOP of pooled compute, 63% of the efficiency of a centralized H100 baseline, and converged to within a small margin of a centralized reference run. --- ### Accuracy-Preserving Stability Regularization for Large-Scale Retail Demand Forecasting Authors: Jize Li, Jiani He, Dishu Yang, Dingyan Shang, Jingjing Liu, Shiqi Huang Published: 2026-07-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.13331v1 Retail demand forecasts are reused across replenishment, capacity, labor, and transportation planning cycles. Point-error objectives do not constrain abrupt movement between adjacent forecasts, while post-hoc smoothing acts only after model fitting. We ask whether a training-time penalty on consecutive within-series movement can improve horizontal forecast-path stability without materially changing point accuracy. The penalty is evaluated in a temporal-structured pipeline combining recent-demand embeddings with calendar, price, hierarchy, item, and store features. On selected M5 demand series at 1000, 3000, and 4000-series scales, the stability-aware hybrid model improves Forecast Stability Score over XGBoost by 6.91%, 6.66%, and 7.68%, respectively, while RMSE changes remain within 0.72% across three random seeds. Post-hoc exponential smoothing attains lower raw movement but incurs a larger RMSE cost; training-time regularization preserves more point accuracy and performs favorably under normalized stability. These findings extend forecast evaluation from point-error minimization toward an accuracy-stability trade-off perspective for operational retail forecasting. --- ### Efficient Text-to-Audio Generation via Pruning Authors: Arshdeep Singh, Yi Yuan, Yun Chen, Wenwu Wang, Mark D. Plumbley Published: 2026-07-14 Categories: eess.AS, cs.AI arXiv: https://arxiv.org/abs/2607.13330v1 Diffusion-based text-to-audio generative models such as AudioLDM achieve high perceptual quality and strong semantic consistency; however, their practical deployment is hindered by the substantial computational cost of the U-Net denoising backbone. In this work, we apply model pruning to improve the computational efficiency of AudioLDM, a U-Net-based text-conditioned audio latent diffusion model. We analyse parameter redundancy across U-Net convolutional blocks and evaluate a filter-pruning strategy. Pruning is guided by norm-based criteria and followed by lightweight finetuning to recover performance losses. Experimental results demonstrate that up to 83% of the parameters and 39% of the multiply-accumulate operations of U-Net have been reduced while maintaining, and in some cases improving, generation quality compared to the baseline unpruned network. We find that pruning affects AudioLDM's ability to generate certain sound events including safety-critical sounds such as gunshots, sirens, and explosions, as well as mechanical sounds such as drills and sewing machines, and other sounds such as sprays and tick-tocks, which are mostly recovered by lightweight finetuning of the pruned model. --- ### Privacy Preserving Recommender Systems Balancing Personalization with Privacy Authors: Ranjeet K Jha, Venkata Suresh Gummadilli Published: 2026-07-14 Categories: cs.CR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.13328v1 Personalized recommendation systems are central to modern e-commerce and retail platforms, but they typically rely on centralized storage of detailed user interaction data, creating significant privacy and regulatory challenges. With increasing requirements from regulations such as GDPR, CCPA, and CPRA, organizations must develop recommendation systems that preserve user privacy without substantially degrading recommendation quality. This work presents and evaluates a privacy-preserving recommendation framework that combines federated learning, differential privacy, cohort-level modeling, and privacy-aware intelligent agents. The framework keeps raw user data decentralized while introducing mathematically bounded noise to model updates. Experiments were conducted on synthetic retail datasets that emulate customer clickstream and purchase behavior. Recommendation quality was evaluated using Click-Through Rate (CTR), Precision@K, Recall@K, and Normalized Discounted Cumulative Gain (NDCG@K) across multiple differential privacy budgets. We evaluate matrix factorization, neural collaborative filtering, and GRU4Rec under varying privacy constraints and analyze the trade-off between privacy and utility. An interactive Streamlit dashboard was developed to visualize recommendation performance, ranking stability, privacy-utility trade-offs, and fairness metrics. Results show that the proposed framework maintains competitive recommendation quality at moderate privacy budgets (approximately $ε\approx 5$), demonstrating that strong privacy guarantees can be achieved with limited impact on recommendation effectiveness. This work provides a practical framework for deploying privacy-preserving recommendation systems that balance personalization, regulatory compliance, and business objectives, offering a scalable approach for next-generation AI-driven retail platforms. --- ### SARFA: Segment Anything with Radiomic Feature Alignment Authors: Tyler Ward, Abdullah Imran Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.13323v1 The Segment Anything Model (SAM) has demonstrated strong generalizability across a variety of segmentation tasks. However, SAM often struggles in situations where the target to be segmented is ambiguous. This poses a problem in medical imaging, where accurate delineation of targets such as tumors is vital, but even expert radiologists can disagree on the appropriate boundary for a target. Addressing this, we propose SARFA (Segment Anything with Radiomic Feature Alignment), a novel framework for improved medical image segmentation. Via probabilistic prompting, SARFA generates a diverse set of plausible masks for each input image and optimizes them with a radiomics-driven training objective based on Fréchet Radiomic Distance (FRD) and Direct Preference Optimization (DPO). By minimizing the FRD between masked predicted and ground truth regions within each image, SARFA encourages segmentation outputs whose anatomical and textural characteristics align with clinically meaningful ground truth representations, without relying solely on pixel-level overlap. Evaluated on computed tomography (CT) and magnetic resonance imaging (MRI) benchmarks, SARFA outperforms existing ambiguous segmentation methods, demonstrating the effectiveness of radiomic feature alignment and DPO-style candidate mask ranking as a training objective. Our code is available at https://github.com/tbwa233/SARFA. --- ### Adapting Generalist Vehicle Models for High-Speed MPC Across Terrains Authors: Rwik Rana, Jesse Quattrociocchi, Christian Ellis, Nathan Tsoi, Garrett Warnell, Joydeep Biswas Published: 2026-07-14 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.13319v1 High-speed off-road autonomy requires precise closed-loop control for a target vehicle while remaining robust across changing terrains. Recent forward kinodynamic (FKD) prediction foundation models suggest a promising path, starting from a generalist model and specializing it to the target platform. However, effective specialization remains challenging, as it often requires substantial real-world data, and models adapted to one setting can still overfit to specific terrains or driving regimes. We present OptCar (Optimized Car), a recipe for bridging the gap from generalist to specialist FKD models that preserves cross-terrain generalization while optimizing performance for a specific vehicle. $\texttt{OptCar}$ introduces a history-conditioned dynamics adaptation module that encodes recent state-action observations into a dynamics context token, and then fine-tunes the generalist model using limited real-world data together with targeted synthetic rollouts from environment-specific system identification. In closed-loop model predictive control (MPC) experiments across three terrains and an out-of-distribution cart-pulling task, the largest gains appear at 6~m/s, the highest speed evaluated and the regime in which slip dominates tracking error. On vegetation and dirt, the most slip-diverse terrain, OptCar reduces 6~m/s trajectory tracking error by roughly 55% relative to a fine-tuned AnyCar baseline, and remains the most accurate even when an unseen cart payload changes the dynamics. With only 5 minutes of real data per terrain, OptCar is competitive on road with a specialist trained on 30 minutes of road data, and substantially outperforms it once the terrain changes. --- ### Reflecting Process Expertise in Procedural Material Generation Authors: Kunal Gupta, Gaurav Joshi, Yen-Ru Chen, Seemandhar Jain, Ishit Mehta, Manmohan Chandraker Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.13318v2 Procedural material creation underpins applications in digital content creation, visual effects, and 3D asset design. Achieving high-quality results requires more than reproducing node graphs -- it demands understanding the process by which experts construct materials. We formulate procedural material generation as retrieval-time process reasoning over expert demonstrations, elevating process to a first-class representation beyond graph-only synthesis. Concretely, we represent expert workflows as process traces: textual records of construction steps, parameters, and design intent. To instantiate this idea, we use a pretrained LLM-based ProcessSynthesizer to synthesize a process trace aligned with a user's intent and a pretrained LLM-based Compiler to ground the process trace into an executable Blender material graph. Because procedural expertise is most naturally conveyed through demonstrations, we leverage tutorial videos as a source of process knowledge and extract textual, LLM-compatible traces using automated video analysis tools. In an expert study with five Blender artists (avg. 7.5 years of experience), materials generated by reflecting expert demonstrations were found to produce workflows requiring fewer edits, and more closely match professional design strategies than methods operating solely on static artifacts. A user study with 150 participants further shows that our approach achieves superior generation and editing performance compared to prior procedural systems. All code, models, and data will be available at https://materialapprentice.github.io --- ### Meta-Learning Preferences for Multilingual LLM Alignment Authors: Jiaying Lin, Seongho Son, Nam Phuong Tran, Long Tran-thanh, Ilija Bogunovic, Debmalya Mandal Published: 2026-07-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.13315v2 Unequal availability of human preference data across languages poses a significant challenge for aligning large language models in multilingual settings. To address the lack of sufficient data in low-resource language alignment, we propose a meta-learning framework for Reinforcement Learning from Human Feedback and Direct Preference Optimization. By leveraging preference data from other languages, our framework learns a transferable initialization that enables effective adaptation to a target language with minimal data. We provide theoretical guarantees for both the meta-reward modeling and meta-policy optimization settings, and empirically demonstrate the effectiveness of our approach on multilingual benchmarks. In an extremely low-resource setting with only 100 target-language preference samples, our approach achieves up to $28\%$ win-rate improvements over baseline methods, and consistently outperforms baselines across multiple target languages and model scales. Our approaches retain these advantages across different combinations of meta-training languages and varying linguistic distances from the target languages. --- ### Tabular Foundation Models for Discrete Choice Estimation Authors: Liu Liu, Dan Zhang Published: 2026-07-14 Categories: cs.LG, cs.AI, econ.EM arXiv: https://arxiv.org/abs/2607.13314v2 Tabular foundation models (TFMs) generate predictions on structured data via in-context learning, without task-specific estimation. We ask whether TFMs can be effectively applied to discrete choice, a central demand estimation framework in marketing and operations, and find that directly applying TFMs yields limited performance. The gap is structural: TFMs assume row-independent observations, whereas discrete choice is inherently set-valued and subject to persistent consumer preference heterogeneity. We propose a reformulation that encodes both choice-set dependence and individual heterogeneity within a row-based learning framework. Evaluated on a yogurt scanner panel, individual-level heterogeneity encoding is the dominant driver of predictive accuracy. The best reformulation outperforms hierarchical Bayesian estimation on both holdout log-likelihood and hit rate, running 16 times faster, a practical advantage for large-scale demand estimation. The advantage is largest in the medium-data regime (10--40 purchase occasions per consumer), where parametric Bayesian shrinkage most distorts estimates for atypical consumers. Fine-tuning on population choice data provides additional gains for consumers with shallow purchase histories, where in-context learning has limited individual-specific signal to condition on. These results establish a principled approach for applying foundation models to consumer choice problems more broadly. --- ### Finding the Right Tables and Columns: A Benchmark and Corpus-Adaptive Embeddings for SQL Schema Retrieval Authors: Qingcheng Zeng, Puxuan Yu, Aman Mehta, Fuheng Zhao, Rajhans Samdani Published: 2026-07-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.13311v1 Retrieval in the SQL setting has largely been studied as the task of finding, within a large collection of SQL statements, the statement that answers a natural-language question. At scale, however, a more fundamental retrieval problem precedes generation: schema retrieval, identifying the tables and columns a question requires in a database that may contain thousands of them, far more than fit in a model's context. We argue that this step warrants first-class evaluation. To this end, we recast five text-to-SQL datasets (Spider, BIRD, BEAVER, and two LiveSQLBench variants) as retrieval tasks at both table and column granularity, covering realistic and enterprise-scale schemas under two document representations, and we show that off-the-shelf text and code embedders transfer poorly to this setting. We then propose corpus-adaptive fine-tuning: natural-language queries are synthesized directly from the target schema corpus, granularity-aware hard negatives are mined, and a 305M-parameter embedder is fine-tuned contrastively. This procedure raises average recall@10 from 60.4 to 75.6 (nDCG@10 from 51.9 to 68.0), making the 305M model the strongest retriever under one billion parameters and competitive with state-of-the-art embedders of 4-8B parameters, more than an order of magnitude larger. The same recipe improves an 8B state-of-the-art embedder from 77.8 to 78.4 recall@10, matching the best result on the benchmark and indicating that the adaptation is backbone-agnostic. Leave-one-corpus-out experiments and a leakage audit show that these gains reflect a transferable schema-retrieval ability rather than memorization of the evaluation data. Our results establish schema linking as a standalone retrieval task and lightweight, label-free corpus adaptation as a practical route to deploying it at enterprise scale. --- ### Accuracy Without Grounding: Diagnosing Visual Dependency Dissociation in Video LLM Benchmarks Authors: Jae Joong Lee Published: 2026-07-14 Categories: cs.CV, cs.AI, cs.MM arXiv: https://arxiv.org/abs/2607.13305v1 Benchmark accuracy in video large language models (LLMs) is often treated as evidence of visual understanding. We audit this assumption across twenty models spanning 2-78B parameters and ten architecture families. We introduce the Visual Dependency Gap (VDG), the difference in per-question correctness between original-video and black-screen conditions. Paired McNemar tests on MVBench show that accuracy and visual dependency are separable: models differ on original video (p = 0.0003) but not on black screens (p = 0.53). Across models, task-type rankings are stable: Attribute Perception is strongly visual, whereas Temporal Reasoning approaches the language-only baseline. A diagnostic ladder from black screen to single frame, shuffled frames, and original video reveals that frame diversity supplies most of the visual benefit, while temporal order contributes near-zero accuracy across sixteen open-weight models. An ablation from 0.5 to 24 FPS rules out sparse sampling as the cause. H.264 experiments further show that stable aggregate accuracy conceals bidirectional question-level answer flips. The diagnostic also generalizes to four API-accessed models, whose VDG values range from 0.025 to 0.315. These results motivate VDG as a standard audit for whether video benchmarks measure visually grounded capability. Code is available at https://github.com/JaeLee18/accuracy-without-grounding. --- ### Where Does the Noise Come From? A Variance-Components Decomposition of Non-Determinism in LLM Brand Answers Authors: Dmitrij Żatuchin Published: 2026-07-14 Categories: cs.IR, cs.CL arXiv: https://arxiv.org/abs/2607.13304v1 Teams measuring whether large language models (LLMs) recommend a brand face a reproducibility problem: ask the same question twice and the answer moves. Practice resamples each prompt a few times (commonly five) and averages, treating within-prompt resampling as the source of the noise. But a measured brand score moves for at least four separable reasons: within-prompt resampling, prompt paraphrase, model identity, and query language. We specify a crossed random-effects (generalizability-theory) decomposition that partitions the total variance of a response-level brand outcome into these four sources, and embed the components in a decision-study allocation that returns how many repeats, paraphrases, models, and languages to buy for a target reliability. We apply it to a fully crossed corpus of 12,933 LLM responses on 20 Central and Eastern European brands, 8 languages, and 3 models (GPT-5.2 and Gemini 3 Flash in parametric mode, Perplexity in grounded retrieval), with a stability subset of 1,435 cells resampled about five times. The outcome is per-response multilingual sentiment polarity. Query language is the largest systematic facet (26.5% of the variance of one response) against 1.5% for brand identity (ICC 0.0146), so a single AI answer carries almost no brand-discriminating signal. Once a cell term isolates pure resampling, resampling is 34.8% of variance and the brand-in-context interaction 29.6%; brand-by-language is 8.6% (a bilingual penalty) while brand-by-model and brand-by-prompt are near zero. Per unit of query budget, adding languages and models reduces relative-error variance far more than adding repeats: a repeat past the fifth reduces it by only 0.0003. Brand-ranking reliability stays low, near 0.01 for a single answer and about 0.36 at the full crossed design, so reliability is bought by spreading across languages and models, not by repeating one prompt. --- ### Faithful Autoformalization of Natural Language Assertions Authors: Hongyi Liu, Madhusudan Parthasarathy, Adithya Murali Published: 2026-07-14 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2607.13303v2 Formal contracts are essential for software testing and verification, yet writing them remains labor-intensive and error-prone. LLMs offer a promising path toward autoformalization: synthesizing executable assertions from natural-language specifications and thereby bridging the gap between informal developer intent and formal executable specifications. We present Monty: an autoformalization framework for assertions that tackles the challenges of expectations of validity of assertions and ambiguity in natural-language. Our techniques are based on filtering formalizations using a novel conformance score metric and validity scores obtained from testing the code against formalized assertions. We evaluate our approach on 541 assertion-generation tasks derived from 22 collection-like Java classes, and show that our technique produces the ground truth more reliably (improving upto 20 points in precision on average) than when using LLMs naively to translate assertions. --- ### Improving Medical Image Generative Models with Fréchet Distance Loss Authors: Andrew Marshall, Xuanang Xu, Xiaoran Zhang, Rui Wang, Lawrence Staib, James Duncan Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.13300v1 Diffusion generative models have demonstrated immense potential for synthetic medical image generation. However, these models often struggle to capture complex morphological characteristics of heterogeneous tumors with irregular boundaries, limiting their utility for downstream clinical tasks such as segmentation. This limitation stems from the standard denoising objective: minimizing a per-pixel error, which smooths high-variance irregular structures characteristic of tumors. To address this, we propose finetuning these generative models with Fréchet Distance loss (FD-loss). FD-loss aligns the first and second order feature statistics of real and generated images in a pretrained encoder space, encouraging the generator to capture complex structural variations characteristic of heterogeneous tumors. We integrate FD-loss across diverse architectural settings, using both natural- and medical-image encoders on multiple liver and brain cancer datasets spanning CT and MRI modalities. Downstream segmentation networks trained on our FD-regularized synthetic data consistently achieve superior performance, improving tumor DSC by $>$$5\%$ over unregularized synthetic augmentation alone. Qualitative analysis suggests these gains are associated with more faithful tumor synthesis and fewer segmentation hallucinations. Our results show FD-loss as an effective regularizer for medical image generative models to improve clinical workflows. --- ### FOLIO: Focused Semantic Memory for Streaming Video Understanding Authors: Haoyang Fan, Dhruv Parikh, Anvitha Ramachandran, Sameh Gobriel, Nilesh Jain, Rajgopal Kannan, Viktor Prasanna Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.13298v1 In online streaming video understanding, a video stream continues to arrive and queries may be issued at any time. Because streaming frames grow without bound, the system must continuously compress and retain information from the observed video prefix while future frames and future queries remain unknown. The core challenge is deciding what information to retain and how to organize the maintained history: as this history grows with the stream, memory cost increases and many redundant visual details are retained, whereas later queries often depend on specific entities, actions, and their temporal changes. To address this challenge, we introduce FOLIO, a training-free focused semantic memory system that records important parts of the stream in higher detail while keeping surrounding context compact. As the stream arrives, FOLIO updates memory at the segment level, guided by a dynamic focus state, combining a short-term visual buffer with a long-term semantic memory organized around observed entities and linked to a visual-evidence cache. At query time, lightweight hybrid retrieval combines direct matching over the structured memory with semantic query expansion. FOLIO achieves state-of-the-art performance, reaching 82.0/69.1 Perception/Backward accuracy on OVO-Bench with Qwen3-VL-8B and 74.5 overall accuracy on StreamingBench, while substantially reducing the cost of maintaining streaming memory by reserving detailed records for focused entities and storing surrounding context compactly. --- ### Theory-Level Autoformalization: From Isolated Statements to Unified Formal Knowledge Bases Authors: Marcus J. Min, Mike He, Zhaoyu Li, Zixuan Yi, Sharad Malik, Aarti Gupta, Xujie Si, Osbert Bastani Published: 2026-07-14 Categories: cs.AI, cs.CL, cs.LG, cs.PL arXiv: https://arxiv.org/abs/2607.13292v2 Autoformalization translates informal natural language into formal, machine-verifiable languages. While most work focuses on individual statements, real formalization efforts are inherently theory-level: they require an entire web of axioms, definitions, and lemmas before target theorems can even be stated. In this position paper, we argue for theory-level autoformalization: formalizing complete theories, including all their inter-dependencies, as structured libraries. We examine the significance of this shift, address alternative views, identify open challenges, and propose three promising paths forward. Our survey of autoformalization is available at https://github.com/marcusm117/Awesome-Autoformalization. --- ### Harness Handbook: Making Evolving Agent Harnesses Readable,Navigable, and Editable Authors: Ruhan Wang, Yucheng Shi, Zongxia Li, Zhongzhi Li, Yue Yu, Junyao Yang, Kishan Panaganti, Haitao Mi, Dongruo Zhou, Leoweiliang Published: 2026-07-14 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2607.13285v1 The capability of a modern AI agent depends not only on its foundation model but also on its harness, which constructs prompts, manages state, invokes tools, and coordinates execution. As models, APIs, environments, and requirements evolve, the harness must be continually modified. Before such a change can be made, a developer or coding agent must identify all code locations that implement the target behavior. This is difficult because production harnesses are large, tightly coupled, and behaviorally distributed, while modification requests describe what the system should do and repositories are organized by files and modules. Code search, repository indexing, and long-context processing ease inspection, but still leave this behavior-to-code mapping to be recovered by hand. Behavior localization is therefore a central bottleneck in harness evolution. We introduce the Harness Handbook, a behavior-centric representation synthesized automatically from a harness codebase via static analysis and LLM-assisted structuring, linking each behavior to its corresponding source. We also introduce Behavior-Guided Progressive Disclosure (BGPD), which guides agents from high-level behaviors to relevant implementation details and verifies candidate locations against the current source. On diverse modification requests from two open-source harnesses, Handbook-Assisted planning improves behavior localization and edit-plan quality while using fewer planner tokens, with the largest gains on scattered sites, rarely executed paths, and cross-module interactions. Evolving complex agentic systems thus depends not only on generating edits, but also on determining where those edits should be made. --- ### Deconstructing Actor-Critic: A Large-scale Empirical Study of Design Components for Practitioners Authors: Haseeb Shah, Lingwei Zhu, Adam White, Martha White Published: 2026-07-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.13274v1 Reinforcement learning is increasingly being considered for controlling real-world systems, from fusion plasma and autonomous vehicles to drug discovery and drinking water treatment, where reliability is essential and tuning budgets are limited. Actor-critic algorithms share a set of design decisions, such as how the policy is updated, how it represents the distribution over actions, how its gradient is estimated, and how often it is updated relative to the value estimator. Using a control task derived from a real water treatment plant, we analyze over 33,000 experiments to determine how these components affect variability across runs and sensitivity to hyperparameters. Common defaults, such as Gaussian action distributions with pathwise gradient estimators, are among the least reliable configurations, whereas bounded distributions with adaptive update schedules remain robust across a wide range of settings. These findings offer empirical guidance to practitioners across scientific and engineering domains for understanding and making component-level decisions when adapting actor-critic methods to new real-world control settings. --- ### Differentiable Polarized Path Tracing Authors: Pramod Rao, Jérémy Riviere, Xilong Zhou, Abhijeet Ghosh, Abhimitra Meka, Thabo Beeler, Marc Habermann, Christian Theobalt, Delio Vicini Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.13265v1 Physically based differentiable rendering has proven to be a powerful tool for inverse rendering problems (e.g., 3D reconstruction, reflectance estimation, lighting estimation). However, most existing methods operate solely on radiometric intensity, discarding valuable polarization cues that constrain scene geometry and material properties. While forward simulation of polarized light is well-defined via Mueller-Stokes calculus, extending reverse-mode differentiation to this domain presents significant challenges. The rank-deficient nature of common polarimetric operators, such as linear polarizers and diffuse reflections, violates the invertibility assumptions of standard gradient estimators like path replay backpropagation and results in numerical instability. We address this by proposing a robust, polarization-aware differentiable path tracing method. Our approach estimates unbiased gradients through a combination of path replay and local caching. This formulation enables efficient and stable optimization of material and lighting parameters in complex scenes, broadening the applicability of physically based inverse rendering. Project page: https://vcai.mpi-inf.mpg.de/projects/DPPT/ --- ### Discourse-Aware Policy Analysis with Argumentation: A Hybrid LLM-Symbolic Framework for Disaster Governance Authors: Stylianos Loukas Vasileiou, Olga Derendiaeva Published: 2026-07-14 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.13260v1 Policy documents shape governance outcomes, but their reasoning is often implicit. Participatory commitments and managerial control routinely coexist in the same text, and the tensions between them are rarely stated directly. Existing computational approaches to policy discourse cannot express the frame-mediated relations that drive these tensions, where one argument narrows or instrumentalizes another rather than rejecting it. End-to-end summarization by large language models produces fluent text but offers little structure that domain experts can inspect or contest. We present Apaf, a hybrid LLM--symbolic pipeline that operationalizes critical discourse analysis as a quantitative bipolar argumentation framework over policy text. Arguments are first classified into deliberative or managerial frames. Four frame-mediated relation subtypes (agency reduction, agenda shift, instrumental support, and normative support) are then produced by deterministic rules over LLM-extracted features. We release a novel dataset of 100 sub-documents of disaster-risk-reduction policy from the USA, UK, Canada, and Australia, and show that the resulting argument graphs are accurate, interpretable, and stable across jurisdictions. --- ### AffectFlow-DINO: Uncertainty-Aware Multi-Task Affect Estimation via Conditional Rectified Flow Authors: Salah Eddine Bekhouche, Abdellah Zakaria Sellam, Fadi Dornaika, Abdenour Hadid Published: 2026-07-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.13250v1 We present \textbf{AffectFlow-DINO}, a multi-task learning system for the 11th ABAW challenge that extends a standard deterministic architecture with a conditional rectified-flow head to model the inherent ambiguity of in-the-wild facial behavior. Instead of predicting a single affect estimate, the model learns a conditional generative distribution, enabling uncertainty-aware one-to-many predictions through Monte Carlo sampling. The system jointly estimates continuous valence-arousal, classifies eight facial expressions, and detects twelve Action Units from static face images. Built on a frozen DINOv3 ViT-S/16 backbone, extensive ablation studies show that rectified-flow decoding consistently improves deterministic prediction, particularly for valence-arousal estimation (CCC-V $+0.058$). We further show that post-hoc threshold calibration effectively recovers performance on severely imbalanced rare classes (e.g., Fear: $3.8\% \rightarrow 33.1\%$) without retraining. Combined with backbone fine-tuning and flow retuning, the final model achieves $\mathbf{P_{MTL}=1.177}$, substantially outperforming the official challenge baseline of $P_{MTL}=0.45$. --- ### GSM-Plus-BN: A Perturbation-Based Benchmark for Bangla Mathematical Reasoning in Large Language Models Authors: Bidyarthi Paul, Nahida Jannat Mayouree, Md. Asif Karim, Sagar Chandra Nath, Swastika Kundu Published: 2026-07-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.13248v1 The evaluation of mathematical reasoning in large language models (LLMs) has predominantly focused on high-resource languages like English. This has created a significant barrier to the equitable development and deployment of AI in linguistically diverse regions such as Bangladesh, where over 230 million people speak Bengali. Despite this global significance, there has been minimal prior work on mathematical reasoning in Bengali and no existing research that systematically benchmarks a perturbated Bengali mathematical dataset, leaving a critical void in assessing model robustness and true comprehension beyond pattern recognition. This study addresses this gap by introducing GSM-Plus-BN, a novel perturbated Bengali mathematical dataset derived from the English GSM-Plus benchmark and verified by human translators. We evaluate six open-source LLMs Qwen3-32B, Llama-3.1-8B-Instant, Llama-3.3-70B-Versatile, Llama-4-Scout-17B-16E-Instruct, GPT-OSS-120B, and GPT-OSS-20B using a benchmark of 9,000 evaluation samples comprising 1,000 seed questions and 8,000 perturbed variants under both Standard Prompting and Chain-of-Thought (CoT) Prompting. Experimental results show that GPT-OSS-20B achieves the highest seed question accuracy of 96.08% under Standard Prompting, while larger models such as Llama-3.3-70B and GPT-OSS-120B demonstrate superior robustness across perturbation types. Furthermore, CoT prompting substantially improves reasoning for most models compared to Standard Prompting, yet a notable performance gap persists across all models relative to their English benchmarks, underscoring the inherent difficulty of perturbed Bengali text. This research makes a foundational contribution by providing GSM-PLUS-BN as a new resource and baseline for future Bengali mathematical reasoning research. --- ### Reassessing Muon for Matrix Factorization Authors: Ali Parviz, Gal Mishne, Alex Cloninger Published: 2026-07-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.13246v2 Muon has recently emerged as a strong optimizer for large-scale deep learning, where it reshapes gradient updates through approximate orthogonalization and has been reported to outperform Adam and AdamW in large language model training. Its empirical success has motivated a growing body of theoretical work that interprets Muon as steepest descent under the spectral norm. Yet it remains unclear which of Muon's advantages stem from its update rule itself and which are artifacts of the scale, architecture, and data of modern deep networks. In this work, we isolate the optimizer from these confounding factors by studying Muon on a simple, well-understood, and spectrally structured problem: low-rank matrix factorization. Through a controlled comparison against carefully tuned adaptive baselines, we find that Muon does not consistently outperform AdamW in this setting and that several previously reported advantages are sensitive to hyperparameter choices. Our results provide a more nuanced picture of when spectrum-aware orthogonalization is beneficial and argue for evaluating modern optimizers on controlled problems in addition to end-to-end benchmarks. --- ### Just-In-Time Scene Graph Growth: Combating Perceptual Saturation in Long-Horizon Robotics Authors: Yue Chang, Rufeng Chen, Yifan Tian, Dazhi Huang, Zhaofan Zhang, Yi Chen, Wenze Zhang, Li Chen, Hui Xiong, Sihong Xie Published: 2026-07-14 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2607.13245v2 While 3D Scene Graphs (3DSGs) provide crucial structured representations for embodied agents, conventional Ahead-of-Time, "build-everything-then-filter" pipelines conflict with the real-time, low-latency demands of edge platforms, inducing a perceptual saturation effect via severe observation redundancy. To resolve this, we present JITOMA (Just-In-Time On-demand Memory Activation), a closed-loop framework that unifies task reasoning, perception, and memory into a just-in-time growth process. Instead of exhaustively mapping the entire environment, JITOMA leverages a top-down task heatmap at the frontend to filter continuous observations, routing minimal streams to maintain a global foundation of low-cost, dormant anchors. Upon a cognitive query, the backend Large Language Model (LLM) parses the robotic intent to dynamically awaken task-relevant anchors, triggering expensive semantic operations such as dense node captioning exclusively within the activated local subgraph. To evaluate these dynamic capabilities and study perceptual saturation trade offs, we introduce JITOMA-Bench, a benchmark for long-horizon task switching and complex intent grounding. Across JITOMA-Bench, JITOMA maintains only 1--6 active semantic objects and 0.25--0.28 s/frame across all tiers, showing that semantic computation remains bounded by current task demand rather than accumulated scene complexity. --- ### Just-In-Time Scene Graph Growth: Combating Perceptual Saturation in Long-Horizon Robotics Authors: Yue Chang, Rufeng Chen, Yifan Tian, Dazhi Huang, Zhaofan Zhang, Yi Chen, Wenze Zhang, Li Chen, Hui Xiong, Sihong Xie Published: 2026-07-14 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2607.13245v1 While 3D Scene Graphs (3DSGs) provide crucial structured representations for embodied agents, conventional Ahead-of-Time, build-everything-then-filter pipelines conflict with the real-time, low-latency demands of edge platforms, inducing a perceptual saturation effect via severe observation redundancy. To resolve this, we present JITOMA (Just-In-Time On-demand Memory Activation), a closed-loop framework that unifies task reasoning, perception, and memory into a just-in-time growth process. Instead of exhaustively mapping the entire environment, JITOMA leverages a top-down task heatmap at the frontend to filter continuous observations, routing minimal streams to maintain a global foundation of low-cost, dormant anchors. Upon a cognitive query, the backend Large Language Model (LLM) parses the robotic intent to dynamically awaken task-relevant anchors, triggering resource-intensive operations -- such as dense node captioning and functional inference -- exclusively within the activated local subgraph. To evaluate these dynamic capabilities and study perceptual saturation trade-offs, we introduce JITOMA-Bench, a comprehensive suite for long-horizon multi-tasking and complex multi-step reasoning. Extensive experiments demonstrate that JITOMA substantially reduces active graph size and captioning latency, while maintaining stable processing time under long-horizon task switching. --- ### EMAGN: Efficient Multi-Attention Graph Network via Learned Clustering for Scalable Traffic Forecasting Authors: Mingxing Xu, Rakesh Chowdary Machineni, Ke Liu, Xi Cheng, Chengqi Lu, Xin Hu, Lyuhao Chen, Xiangyu Li, Junwei You, Oliver Gao Published: 2026-07-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.13241v1 Traffic forecasting is highly challenging due to complex and nonlinear spatial and temporal dependencies. Self-attention mechanisms have been widely adopted to model dynamic and long-range dependencies, achieving state-of-the-art performance, but suffer from limited scalability due to quadratic computational and memory complexity. To address this, we propose an Efficient Multi-Attention Graph Network (EMAGN) that linearises the spatial attention mechanism itself, inspired by the theory of fast high-dimensional Gaussian filtering. Two learned clustering matrices C_k and C_v adaptively group key and value vectors into M super-clusters, reducing complexity from O(N^2 d) to O(NMd) without sacrificing the flexibility of attention for dynamic dependency modelling. Experimental results on PEMS-BAY and METR-LA show that EMAGN achieves accuracy within 2.7-3.2% MAE of full-attention GMAN while reducing training time by 32%, inference time by 38%, and GPU memory by 58%. Critically, at K=16 attention heads, full-attention GMAN runs out of memory on a standard 11 GB GPU entirely while EMAGN continues to operate, demonstrating a categorical expansion of feasible model configurations. EMAGN also surpasses Linformer and Performer in both accuracy and efficiency within the same backbone, owing to its traffic-network-aware adaptive clustering. --- ### Cost-Optimal Foundation Model Deployment Portfolio for Transportation Management Authors: Xi Cheng, Ke Liu, Siyuan Feng, Jane Lin, H. Oliver Gao Published: 2026-07-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.13239v1 Foundation models, including large language models (LLMs) and vision-language models (VLMs), are increasingly used for transportation management center (TMC) tasks such as anomaly detection, incident reporting, and traveler information. Deploying multiple such models across TMC functions raises a portfolio question: which model should serve each function, in which deployment mode, and under what shared hardware budget? We formulate this as the Foundation Model Deployment Portfolio (FMDP) problem, a mixed-integer program minimizing total cost of ownership (TCO) subject to per-function quality, latency, and safety constraints over shared GPU capacity. We prove the problem NP-hard by reduction from the 0-1 knapsack problem and propose a polynomial-time greedy heuristic. In an illustrative case study with five TMC functions and 19 candidate (model, mode) pairs, FMDP identifies a mixed portfolio costing $34/mo (97% below the cheapest feasible all-closed-API baseline) by routing four functions to open-source APIs and the one function whose quality floor no open-source model meets to a closed API. Break-even analysis shows that on-premise GPU investment becomes reasonable only above approximately 309 vision queries/hour or if API prices double. --- ### Active Learning for Efficient Annotation of Surgical Videos with Weak Supervision Authors: Manasa Dendukuri, Matjaz Jogan, Daniel A. Hashimoto, Guiqiu Liao Published: 2026-07-14 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2607.13237v1 Precise spatial-temporal annotation of laparoscopic videos is time-consuming and requires expert knowledge. We propose a human-in-the-loop knowledge acquisition framework that combines active learning with dual-loss optimization to significantly reduce the annotation effort needed for automatic localization and segmentation of objects in the surgical field. Our method employs a foundation model to generate temporally consistent class activation maps (CAMs) from video using two complementary training objectives: a weak supervision loss on video-level tool presence labels for weakly annotated data, and an image-level mask loss on human-corrected annotations obtained through active learning. Rather than requiring dense pixel-level annotation upfront, our pipeline iteratively proposes pseudo-masks that guide the expert annotator to refine the knowledge previously captured by the model. We demonstrate that our framework reduces the effort of surgical video annotation by 50% by the end of training in comparison to fully manual annotation. Through eliminating the need for large, fully annotated datasets from the start, this framework enables scalability to the development of surgical tool segmentation models. This iterative human-in-the-loop refinement supports efficient knowledge acquisition with minimal expert input, providing a practical and deployable strategy for expanding tool segmentation to larger, more diverse datasets and real-world clinical settings. --- ### Continuously Evolving Deepfake Detection: An Architecture and Public-Benchmark Evaluation of a Dynamic Detection System Authors: Ken Jon Miyachi, Dylan Uys Published: 2026-07-14 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2607.13234v1 Deepfake detectors that achieve near-perfect scores on academic benchmarks collapse on real-world content: recent in-the-wild evaluations report AUC drops of 45-50% for state-of-the-art open-source models. We argue this gap is structural: static detectors are trained once against a moving generative frontier. We present BitMind Forensics (BMF), trained through Bittensor SN34, an open adversarial competition that continually refreshes the training distribution. We evaluate one dated export comprising image, general-video, and human-video checkpoints across nineteen public datasets: the canonical face-swap suites (FaceForensics++, Celeb-DF v1/v2/++, DFDC, DFD, UADFV, DF40) and recent in-the-wild and AI-generated-media benchmarks (Sumsub, Deepfake-Eval-2024, WildRF, Community Forensics, AIGCDetectBench, GenImage, AI-GenBench, AIGIBench, RAID, GenVidBench, GenVideo-100K). BMF reaches 0.936 AUC on Sumsub's original images and 0.872 pooled AUC over its full four-condition manipulation battery (1.4M images), staying robust under perturbation (0.855 JPEG, 0.799 downscaled), while GPEN enhancement improves detection (0.996). On Deepfake-Eval-2024, it matches the best commercial detector on images (0.915 vs 0.90) and exceeds it on video (0.822 vs 0.79), far above the best open-source detectors (0.56 and 0.63). It reaches 0.991 AUC on a 21-generator AI-image panel and 0.918 on GenVidBench, and exceeds the FF++-trained frontier on DFDC (0.947 vs 0.843) and Celeb-DF v2 (0.9985 vs 0.956), both contamination-audited, with statistical parity on Celeb-DF++. In a temporal study, successive dated exports improve on held-out media from generators absent from the static baseline's training (image 0.842 to 0.902; video 0.864 to 0.936). Our evaluation harness is public, and at publication the production API serves the exact evaluated snapshot for independent verification. --- ### AI-Native Insurance for Agentic AI: Pricing, Underwriting, and End-to-End Automation Authors: Quanyan Zhu Published: 2026-07-14 Categories: cs.AI, cs.CR, cs.CY arXiv: https://arxiv.org/abs/2607.13230v1 Agentic AI introduces new insurance challenges because autonomous AI systems can make decisions, invoke tools, modify external environments, and interact with third-party services. This paper develops an AI-native mathematical framework for underwriting, pricing, and contract design for agentic AI deployments. A deployment is represented by a risk state that captures autonomy level, operational authority, permission exposure, governance maturity, and dependency concentration. The framework maps the risk state to event probabilities, loss severities, governance costs, premiums, deductibles, coverage allocation, and policy covenants, and formulates an optimization problem for insurance contract design under participation, profitability, and incentive compatibility constraints. The paper establishes structural properties of insurability, including characterization of an insurability region, monotone deterioration of feasibility with increasing exposure, and governance certification thresholds. Insurance is further interpreted as both an operational cost and a regulatory mechanism for AI deployment. A healthcare case study illustrates contract optimization, sensitivity analysis, and automated claims processing for agentic AI systems. --- ### Audited Selective Verification for Risk-Controlled N-1 Thermal Contingency Screening under Deployment Shift Authors: Jayakumar Manoharan Published: 2026-07-14 Categories: eess.SY, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.13221v1 Real-time N-1 contingency screening in an energy management system trades assurance against cost: verifying every credible outage with full power flow is too slow, while fast linear-sensitivity screening gives no statistical guarantee and can silently pass unsafe operating points, especially when a controller drives the system into unfamiliar regimes. This paper introduces Audited Selective Verification, a risk-budgeted screening and triage layer for any controller's output (optimization, model-predictive, or learned). A cheap surrogate proposes which outages to skip; an online audit runs full power flow on a small random sample each window; and a calibrated threshold certifies a thermal-violation-rate bound for the skipped set at a chosen budget and confidence, with a corresponding bound for the unverified trusted subset. Validity rests on real verification and the audit rather than on surrogate accuracy, so it holds under arbitrary deployment shift. It is a risk-budgeted screen, not a replacement for deterministic verification when policy requires checking every credible contingency. On three public transmission systems up to 1354 buses, the realized violation rate stays within budget, standard deterministic and calibrated screens become unsafe under shift, and the method cuts full power-flow studies by 29 to 75 percent per real-time operating point. --- ### Networked Intelligence: Active Shared Context Graphs for Human-AI Team Science Authors: Sutanay Choudhury, Jeffrey J. Czajka, Lummy M. O. Monteiro, Erin Bredeweg, Jason McDermott, Katherine Wolf, Alex Beliaev, Josh Elmore, Paul Piehowski, Kylee Tate, Yuqian Gao, Aivett Bilbao et al. Published: 2026-07-14 Categories: cs.AI, cs.CE, cs.HC arXiv: https://arxiv.org/abs/2607.13220v3 Most AI-for-science systems focus on scaling a single reasoning process by using better models, larger context windows, long-horizon agentic execution, or digital co-scientists working with one principal user. However, challenging scientific problems are rarely solved by one reasoner alone. They are solved by teams whose members carry different priors, experimental background, tacit knowledge, and domain-trained intuitions. The open problem is therefore not only how to scale models, but how to develop "networked intelligence", scaling the connections between humans and AI systems so that a result or hypothesis produced in one context reaches another person, agent, instrument or robot that can act on it. We introduce Mycelium, an active shared workspace that automatically connects researchers and AI agents. As human users and agents work, the system captures important observations and hypotheses, tracks how they relate to the team's evolving knowledge model, and routes them to the person or agent whose next decision they can inform. We evaluate Mycelium through a real-world scientific discovery use case: a biological multi-omics campaign where shared context turned a local analytical finding into a cross-expert mechanistic constraint and ultimately into an experimental design. Finally, we describe networked intelligence as sparse conditional computation over distributed scientific contexts. This framework establishes when a scaled standalone agent is sufficient, and when isolated data and specialized expertise make a networked approach essential. --- ### CayleyR: Solving the TopSpin puzzle via cycle intersection Authors: Yuri Baramykov Published: 2026-07-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.13219v1 We present cayleyR, an R package for solving permutation puzzles by detecting cycle intersections in Cayley graphs. The core algorithm performs an iterative bidirectional search: from both the initial and target permutation states, random operation sequences generate cycles in the Cayley graph of the symmetric group Sn; their intersection yields a connecting path. When no direct intersection is found, a distance-guided bridge selection narrows the gap, and the process repeats. The package targets the TopSpin(n,k) puzzle, whose state space is a Cayley graph of Sn generated by a cyclic shift and a prefix reversal. We describe the mathematical framework, the algorithm, and its implementation, which combines a C++ hash-indexed state store with optional Vulkan GPU acceleration. The software is publicly available on CRAN. --- ### Graph Partitioning with Demands: Generalized Conductance and its Applications Authors: Michał Szyfelbein, Dariusz Dereniowski Published: 2026-07-14 Categories: cs.DS, cs.LG arXiv: https://arxiv.org/abs/2607.13218v1 In this work, we study various graph partitioning problems under a general demand model. In each such task, we are given a graph $G=(V,E,c,w)$ with a capacity function $c\colon E\to \mathbb{N}$ and a demand function $w\colon V\times V\to \mathbb{N}$. Our main focus is the problem of finding a cut $(S, \bar{S})$ minimizing the quantity \( ψ_w( S ) = \frac{c( S, \bar{S} )}{w( S, V )\cdot w( \bar{S}, V )}. \) Here, $c( S, \bar{S} )$ is the cost of edges between $S$ and the complement of $S$, $\bar{S}$, and $w( S, V )=w( S )+w( S, \bar{S} )$ is the sum of the internal demand within $S$, $w( S )$, and the demand between vertices of $S$ and $\bar{S}$, $w( S, \bar{S} )$. We call $ψ_w( S )$ the \emph{generalized conductance} of the cut $(S, \bar{S})$, and the task of minimizing $ψ_w( S )$ the Generalized Conductance Problem. Our main contribution is an algorithm with an $\mathcal{O}(\log n)$-approximation guarantee for this objective. Our result is achieved via a two-way reduction: first to the well-known Generalized $k$-Multicut Problem, and then to a constrained variant of the classic Sparsest-Cut Problem, with an additional upper-bound constraint on the amount of demand that may be cut. Moreover, we show that the above procedure can be used to obtain an $\mathcal{O}(\log n)$-bicriteria approximation for Graph Partitioning with Demands, where the goal is to find a minimum-cost subset of edges $C$ such that for every component $H$ of $G\setminus C$, $w( H )\leq ρ\cdot w( V )$. This, in turn, yields an $\mathcal{O}(\log n)$-approximation for Hierarchical Clustering with Demands, the problem of finding a hierarchy of cuts that partitions the graph into increasingly refined clusters. For multiplicative demand functions, we improve these guarantees to $\mathcal{O}(\sqrt{\log n})$ and for trees we get an $\mathcal{O}(1)$-approximation for all of our objectives. --- ### Hierarchical $\mathcal{F}$-Clustering: Approximation and Hardness of Clustering into Trees and Bounded Diameter Graphs Authors: Michał Szyfelbein, Dariusz Dereniowski Published: 2026-07-14 Categories: cs.DS, cs.LG arXiv: https://arxiv.org/abs/2607.13217v1 Consider the following variation on the Hierarchical Clustering problem: Usually, while building a hierarchical clustering, one recursively partitions the data until each cluster becomes a singleton. We relax the halting condition of the recursive process to stop whenever the remaining cluster is a graph belonging to a class $\mathcal{F}$. We call this problem Hierarchical $\mathcal{F}$-Clustering and we measure the quality of any solution using adapted Dasgupta's clustering objective. We study two natural choices of $\mathcal{F}$: trees and graphs of bounded diameter. We present the first polynomial time $\mathcal{O}(\log n\cdot\log\log n)$ and $\mathcal{O}(\log n)$-approximation algorithms for clustering into trees and bounded diameter graphs respectively. Our main technical contribution is a framework for approximating such problems based on linear programming. In fact, we characterize graphs classes $\mathcal{F}$ for which our approach can be applied and show that it includes both trees and bounded diameter graphs. However, our ideas are not limited to them and might be useful for other structures as well. Broadly speaking, our framework applies whenever the corresponding flat clustering problem, which we call $p_{\mathcal{F}}$-Partitioning, admits a natural ILP formulation together with a rounding procedure with provable approximation guarantees. Intuitively, given a set of vertices called terminals, the problem is to find an edge set whose removal results in satisfying certain vertex-dependent structural predicate for each terminal. We then use these ingredients to build clustering trees with the aforementioned approximation guarantees. To complement these results, we show that both Hierarchical Clustering into trees and into bounded diameter graphs cannot be approximated within any constant factor under the Small Set Expansion Hypothesis. --- ### Classifying daily activities needs posture, reconstructing them needs motion Authors: Arefeh Farahmandi, Gunnar Blohm Published: 2026-07-14 Categories: q-bio.NC, cs.AI, cs.IR, cs.LG arXiv: https://arxiv.org/abs/2607.13216v1 Humans recognize movements effortlessly, even from noisy and complex visual input. But what information in the stimulus allows humans to rapidly classify movements? No framework has systematically compared different strategies of movement analysis to address this question. Here, we used videos of 16 daily activities from the MoVi dataset and compared three strategies: Temporal Movement Primitives (TMPs), which decompose movements into weighted sums of temporally smooth basis functions; Legendre polynomial coefficients, which project joint-coordinate trajectories onto an orthogonal polynomial basis; and Autoencoder latent embeddings. Legendre coefficients and TMPs achieved the highest classifier accuracy, followed by autoencoders. We found two discriminative features for movement classification. The most informative is the general posture of the body, the average spatial configuration that distinguishes one activity from another. Additionally, we identified 9 critical joints that are most predictive for movement classification. Interestingly, good classification accuracy did not automatically lead to good movement generation: when we reconstructed movements for each activity, TMPs preserved the temporal dynamics and produced perceptually natural motion, whereas reconstructions from Legendre coefficients retained only the average posture and appeared frozen. These results reveal a dissociation in how movement information is organized: the static configuration of the body suffices to classify what activity is performed, but the temporal dynamics of movement are required to reconstruct how it unfolds. This distinction clarifies which features the visual system may rely upon for rapid action recognition, and suggests that postural features could enable efficient movement screening in clinical applications, while dynamic information remain essential wherever movement generation is the goal. --- ### Adaptive Filtering of the KV Cache: Diagnosing and Correcting Structural-Role Bias in LLM Inference Authors: Soumil Mandal Published: 2026-07-14 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.13205v2 Attention-based KV cache eviction (H2O and its descendants) compresses the memory-constrained state of a long-context model by ranking tokens on accumulated attention mass, treated here as signal energy, and keeping the heaviest. On schema-dense input streams such as nested JSON, this score acts as a non-stationary filter that disproportionately retains noise: a non-content sink role (delimiters or whitespace) carries an order of magnitude more energy than any content role, and structural KEY tokens are over-retained at roughly 1.8x the rate of the answer-carrying VALUE tokens, collapsing exact-match accuracy from 88% to 0% at a 5% budget as the signal-to-noise ratio of the retained state degrades. A counterfactual experiment establishes that suppressing KEY tokens is the best deployable filter. Our retraining-free, role-conditional allocation over SnapKV's windowed score, governed by a single tuned hyperparameter, closes 63-98% of the H2O gap at sub-20% budgets and, at higher budgets, modestly matches or exceeds full-cache accuracy -- a small, seed-sensitive denoising effect (borderline significant at B=0.50; not distinguishable from zero at B=0.30 over four seeds). A 15 MB linear role probe supplies these labels at negligible inference cost, though matching parser-level downstream accuracy remains open. --- ### Adaptive Filtering of the KV Cache: Diagnosing and Correcting Structural-Role Bias in LLM Inference Authors: Soumil Mandal Published: 2026-07-14 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.13205v1 Attention-based KV cache eviction (H2O and its descendants) compresses the memory-constrained state of a long-context model by ranking tokens on accumulated attention mass, treated here as signal energy, and keeping the heaviest. On schema-dense input streams such as nested JSON, this score acts as a non-stationary filter that disproportionately retains noise: a non-content sink role (delimiters or whitespace) carries an order of magnitude more energy than any content role, and structural KEY tokens are over-retained at roughly 1.8x the rate of the answer-carrying VALUE tokens, collapsing exact-match accuracy from 88% to 0% at a 5% budget as the signal-to-noise ratio of the retained state degrades. A counterfactual experiment establishes that suppressing KEY tokens is the best deployable filter. Our retraining-free, role-conditional allocation over SnapKV's windowed score, governed by a single tuned hyperparameter, closes 63-98% of the H2O gap at sub-20% budgets and, at higher budgets, modestly matches or exceeds full-cache accuracy -- a small, seed-sensitive denoising effect (borderline significant at B=0.50; not distinguishable from zero at B=0.30 over four seeds). A 15 MB linear role probe supplies these labels at negligible inference cost, though matching parser-level downstream accuracy remains open. --- ### Efficient Computing for Medical Image Acquisition and Reconstruction Authors: Xiao Wang, Jayasai Rajagopal, Md Safaiat Hossain, Peng Chen, Mohamed Wahib, Enzhi Zhang, Emma J. Reid Published: 2026-07-14 Categories: eess.IV, cs.CV, math-ph arXiv: https://arxiv.org/abs/2607.13204v1 Medical imaging systems such as CT, MRI, PET, and SPECT do not directly acquire images. Instead, they measure physical signals that encode anatomical or physiological information, and image reconstruction recovers the underlying image by solving an inverse problem. Although these imaging modalities are governed by different imaging physics, they share a common computational framework that naturally connects medical physics, linear algebra, probability, numerical optimization, and efficient computing. As medical imaging systems acquire increasingly large and higher-dimensional datasets, image reconstruction has become one of the primary computational bottlenecks in modern medical imaging. Advanced reconstruction methods, including analytical reconstruction, iterative optimization, and statistical model-based reconstruction, substantially improve image quality while reducing radiation dose or scan time, but at significantly increased computational cost. Efficient computing has therefore become essential for achieving clinically practical reconstruction times. This chapter presents a unified computational perspective on medical image acquisition and reconstruction across CT, MRI, PET, and SPECT. It first reviews the imaging physics and data acquisition process for each modality and derives a generalized mathematical framework for image reconstruction. Building on this framework, the chapter discusses analytical, iterative, and statistical reconstruction methods together with their computational characteristics. Finally, it examines efficient computing considerations, including optimization algorithms, physics-aware forward operators, memory-efficient implementations, and parallel computing strategies. Together, these topics demonstrate how the integration of imaging physics, mathematical modeling, and efficient computing enables accurate and scalable medical image reconstruction. ---