# Research Papers — 2026-07 · bundle page 97 of 222 > Full-text markdown of papers #4801–4850 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/97 [← Previous bundle](https://aipod.dev/papers/archive/2026-07/page/96.md) [Next bundle →](https://aipod.dev/papers/archive/2026-07/page/98.md) ### Autoregressive B-Rep Shape Generation with Parametric Surfaces Authors: Dafei Qin, Rui Xu, Zeyu Shen, Kaichun Qiao, Hongyang Lin, Qixuan Zhang, Huaijin Pi, Lan Xu, Jingyi Yu, Wenping Wang, Taku Komura Published: 2026-07-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.17093v1 Generative CAD modeling has broad design and application potential. Despite significant advances in Boundary Representation (B-Rep) generation, the dominant representation in CAD, existing methods largely depend on uniformly sampled point- or grid-based geometry representations, sacrificing native surface types and parameters and thereby limiting geometric fidelity and downstream usability. We present ParaCAD, an autoregressive framework for point-cloud-conditioned B-Rep generation that directly operates on native parametric surfaces. ParaCAD introduces a surface-centric tokenization that explicitly encodes each face by its exact surface type and continuous parameters, preserving the intrinsic semantics of CAD geometry. Our model first generates parametric surfaces with constrained UV domains, and then constructs a valid B-Rep by globally intersecting these surfaces to recover edges and vertices. ParaCAD places point-cloud-conditioned generation at the core of B-Rep synthesis, making it practical for user-guided reconstruction and seamless integration into existing 3D generation pipelines. Extensive experiments demonstrate that ParaCAD produces accurate B-Reps with faithful point-cloud alignment, outperforming point-based baselines in geometric precision, robustness, watertightness and downstream usability. --- ### ThRIve: Thermally Robust CNN Inference via Low-Rank Adaptation in Heterogeneous PIM Architectures Authors: Vibhanshu Sharma, Pratyush Dhingra, Janardhan Rao Doppa, Partha Pratim Pande Published: 2026-07-19 Categories: cs.AR, cs.ET, cs.LG arXiv: https://arxiv.org/abs/2607.17091v1 Processing-In-Memory (PIM) has emerged as a promising technology for accelerating machine learning (ML) workloads. Specifically, non-volatile memory-based PIM architectures have enabled effective ML acceleration due to their ability to perform energy-efficient matrix-vector multiplication operations. However, these devices suffer from non-idealities such as thermal noise. This noise alters the stored values in the memory cells which correspond to actual model weights, compromising the inference accuracy. In this work, we introduce ThRIve, a noise-aware training methodology that leverages low-rank adaptation to enable thermally robust inference on heterogeneous PIM architectures. ThRIve selectively stores these low-rank noise-aware parameters on a hardware that is less susceptible to thermal noise, enabling robustness against temperature-induced noise variations. ThRIve mitigates the effects of thermal-noise and prevent the drop in inference accuracy across the entire operating temperature range. Experimental results demonstrate that ThRIve-enabled architectures maintain consistent inference accuracy, with the mean accuracy staying within 2% of the ideal (i.e., noise-free) accuracy, and the variation in accuracy across the entire operating temperature range remaining within 2% of the mean. The proposed methodology achieves accuracy and robustness comparable to thermally-resilient Static Random-Access Memory (SRAM)-based PIM systems, while delivering up to 5.4x reduction in energy-delay product (EDP) during CNN model inferencing. --- ### Expressivity of Shallow Neural Networks Over Finite Fields Authors: Maksym Zubkov, Carol Wu, Shiwei Yang, Param Mody, Yifei Chen Published: 2026-07-19 Categories: math.AG, cs.NE arXiv: https://arxiv.org/abs/2607.17090v1 We study the expressivity of shallow polynomial neural networks (PNNs) with monomial activation functions over finite fields. For a given architecture, we define a neuromanifold as the image of the map from all possible network weights into the product of polynomial rings. We quantify the expressivity by the cardinality of the neuromanifold, and derive a natural lower and upper bound. This leads to counting rational points over finite fields, a problem closely linked to the Weil conjectures. Finally, we present an architecture that exhibits a striking difference in the neuromanifolds when considered over a characteristic zero versus a finite-characteristic field, illustrating the critical role of field characteristic in the notion of expressivity. --- ### OTAP: Structure-Aware Optimal Transport for Evaluating Planning and Execution in Agent Trajectories Authors: Babak Barazandeh, Subhabrata Majumdar, George Michailidis Published: 2026-07-19 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2607.17082v2 Large language model agents solve tasks by generating trajectories that interleave planning, tool calls, and intermediate results. Current evaluation metrics reduce such a trajectory to a binary success flag, compare it against a reference by exact matching, or delegate judgment to another language model. A success flag cannot distinguish a sound solution from one that succeeds by luck, and says nothing about why a failed run went wrong. Exact matching penalizes plans that are valid but reordered or decomposed differently from the reference. We reframe trajectory evaluation as a distance between the agent's execution graph and a set of valid solution graphs, and instantiate it via an unbalanced fused Gromov-Wasserstein transport problem over attributed dependency graphs. The resulting score, termed OTAP (Optimal Transport for Agentic Planning), is a pseudo-metric that is provably invariant to dependency-preserving reorderings and has bounded sensitivity to redundant steps. Its unbalanced marginals handle missing or hallucinated steps without forcing a match, and its soft coupling accommodates variation in plan granularity. On controlled perturbations and three public benchmarks, OTAP separates valid from invalid trajectories in a regime where semantics-only metrics score below chance. Its advantage tracks the fidelity of the dependency graph: largest where edges follow from operator semantics, smallest where they are inferred from free text. Where a formal verifier exists, strict surface metrics predict validity better than OTAP does, which places OTAP in open-ended domains where no verifier is available. --- ### ALLUDE: A Unified Evaluation System for Configurable Attacks in Differentiable Environments Authors: Mansi Phute, Alexander Greenhalgh, Matthew Hull, Haoran Wang, Alec Helbling, ShengYun Peng, Elliott Faa, Willian Lunardi, Martin Andreoni, Wenke Lee, Duen Horng Chau Published: 2026-07-19 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2607.17077v1 Adversarial attacks against vision models like object detectors are often evaluated under limited conditions, leaving their performance under-characterized. Bridging simulation and differentiable rendering enables more robust, end-to-end evaluation of these adversarial attacks, yet there is no easy-to-use, unified system that offers a rich set of customizable configurations for adversarial attacks across multiple scenes, objects, environmental and lighting conditions, and camera trajectories. We present ALLUDE, which addresses these gaps, offering first-of-its-kind evaluation capabilities across Linux and Windows. We comprehensively demonstrate ALLUDE's evaluation breadth through a two-pronged strategy: (1) using Latin Hypercube Sampling, we draw a representative subset from 5,400 configurations spanning 10 scene-object pairs, 9 weather conditions, 4 optimizers, 5 camera trajectories, and 3 detection models; (2) we stress-test existing attacks (CAMOU, RAUCA, FCA) under diverse weather conditions and continuous camera trajectories, revealing degradation of attack success across every attack, exposing evaluation gaps in prior work. Through ALLUDE's end-to-end differentiable rendering, adversarial attacks can be optimized against shifting real-world deployment conditions. Our cross-platform code is open source. --- ### A Systematic Evaluation of Traditional Privacy Policy Analysis Tools Against LLMs Authors: Madhav Aryal, Sudipa Saha, Sunil Manandhar, Anshuman Chhabra, Kaushal Kafle Published: 2026-07-19 Categories: cs.CR, cs.CL arXiv: https://arxiv.org/abs/2607.17075v2 The advent of LLMs has significantly changed the research on privacy policy and data compliance analysis by enabling tasks that previously required specialized, domain-specific tools. However, it remains unclear to what extent LLMs can truly replicate the diverse functionalities, and the wide range of methodologies and analysis offered by prior work. In this paper, we conduct the first systematic evaluation of whether off-the-shelf LLMs can replace specialized privacy analysis tools. We study six representative tools spanning three major functionalities: contradiction detection, regulatory compliance analysis, and privacy policy summarization and aggregation, and across three intermediate tasks: structured data extraction using tuples, Semantic Role Labeling (SRL) and manual privacy policy labeling. We compare the performance of two state-of-the-art LLMs (GPT-5.2 and Gemini-2.5 in various configurations) against the tools by directly prompting the models to perform corresponding functionalities and tasks on a custom dataset of 10 privacy policies, allowing us to assess whether off-the-shelf models can produce tool-specific functionalities without further engineering or domain-specific training, major limitations in prior work. Our results show that LLMs consistently match or exceed the capabilities of existing tools across the functionalities. In manual labeling of first-party collection entities, LLMs achieved an average precision of 81.8% and recall of 70.9%, while for labeling of third-party sharing entities, they achieved an average precision of 91.4% and recall of 70.8% compared to the OPP-115 dataset. Overall, our findings indicate that LLMs can effectively perform a broad range of functionalities and tasks in privacy policy and regulation analysis that previously required specialized tools. --- ### ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Authors: Pratyush Dhingra, Pramit Kumar Pal, Janardhan Rao Doppa, Partha Pratim Pande Published: 2026-07-19 Categories: cs.AR, cs.AI, cs.DC arXiv: https://arxiv.org/abs/2607.17074v2 Mixture of Experts (MoE) architectures have emerged as a dominant paradigm for scaling Large Language Models (LLMs). However, MoE inference on conventional hardware is constrained by three fundamental bottlenecks. These encompass the massive memory bandwidth required to fetch non-contiguous expert weights, the non-deterministic scatter-gather traffic generated by input-dependent token routing, and the tail-latency dependency imposed by synchronous expert output aggregation. To address these challenges, we propose ThAME, a three-dimensional (3D) heterogeneous multi-chiplet architecture for MoE inference. ThAME employs Ferroelectric Field-Effect Transistor (FeFET)-based non-volatile and DRAM-based volatile memory chiplets with a co-designed compute mapping strategy that aligns the distinct computational profiles of attention mechanisms and expert routing. Furthermore, we design a specialized Network-on-Chip communication backbone optimized to mitigate the bottlenecks associated with non-deterministic token routing traffic across the combinatorial space of input-dependent MoE traffic patterns. Experimental results demonstrate that ThAME outperforms state-of-the-art counterparts by up to 15.7x in terms of speedup and improves energy efficiency by up to 9.8x. --- ### Bridging the Information Gap: Semantic Densification and Hindsight Distillation for Cold-Start Prediction Authors: Hao Duong Le, Yifei Gao, Huan Li, Lun Jiang, Chen Bai, Ke Xing, Chen Zhang Published: 2026-07-19 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.17070v1 New-user cold-start is a critical bottleneck for e-commerce platforms: predicting user lifetime value (LTV) and conversion rate (CVR) for users with sparse interaction history. Two prior directions -- LLM-based semantic augmentation and learning using privileged information (LUPI) -- each face a key limitation. First, LLM augmentation produces unstructured rationales that are noisy and hard to operationalize in production. Second, naive student-teacher distillation can be brittle due to an information gap between the privileged teacher and the sparse student; moreover, this gap is heterogeneous across users. We propose SemRaD, a Semantic Reasoning-aware Distillation framework addressing both limitations. First, a Structured Semantic Reasoning Pipeline replaces free-form rationales with a structured schema built via a discover-curate-audit workflow, producing per user a Densified Semantic Profile (consumed by the deployed student via a Semantic-Gated Encoder that focuses on the most informative dimensions) and a Hindsight Distillation Target reconciled from pre- and post-conversion reasoning (used only at training). Second, to bridge this gap and handle its heterogeneity, a Hindsight-Aware Distillation Network transfers privileged knowledge via the hindsight target, with Distillation Experts improving transfer under per-user variability. On a large-scale industrial dataset, SemRaD lifts +1.9% LTV (Gini) and +1.0% CVR (AUROC) over a production-grade base; a four-week online A/B at Keeta confirms +1.0% LTV / +0.43% CVR. SemRaD also matches the production system's LTV using only 9% of the training data while improving CVR by 0.8%. --- ### AdvSerial: Physical Adversarial Attacks on Infrastructure-mounted Pedestrian Detectors via Semantic Feature Suppression Authors: Yuanhao Huang, Yilong Ren, Jinlei Wang, Xuesong Bai, Jinchuan Zhang, Haiyang Yu Published: 2026-07-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.17069v1 AI-based visual perception systems are increasingly deployed in infrastructure surveillance, including roadside monitoring units, highway cameras, and smart-city pedestrian management systems. The security vulnerability of these systems to physical adversarial attacks poses a direct threat to the reliable operation of transportation infrastructure. We propose AdvSerial, a dynamic 2D--3D joint optimization framework for generating continuous high-angle physical adversarial patches against pedestrian detectors in infrastructure-based scenarios. We UV-map a boundary-aware quilted texture onto 3D garments, combine 2D digital attacks with 3D sparse- and continuous-frame rendering, and explicitly suppress person-specific semantic features while enforcing temporal continuity. A Feature Smooth Quilting strategy reduces visible patch boundaries and bounds cross-seam feature discontinuities. A serial-frame loss encourages long uninterrupted sequences of detection failures. In physical world experiments, AdvSerial achieves a 74.8% attack success rate on YOLO-v5 and degrades mean detection confidence from 84.30% to 39.38%. Experiments spanning eight detectors with different architectures demonstrate strong transferability. Notably, it achieves an $89.71%$ attack success rate on YOLO-v2 and resists both patch-detection defenses (NapGuard) and 3D-temporal perception (Sparse4D-v3). The results reveal persistent, temporally consistent failure modes under high-angle surveillance, and motivate the design of motion-aware and 3D-aware defenses for security-critical infrastructure deployments. --- ### When LLMs Over-Answer: Measuring and Mitigating Quality Issues in LLM-Based Hardware Description Language Question Answering Authors: Ziteng Hu, Jiachi Chen, Wenhao Lv, Huan Zhang, Yingjie Xia Published: 2026-07-19 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.17063v2 The rapid advancement of large language models (LLMs) has led practitioners to increasingly rely on them for answering questions about hardware description languages (HDLs). Because HDL is ultimately synthesized into physical hardware, an imprecise or redundant answer can propagate into timing violations or non-synthesizable logic that surface only late in the design flow, making the quality of HDL answers especially consequential. However, the quality of LLM-generated responses, particularly in comparison with answers provided by human experts, remains unclear. To investigate this question, we collect 6,246 HDL Q&A posts with accepted answers from Stack Overflow and curate them into a dataset, organized into a taxonomy of four main categories (Conceptual, Debugging, Generation, and Optimization) and ten subcategories. Using this dataset, we design a user study conducted with 19 HDL engineers with one to three years of experience. Our findings reveal a pervasive over answering tendency: LLMs supply correct content but bury it under redundant alternatives (65.7%) and verbose padding (69.1%), while nearly half of answers (49.0%) fail to fully align with expert answers yet participants still preferred LLM responses for readability (58.3%). Motivated by these findings, we propose a multi-agent framework for improving LLM-based HDL question answering. We evaluate answer quality using an LLM-as-Judge and two structural metrics: the number of core answers, which reflects redundancy since LLMs often provide multiple alternative solutions, and the length of non-core content, which reflects verbosity. Evaluated on the four mainstream LLMs, our framework increases the average core-answer quality score from 3.71 to 4.67 (+0.96) and the non-core content quality from 3.72 to 4.23 (+0.51), on a five-point scale. --- ### What does a Bayes-filtered transformer believe? A predictive Monte Carlo approach Authors: Afiq Abdillah Effiezal Aswadi, Haotong Ma, Susan Wei Published: 2026-07-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.17060v1 A Bayes-filtered transformer (BFT) is a transformer trained on sequences that are generated in two steps: first a latent task is drawn from a prior, then observations are drawn conditional on that task. Trained under autoregressive log loss, the BFT's next-token prediction, in the idealized limit, is the Bayesian posterior predictive distribution (PPD) induced by that prior and that conditional law. In practice the trained BFT is only an approximation of this ideal PPD, raising an interpretive question: what prior and posterior over the latent task has the trained BFT actually internalized? Existing work answers this question by comparing the trained BFT's predictions against the predictions of various "reference" posteriors, each standing in for a different candidate algorithm or computation the BFT might be implementing. This prediction-space comparison is fragile: different posteriors can share the same posterior-mean predictions. We use predictive Monte Carlo (PMC) as a general interpretability tool for any BFT: using only next-token generation, PMC returns an approximation to the implicit prior and posterior over the latent task, answering the interpretive question directly in latent space. We apply PMC to three stylized task families spanning 0-Markov and 1-Markov exchangeability. The phenomena previously reported in these settings remain visible in latent space. Code is available at https://github.com/afiq-aswadi/bft-pmc --- ### DROID-ANCHOR: Odometry-Anchored Recurrent Metric Depth Estimation Authors: Yuxuan Chen, Brook Du Published: 2026-07-19 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2607.17058v1 Precise metric depth estimation is fundamental for autonomous robot navigation, yet monocular systems inherently suffer from scale ambiguity and scale drift. While recent recurrent flow-based SLAM systems have demonstrated state-of-the-art robustness, they remain scale-ambiguous. In this paper, we propose Metric-DROID, an end-to-end recurrent architecture that anchors visual SLAM to physical reality by integrating proprioceptive odometry. Our framework introduces the following innovations: (1) A LSTM Update Operator that encodes high-frequency odometry sequences into spatial feature maps, providing a persistent metric bias for iterative refinement. (2) An Uncertainty-Aware Metric Backend ($BA_{odom}$) that treats odometry as a geometric anchor with learned heteroscedastic covariance. By regressing a time-varying metric uncertainty $Σ_{o}$, our system intelligently balances visual re-projection and metric translation residuals, effectively mitigating the impact of wheel-slip and sensor noise. (3) We further propose a selective residual fine-tuning strategy to preserve pre-trained geometric priors while enabling zero-shot metric alignment. --- ### High-Capacity Robust Watermarking Technology for High-Resolution Images Authors: Shaowu Wu, Wei Lu, Qing Qian, Mingsen Deng Published: 2026-07-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.17056v1 Most existing watermarking techniques are primarily designed for low-resolution images, with few methods tailored for high-resolution images. Moreover, the embedding capacity is often limited to fixed lengths (e.g., 30, 100, 256 bits, etc.), which struggles to meet practical demands. To address these issues, this paper proposes a high-capacity robust watermarking method for high-resolution images, capable of embedding a watermark of 4 KB (32,768 bits) into images with a resolution of 1024*1024, achieving an embedding rate of 0.0313 bpp. Specifically, this paper adopts a block-wise strategy to effectively embed the watermark into local regions, enabling the network to train and learn normally even under low-resource conditions. The encoder and decoder structures respectively employ a reversible symmetric architecture with three convolutional and three deconvolutional layers, ensuring consistency in the coupling and decoupling of the watermark and image features. Additionally, the loss function combines global and local losses with weighted contributions. By incorporating constraints on the visual quality and robustness of local block regions, the overall imperceptibility and robustness of the image are further enhanced. Extensive experimental results verify that the proposed method is effective and feasible in high-resolution image scenarios with high-capacity watermarking, while demonstrating strong robustness against various noise attacks. --- ### Searching for Task-Specific Vision Paths: Evolutionary Block Pruning Across Vision-Language Models Authors: Tarun Tomar Published: 2026-07-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.17052v1 Vision-language models normally execute the same complete vision encoder for every question, even when OCR, counting, object, attribute, and spatial queries may not require identical computation. We study whether fixed-budget combinations of vision blocks can be skipped without fine-tuning. A shared K-block route skips one searched set of exactly K blocks for every question, while a capability-specific K-block policy selects one same-size route using a known capability label. We introduce a source-balanced evolutionary search and compare it with independent ranking, contiguous removal, and random routes at matched budgets. Experiments use Qwen2.5-VL-3B-Instruct, SmolVLM2-2.2B-Instruct, and an 876-example image-disjoint selection split. Search transfers across architectures: on SmolVLM2, the searched shared four-block route beats independent construction by 4.91 percentage points. Capability specialization is less stable. On Qwen, the six-block capability policy beats the shared route by 2.17 points, driven by a 7.10-point OCR gain. On sealed IIIT5K, however, the SmolVLM2 OCR-specific route trails its shared route by 13.6 points. Combinatorial search reliably improves route construction, but capability labels do not define universally transferable vision pathways. --- ### EvoGUI: An Evolution-Aware Benchmark for GUI State-Transition Understanding Authors: Yaohan Yang, Minglei Shi, Borui Zhang, Jie Zhou, Jiwen Lu Published: 2026-07-19 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2607.17050v1 GUI agents must reason about how actions transform interface states, but end-to-end success rates entangle this ability with perception, grounding, planning, and recovery. We introduce EvoGUI, a diagnostic framework that converts normalized GUI trajectories into three complementary visual question answering probes: temporal ordering, inverse action/value prediction, and contrastive one-step successor discrimination. Their labels are derived from trajectory order and logged actions, requiring no additional task-label annotation after trajectory normalization. We instantiate EvoGUI-Bench from Mind2Web and WebLINX, yielding 3,000 instances across 120 domains, and evaluate 28 vision-language model configurations zero-shot. The strongest model reaches only 60.4 EvoGain, while model scale and GUI specialization do not reliably predict performance. These results establish EvoGUI-Bench as a scalable diagnostic complement to end-to-end GUI-agent evaluation while exposing substantial headroom in state-transition understanding. The source code is publicly available at https://github.com/Yyhhh6/EvoGUI. --- ### Solver-Hard Is Not Model-Hard: A Hardness-Controlled Diagnostic for LLM Constraint Reasoning Authors: Lucky Verma Published: 2026-07-19 Categories: cs.LG, cs.AI, cs.LO arXiv: https://arxiv.org/abs/2607.17047v1 LLM constraint reasoners are often evaluated near the random-SAT phase transition, confounding density and solver hardness. We test instance-level transfer while near-matching clause density. At aligned size bins, with near-matched density and matched maximum clause width, we compare proof-hard expander-Tseitin and proof-easy ladder-Tseitin formulas, pigeonhole anchors, and density-mismatched controls. Theory separates their resolution hardness; a solver-specific Glucose mean-conflict proxy differs by up to $51\times$, and five other solvers preserve the direction. Across three included models (243 instances each; a fourth is excluded for abstention), the near-matched-density accuracy gaps range from $-32$ to $+20$ points, with a pooled gap of $+1.7$ points ($p=0.74$) and a wrong-signed correctness-versus-conflict association ($r=+0.15$). A proof-preserving relabeling lowers accuracy in all five clusters for one model (mean $-93$ points) but not another, exposing model-surface sensitivity. In a preregistered extension, provider-reported completion-token spend does not consistently increase with the proxy after accounting for formula length and censoring. At 16k, the reasoning model spends more on proof-easy matched formulas and exhausts its budget on the solver-easiest UNSAT family; the 32k C1 gap is absent. These scoped dissociations concern verdict accuracy and observed token spend, not certificate solving, exact proof length, or allocation efficiency. --- ### Where Does Agent Reliability Come From? A Cross-Benchmark Decomposition of Verification Loops, Specialist Models, and Scaffolding in a Production Enterprise Agent Authors: Arunabh Dastidar Published: 2026-07-19 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2607.17044v1 Multi-step enterprise agent tasks fail in a characteristic way: single-pass inference has no checkpoint between deciding an answer and committing to it. We study one production system (Leni) whose architecture installs such checkpoints: verification loops (execute, observe, compare, correct) staffed by lightweight task-specialized post-trained models. We evaluate the unmodified production configuration on three public benchmarks stressing distinct failure modes: SpreadsheetBench Verified (silent computation error), BullshitBench v2 (premise confabulation), and the GAIA validation split (cascade error over long tool chains). The full system improves over its frontier base model by +11.0 percentage points on SpreadsheetBench (91.25% vs 80.25%, n=400, p<0.001), +7 to +10 percentage points on BullshitBench (98% vs 91%, n=100), and roughly +15 points on GAIA validation (75.2% pass@1, n=165; 83.0% best-of-k). Our central contribution is a decomposition of that uplift: most of it comes from scaffolding, routing, and specialist models rather than from the verification step itself, whose isolated contribution is small (+1.5 points) but concentrated at the top of the score distribution, where it converts otherwise-failing tasks. We instrument the loop end-to-end, yielding an empirical verifier confusion matrix (catch rate about 0.20, fix rate 0.75, no false-alarm regressions) that grounds a compounding-reliability model. Specialist-swap ablations suggest that the loop's value depends on who observes it: replacing the small trained verifier with the generating frontier model eliminates most rescues. A valid-premise control shows zero over-rejections in 100 expert-level questions. --- ### Learning from Synthetic Data without Model Collapse in Iterative Instruction Tuning Authors: Xiaonan Luo, Yue Huang, Kehan Guo, Ping He, Chuan Zou, Ting Hua, Xiangliang Zhang Published: 2026-07-19 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.17043v1 Model collapse is a central challenge in learning from synthetic data: as later-generation large language models (LLMs) are trained on an increasing proportion of model-generated data, performance can degrade due to narrowed coverage and accumulated bias. Existing work mainly studies how to bound this degradation. In iterative model evolution, however, the more meaningful objective is to ensure that each successive model improves over its predecessor, which requires diagnosing collapse at a granularity that is actionable for data curation. We study this problem in synthetic data self-improving for instruction tuning. We show that collapse in this setting is not simply uniform performance degradation, but can appear as a polarization of competence, where synthetic training reinforces already strong skills while further degrading weak ones. Motivated by this observation, we propose KITE (Knowledge-boundary Instruction Tuning via Exploration), a two-stage framework that combines failure-guided data generation with boundary-aware uncertainty curation. Experiments across several datasets and multiple open-source LLMs show that KITE yields more stable improvement than strong synthetic-data baselines. --- ### Articulated Humanoid Head for a Robot Receptionist Capable of Natural Human Interaction Authors: Tharusha Fonseka, Charuka Bandara, Moshintha Hewavitharana, Melisa Arukgoda, Wageesha N. Manamperi, Udaya S. K. Perera Miriya Thanthrige, Peshala Jayasekara Published: 2026-07-19 Categories: cs.RO, cs.CV, eess.IV arXiv: https://arxiv.org/abs/2607.17042v1 Humanoid robots have become increasingly popular in applications such as social interaction, education, and service roles, which drives the need for more natural and efficient human-robot interactions. However, currently available humanoid heads often face limitations, including high costs, mechanical complexity, and limited adaptability across diverse environments. To address these challenges, we present an articulated humanoid robot head designed for a receptionist role, integrating a mechanical structure with 21 degrees of freedom (DoF), including mechanisms for the mouth, eyes, eyebrows, and neck, and covered with realistic silicone skin to achieve a human-like appearance and expression. The system integrates a model-based architecture that combines SCRFD, ArcFace, and ByTetrack for face recognition and Llama and Whisper for natural language processing, with hardware support enabling real-time operations and human re-identification. The conversational ability and re-identification capabilities of the humanoid robot head were quantitatively measured, while its emotional expressiveness and human likeness were evaluated through a user study, achieving an average human likeness score of 4.13 out of 5. --- ### Reward-Driven LLM Agent Workflows: Synthesizing POMDP Routing and Self-Correction for Autonomous Decision-Making Authors: Amez Amanj Ali, Kuo-Kun Tseng Published: 2026-07-19 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.17038v1 This paper addresses key technical challenges in current large language model (LLM) agent applications, including long-horizon planning, sparse reward attribution, and dynamic environmental interaction, by designing and optimizing an intelligent agent workflow. The proposed architecture is based on the synthesis of core AI paradigms: Visual, Language, Generative, Graph, Multimodal, Reinforcement, and Agent Intelligence. Unlike conventional baseline models that rely on static prompting and lack robust perception-action loops, our approach introduces a Partially Observable Markov Decision Process (POMDP) routing mechanism. This mechanism is augmented with an internal, self-correcting reward model that evaluates decision trajectories before execution. By integrating multimodal inputs and advanced reinforcement learning principles (such as proximal policy optimization and value function approximation), the agent maintains long-term structural memory and dynamically adapts its reasoning pathways to mitigate error accumulation. Empirical experiments on the ALFWorld embodied simulation environment and the WebShop online navigation benchmark demonstrate a 24.5% absolute improvement in task success rate and trajectory efficiency over mainstream baselines like the standard ReAct framework. Comprehensive ablation studies confirm the significant contribution of the reward-driven critique module in suppressing hallucination rates. This research bridges theoretical foundations of reinforcement learning and graph-based memory with autonomous agent workflows. Ultimately, the resulting architecture offers a practical, scalable reference framework for developing artificial intelligence technologies in complex, multi-step autonomous systems. Code is available at https://github.com/01Amez/RLAW_Implementation. --- ### Apeliotes: A Diffusion-Based Modeling Framework for km-scale Multi-Level Atmospheric Fields Authors: Evangelia Rafaela Frastali, Achyut Paudel, Maryam Golbazi, Frank Liu Published: 2026-07-19 Categories: cs.LG, cs.CE arXiv: https://arxiv.org/abs/2607.17037v1 High-resolution atmospheric data are required to resolve mesoscale and localized meteorological structures, however such datasets remain limited in many regions of the world. Existing high-resolution weather products are typically produced through dynamical downscaling, which is computationally expensive and difficult to scale across locations, variables, and forecast scenarios. These limitations motivate machine-learning-based downscaling systems that can generate multiple weather variables stochastically while producing new high-resolution fields directly. In this paper we present Apeliotes, a framework for high-resolution weather forecasting. Built on the global re-analysis atmospheric data, a pre-trained global weather foundation model, and a regionally trained generative diffusion model, Apeliotes not only provides accurate kilometer-scale weather variables, but also multi-level atmospheric fields which are not directly available in the existing global atmospheric data. Our comprehensive evaluation demonstrates that Apeliotes achieves highly competitive performance. The model predicts vertical wind profile with less than 3\% error between truth and predicted fields, achieving correlations of 0.91 for 10-m wind speed and 0.99 for 2-m temperature, with NRMSE values of 0.42 and 0.17, respectively. --- ### ChemFusion: A Multimodal Cross-Attention Network for Reaction Yield Prediction Authors: Qiwei Han, Chi Zhou Published: 2026-07-19 Categories: cs.LG, physics.chem-ph arXiv: https://arxiv.org/abs/2607.17033v1 Forecasting the outcomes of transition-metal-catalyzed reactions is notoriously complex due to the interplay of diverse physical and chemical variables. A persistent computational bottleneck has been effectively merging broad electronic descriptors with the localized, three-dimensional geometry of the reactive site. To bridge this representation gap, we present ChemFusion, a hybrid neural network that fuses conventional electronic features with explicit 3D atomic coordinates. Using a cross-attention mechanism, the model enables global electronic states to dynamically attend to specific spatial constraints within un-pooled molecular point clouds. When benchmarked against a diverse library of cross-couplings, this approach delivers exceptional predictive performance, decisively surpassing traditional single-modality frameworks. Importantly, extracting the attention matrices reveals that the architecture autonomously learns to identify and penalize restrictive steric hindrances. This provides a physically grounded interpretability, demonstrating that spatially aware networks can navigate complex reaction sterics that standard statistical models typically miss. --- ### Interpretable Machine Learning for Air Pollution and Respiratory Health Prediction: A Socioeconomic Subgroup Analysis Authors: Maede Azani Hassan Abadi, Shouyi Wang Published: 2026-07-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.17024v1 Air pollution and climate-related stressors are increasingly important concerns for respiratory health, especially in settings with unequal environmental exposure and healthcare capacity. This study evaluates an interpretable machine learning framework for predicting respiratory disease rates and air-quality status using structured country-level weekly data. Two supervised learning tasks were considered: regression of respiratory disease rate per 100,000 population and binary classification of air-quality status. Nine regression models and nine classification models were compared using nested cross-validation. Model interpretation was conducted using SHAP values, and subgroup analysis was performed across income levels and geographic regions. The results showed that PM2.5 concentration was the dominant predictor of respiratory disease rate, with linear and regularized linear models achieving the strongest regression performance. For air-quality classification, models achieved high balanced accuracy when PM2.5 was included, but performance decreased substantially when PM2.5 was removed, indicating strong dependence on pollutant-related information. SHAP analysis showed that, without PM2.5, socioeconomic and meteorological variables such as GDP per capita, precipitation, and healthcare access became more influential. Subgroup analysis showed similar aggregate regression error across income groups, but PM2.5 contributed more strongly to predictions in lower-middle-income countries. These results show that model accuracy alone is not sufficient for climate-health prediction. Interpretable models can help identify dominant pollution-related signals, test whether results depend on key pollutant variables, and show whether prediction patterns differ across socioeconomic groups. --- ### Regularize or Localize: When Training-Time KV-Cache Geometry Pays Under Quantization Authors: Libo Sun, Po-Wei Harn, Zewei Zhang, Peixiong He, Xiao Qin Published: 2026-07-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.17019v1 We study whether \sigreg -- LeJEPA's anti-collapse objective -- can reshape representations during standard autoregressive language-model pretraining, and when the resulting geometry helps \kv-cache quantization. We train 110M-parameter models on 10B FineWeb tokens and report three findings. \textbf{(1)} At $λ{=}0.01$, \sigreg reduces hidden-state pairwise-cosine anisotropy by $38\%$ across three paired seeds. Perplexity increases by less than $0.35\%$ in every pair, with no consistent zero-shot loss. \textbf{(2)} This change does not propagate from hidden states to the \kv cache. Applying \sigreg directly to K and V during continued training, however, reduces mean cache anisotropy by $94\%$ across four checkpoints. A matched continuation without the \kv term leaves cache geometry nearly unchanged, and the frozen-trunk retrofits we tested do not reproduce the effect. \textbf{(3)} Under untransformed symmetric group-free quantization, direct \kv regularization is the only training condition that prefers per-channel scaling in all three seeds, and under that same 3-bit per-channel scheme the baseline incurs $4.3$--$7.9\times$ the directly regularized model's \dnll. Under the full simulated KIVI-style configuration (mixed arrangement, zero-points, grouped scales), however, all models reach near-parity, including when storage overhead is approximately matched. In this 110M regime, the training intervention helps when quantizer scales are coarse; the advantage vanishes under the tested combination of token-local grouping, mixed \kv scaling, and zero-points. To our knowledge this is the first training-time \emph{distributional} regularization of standard \kv-cache geometry evaluated against post-hoc cache quantization. --- ### Robust Chance-Constrained Optimization using a Continuous Parameter Space Wasserstein-2 Ambiguity Set of Gaussian Mixtures Authors: Shibshankar Dey, Sanjay Mehrotra Published: 2026-07-19 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2607.17018v2 We study distributionally robust linear chance-constrained problems in which uncertainty is modeled by a Gaussian mixture model (GMM). Finite-support distributionally robust (FDR) formulations, widely used in data-driven robust optimization, robustify over empirical mixture support points and therefore primarily stress-test the fitted nominal mixture. This can be insufficient when service reliability depends on structural misspecification of the nominal mixture-support parameters. To address this limitation, we describe the ambiguity set of distributions by developing a novel formulation of a Wasserstein-2 metric that uses the Bures-Wasserstein (BW) metric over probability measures with finite second moments. Unlike FDR, which generally sets finitely many empirical support points a priori, the proposed ambiguity set allows the worst-case distribution to endogenously determine both how many mixture components receive mass and where their means and covariances lie within a continuous support. For the resulting ambiguity set, under mild regularity conditions, we prove strong duality for the inner worst-case chance-constraint problem and derive its semi-infinite reformulation. We then develop an adaptive cutting-surface algorithm, which endogenously determines the locations of mixture components receiving mass, and the mean and covariances of the Gaussian distributions at these locations. The algorithm attains any prescribed optimality gap in finitely many iterations, while a block-alternating local search identifies new components. A case study using the electric-vehicle charging-station energy-allocation problem demonstrates the framework's practical value in achieving any reliability targets. CDR also induces structural changes in energy allocations, unlike FDR, whose allocations remain close to the nominal solution. --- ### Robust Chance-Constrained Optimization using a Continuous Parameter Space Wasserstein-2 Ambiguity Set of Gaussian Mixtures Authors: Shibshankar Dey, Sanjay Mehrotra Published: 2026-07-19 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2607.17018v1 We study distributionally robust linear chance-constrained problems in which uncertainty is modeled by a Gaussian mixture model (GMM). Finite-support distributionally robust (FDR) formulations, widely used in data-driven robust optimization, robustify over empirical mixture support points and therefore primarily stress-test the fitted nominal mixture. This can be insufficient when service reliability depends on structural misspecification of the nominal mixture-support parameters. To address this limitation, we describe the ambiguity set of distributions by developing a novel formulation of a Wasserstein-2 metric that uses the Bures-Wasserstein (BW) metric over probability measures with finite second moments. Unlike FDR, which generally sets finitely many empirical support points a priori, the proposed ambiguity set allows the worst-case distribution to endogenously determine both how many mixture components receive mass and where their means and covariances lie within a continuous support. For the resulting ambiguity set, under mild regularity conditions, we prove strong duality for the inner worst-case chance-constraint problem and derive its semi-infinite reformulation. We then develop an adaptive cutting-surface algorithm, which endogenously determines the locations of mixture components receiving mass, and the mean and covariances of the Gaussian distributions at these locations. The algorithm attains any prescribed optimality gap in finitely many iterations, while a block-alternating local search identifies new components. A case study using the electric-vehicle charging-station energy-allocation problem demonstrates the framework's practical value in achieving any reliability targets. CDR also induces structural changes in energy allocations, unlike FDR, whose allocations remain close to the nominal solution. --- ### WHALE: A Scalable Unified Model for Recommendation with Wukong-HSTU Architecture Authors: Renqin Cai, Dawei Sun, Yuanjun Yao, Zhiyong Wang, Velvin Fu, Maggie Zhuang, Yu Shi, Zhongnan Fang, Xuan Cao, Jing Qian, Rui Li Published: 2026-07-19 Categories: cs.IR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.17017v2 As scalability becomes increasingly important in recommendation modeling, recent architectures have advanced the modeling of two broad sources of ranking signals along separate paths: non-sequence features, including user, item, context, and cross features; and sequence features from user behavior histories. Wukong and HSTU have emerged as representative scalable backbones for these paths: Wukong scales high-order non-sequence feature-interaction modeling, while HSTU scales long user-behavior sequence modeling. Despite their complementary strengths, practical architectures that combine these two types of feature modeling remain underexplored. We present WHALE, a scalable unified recommendation architecture that jointly models non-sequence and sequence features on top of Wukong and HSTU. Each WHALE layer contains a Wukong module, an HSTU module, and an attention-based fusion module in which Wukong-derived interaction representations query HSTU-derived behavior representations. This design keeps both backbones active throughout the network and enables progressive Wukong-HSTU exchange, allowing high-order feature crosses to repeatedly retrieve fine-grained evidence from long user histories. To make WHALE practical for industrial deployment, we introduce customized Triton kernels and other model-systems co-design techniques to improve training and inference efficiency. On large-scale industrial recommendation data, WHALE achieves consistent gains in offline experiments. Additionally, it delivers positive online gains with a modest serving-throughput trade-off. The method has been deployed in production systems. Overall, WHALE provides a practical example of how these two sources of information can be scalably unified in an industrial recommendation model. --- ### Alignment of a Total Automation Economy Authors: David McAllester Published: 2026-07-19 Categories: physics.soc-ph, cs.AI arXiv: https://arxiv.org/abs/2607.17015v2 We consider economic theory from the perspective of a total automation economy, one with no human involvement in production either in manufacturing or in management. One can naturally ask whether a total automation economy is fundamentally a centrally planned economy or, alternatively, whether efficiency demands decentralization into local decisions by competing agents -- agentic production. A soviet economist, Leonid Kantorovich, developed linear programming as a method companies or governments can use to optimize production. Ironically, he is also generally credited with showing that the most efficient production is achieved through decentralization -- a free market economy with competing agents. Here we review Kantorovich's dualization in detail. We take the objective of the economy to be maximizing production weighted by (human) market price. A fundamental issue is whether an automated pursuit of this objective might have alignment vulnerabilities as the economy evolves. Another question is whether dualization provides insight into the utility of agentic AI systems (multi-agent AI systems) generally. --- ### Modality Discrepancy Transformer for Ambivalence and Hesitancy Recognition Authors: Shiyu Luo, Yu Wang, Jiawen Huang, Zhaoxiang Xiao, Chenxi Huang, Qi Zhang, Bin Liu Published: 2026-07-18 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2609.19148v1 Ambivalence and hesitancy (A/H) are affective states in which individuals express contradictory signals across facial, vocal, and linguistic channels. Automatically recognising A/H in clinical videos requires detecting cross-modal disagreement -- the signal that standard fusion methods suppress. Based on the conflict-aware multimodal fusion framework of Bekhouche et al., we present the Modality Discrepancy Transformer (MDT). MDT enriches the original 6-token design to a 9-token representation comprising three modality embeddings, three absolute-difference features, and three Hadamard-product discrepancy features learned through linear projections. These nine tokens undergo Transformer self-attention, with FiLM-based text-conditioned modulation and LoRA fine-tuning as core architectural components. A text-guided late fusion branch blends a text-only auxiliary head with the full multimodal output at inference. On the BAH dataset from the 3rd ABAW Challenge, MDT achieves 0.7408 Macro F1 on the labelled test split and 0.7368 on the private leaderboard, outperforming the strongest published baseline by over 10 points while training in under 20 minutes on a single GPU. --- ### TwinMark: A Unified Watermark for Provable Survival Under Feature and Logit Distillation Authors: Redwanul Karim, Tobias Feigl, Christopher Mutschler, Felix Ott Published: 2026-07-18 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.19011v1 We propose TwinMark, a watermarking scheme that reads a single SHAKE128 secret through two complementary linear functionals of model-output summaries: a covariance projector against the carrier-set covariance (cov-Feat) and a class-conditional Fisher-aligned linear carrier decoded from class-mean logits (cc-FALC). The two readouts share one bit vector and cover the two extraction surfaces of a deployed vision model: a classifier API attacked by KL knowledge distillation (KD) (Std. KL-KD), and a representation-only host attacked by feature-matching KD (FM-KD). Each readout admits a teacher-measurable a posteriori certificate that lower-bounds post-distillation detection power, and the two channels combine under a regime-restricted OR rule whose test statistic (calibrated null or bit vote) is selected by the exposed surface. cov-Feat admits a rank-blind operator-norm certificate, cc-FALC admits a centered-logit-gap certificate that decouples bit capacity from class count: at K=1024 in m=100 classes (a 10.24x over-encoding), the bit-vote attains z=23.0 sigma at a teacher-accuracy cost of +0.9+-0.2%p. Across 13 attacks on CIFAR-10, CIFAR-100, and Mini-ImageNet, TwinMark verifies on every cell whose post-attack model retains task utility, survives cross-architecture distillation onto ResNet-18/50, VGG-16, and MobileNet-V3, and ports to GNSS few-shot, VOC detection, ISIC segmentation, and STL-10 SimCLR. --- ### WordPolo: Evaluating Language Models Through Iterative Semantic Feedback Authors: Tyler McDonald, Ali Emami Published: 2026-07-18 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.19006v1 Large Language Models (LLMs) and Large Reasoning Models (LRMs) are typically evaluated on challenging benchmarks through dataset accuracy alone, providing no insight into the quality or faithfulness of their reasoning processes. We present WordPolo, a word-finding task where participants must discover an unknown target word using semantic similarity feedback. Players start with zero knowledge, make guesses, and receive distance scores (1 = correct, higher = further away). Success requires interpreting scores to navigate semantic space and systematically narrow the search. This design makes iterative reasoning and adaptive search strategies both directly observable and necessary for success. We evaluate recent LLMs (GPT-4.1, Llama 4, Claude 3.5 Haiku, Qwen 3), LRMs (o4-mini, Deepseek-R1), humans, and a novel heuristic on 1,500 puzzles. Beyond solve rates (which range from 4% to 62%), we introduce progression-based metrics that reveal models often make meaningful progress, insights that accuracy alone would miss. Our analysis shows how reasoning models can be hindered by overthinking and underthinking, while successful models exhibit human-like strategies. WordPolo demonstrates the need for benchmarks that test both reasoning process and outcomes, providing holistic measurements of model capabilities. Our code and dataset can be found at https://wordpolo-demo.vercel.app/. --- ### Code Consistency Preference Optimization Verification for Language Model Alignment Authors: Yunlong Tan, Mingqiao Mo, Hao Zhang Published: 2026-07-18 Categories: cs.SE, cs.CL arXiv: https://arxiv.org/abs/2609.19002v1 Execution-based verification enhances large language models' mathematical reasoning through computational soundness and dependency-aware filtering. However, prior preference optimization methods relying on Bradley-Terry reward models fail to capture the logical dependencies and execution consistency needed for scientific tasks. We propose a method that generates computationally sound solutions with dependency graphs for execution-consistent preference optimization. We first build a scientific reasoning dataset using UltraFeedback prompts, model generations, verification, and consistency results. Then we extract reasoning step expressions, prerequisites, and derivability relationships to construct dependency graphs and compute execution consistency scores. These scores are appended to each step, creating paired training data. Fine-tuning Llama-3-8B and DeepSeekMath-7B yields significant gains: +17.0% on MATH and +15.1% on GSM8K. Extending our Scientific Feasibility Control framework achieves 50.1% accuracy on PhyX multimodal physics reasoning, surpassing DeepSeek-R1 (49.8%) and OpenAI o3-mini (48.2%), with 91.7% scientific validity coverage at alpha=0.10 and 73% fewer scientific law violations, resulting in the CCPO model family. --- ### When Literature Data Mislead Artificial Intelligence in Materials Discovery Authors: Qian Wang, Ying Li, Ryuhei Sato, Hidemi Kato, Shin-ichi Orimo, Hao Li, Eric Jianfeng Cheng Published: 2026-07-18 Categories: cs.IR, cond-mat.mtrl-sci, cs.CE, cs.LG arXiv: https://arxiv.org/abs/2609.01621v1 Artificial intelligence (AI) increasingly treats scientific literature as a data source for building databases, training predictive models, and guiding discovery. Yet literature-derived datasets often assume that reported experimental values are internally consistent and directly reusable. Here, we analyze this assumption using solid electrolyte (SE) conductivity data as a representative materials-science case. By tracing values from source articles to curated datasets, we identify recurrent text-figure mismatches, ambiguous axis annotations, unit inconsistencies, and missing measurement context. These discrepancies are often numerically plausible and therefore difficult to detect through routine preprocessing, but they can propagate as structured label noise during database construction and machine-learning reuse. A cross-database example shows how ambiguous reporting can create a 100-fold conductivity error. Our analysis reframes data accuracy as an infrastructure requirement for artificial-intelligence-driven discovery and motivates traceable reporting, curation, and validation practices for reusable scientific data. Keywords: AI for science; Data reliability; Scientific databases; Structured label noise; Literature-derived data; Materials informatics; Solid electrolytes --- ### Gurukul AI: An Interactive AI-Driven Educational Platform for Indian Education System Authors: Isha Narang, Sneh Gosai, Mayank Singh Published: 2026-07-18 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2608.28611v1 Recent advances in large language models (LLMs) like ChatGPT and LLaMA have transformed AI-driven education, but these systems are predominantly trained on Western-centric data, making them ill-suited for regional curricula like India's. The Indian education system is linguistically diverse, exam-oriented, and structured around standardized syllabi, not addressed by existing datasets or tools. In this work, we curate a syllabus-aligned QA dataset based on NCERT (National Council of Educational Research and Training) textbooks for classes 9-12, capturing the content, context, and teaching style of Indian curricula. The final dataset, comprising 18,720 question-answer pairs across five subjects, is publicly available at https://huggingface.co/datasets/LingoIITGN/Gurukul. We fine-tune the LLaMA 3.1 8B model using this dataset and deploy it in a Retrieval-Augmented Generation (RAG) framework tailored to educational needs. We introduce GurukulAI, an open-access platform that enables Indian students to chat with the model, get doubts cleared, practice exam-style questions, receive contextual answers, and interact in both English and Hindi. By localizing AI for Indian classrooms, our work bridges the gap between global LLM capabilities and regional educational demands. The code is available at https://github.com/lingo-iitgn/GurukulAI. --- ### Local AI pre-screening for human triple-blind peer review in health sciences Authors: Rodrigo Martins Boos Published: 2026-07-18 Categories: cs.CY, cs.AI, cs.DL arXiv: https://arxiv.org/abs/2608.14625v1 Academic peer review is under mounting strain: NeurIPS 2025 received 21,575 submissions, ICLR 2025 received 11,603, and ICML 2025 received 12,107. This volume has outpaced the supply of qualified reviewers, and large language models (LLMs) are already filling the gap, largely undisclosed. An independent analysis of ICLR 2026 found roughly 21% of its 75,800 peer reviews were fully AI-generated, with over half showing some AI involvement (up from 15.8% in 2024). Documented risks include hallucinated citations in accepted papers and hidden prompt-injection instructions embedded in manuscripts to manipulate AI reviewers into favorable assessments. We propose a triple-blind, multi-LLM pre-screening framework for peer review, developed for a health sciences journal, that formalizes and discloses AI involvement while preserving human reviewers as the final decision-making authority. The framework routes a submission through five stages -- sanitization/anonymization, parallel AI pre-screening, an automated check gate, blinded human review, and editorial adjudication -- with return-to-author loops at the check and editor stages. Addressing the confidentiality concerns behind NIH/NSF bans on submitting unpublished proposals to third-party generative AI, all three AI reviewers run on locally-hosted, open-weight LLMs, keeping manuscript content within the journal infrastructure. The closest precedent, Shen et al., benchmarked five open-source LLMs on quartile classification of 200 manuscripts and found accuracy insufficient (35% exact-match) for autonomous use, supporting our decision to retain mandatory human adjudication. This transparent, human-supervised design offers a defensible alternative to today's opaque, unregulated AI use in peer review, potentially reducing the substantial delay of traditional review (avg. 13 weeks to first decision) without displacing human judgment. --- ### MIME: Multimodal Interactive Motion Encoder Authors: Addison Zucek, Prerit Gupta, Kamila Kuatova, Aniket Bera Published: 2026-07-18 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2607.22702v1 Text-motion representation learning has advanced rapidly, with growing interest in multi person interactions for animation, AR/VR, and embodied AI. These settings require representations that align language with both individual actor dynamics and the relationships between actors. We introduce the Multimodal Interactive Motion Encoder (MIME), which, to our knowledge, represents the first dedicated multimodal encoder designed specifically for two person interactive motion. MIME captures individual and shared structure using stream based co-attention with explicit interaction features and curriculum based contrastive training. On Inter-X text-motion retrieval, MIME consistently outperforms early and late fusion baselines across gallery sizes, achieving a 12.8% relative improvement in text-to-motion R@1 at a 2,000-sample gallery. We further evaluate MIME as a frozen auxiliary prior within TIMotion and InterMask on the unseen InterHuman dataset. MIME improves semantic alignment metrics while maintaining comparable FID in TIMotion. These results show that interaction aware multimodal encoding improves multi person motion retrieval and transfers across datasets to support downstream motion generation. --- ### RoleMix: Unifying Sequential and Non-Sequential Features via Semantic Tokenization for Post-Click Conversion Rate Prediction Authors: Wenan Wang, Qin Zhao, Zhixiang Lu Published: 2026-07-18 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.22700v1 Post-click conversion rate (PCVR) prediction is central to industrial recommendation, but remains challenged by the structural mismatch between sparse, unordered multi-field features and long, domain-specific behavior histories. Existing models often process these signals through separate pathways and fuse them late, weakening semantic roles and limiting cross-signal refinement. We propose RoleMix, a unified interaction architecture that represents sequential and non-sequential evidence through a shared, role-preserving token interface. Non-sequential fields are converted into explicit semantic tokens that preserve user, item, pairwise, dense, contextual, and cross-feature roles, while long behavior domains are compressed into item- and context-aware sequence-query tokens through two-stage hierarchical window attention. The resulting global, semantic, and sequence-query tokens are jointly refined by stacked UniMixing-Lite blocks for PCVR prediction. On the large-scale KDD Cup 2026 Tencent UniRec Challenge, RoleMix achieves 83.648% online AUC, outperforming the official industrial baseline by 1.953%. Ablation studies show that semantic tokenization yields the largest isolated gain, highlighting a key principle for large-scale PCVR modeling: preserving field semantics at the token-interface level is as important as scaling the interaction backbone. --- ### Similarity All The Way Up: Multilingual Generalization in LLMs Relies on Language-Level Similarity Structures Authors: Supantho Rakshit, Adele Goldberg, Henry Conklin Published: 2026-07-18 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2607.22699v2 As Large Language Models (LLMs) grow more capable across diverse tasks, their (in)ability to generalize remains difficult to quantify and poorly understood beyond limited domains. In particular, LLMs are known to struggle generalizing multilingually, to languages outside of English, and that are poorly attested in their training data. To understand why this may be, and what enables some models to perform better than others, we turn to a long history of work across the cognitive sciences, arguing that successful generalization derives from appropriate representations in similarity space. We look at how well LLMs' representations capture the hierarchical similarity structure between distinct languages. Strikingly, we show LLMs' latent representations largely recover the hierarchical structure of the Indo-European language family tree -- grouping languages that are members of the same subfamily closely together in representation space. Furthermore, we show that the degree to which models reflect the similarity structure of languages correlates with their performance on XNLI, a multilingual natural language inference benchmark. This extends classic work on similarity-driven generalization at scale, showing how models that represent similar languages similarly generalize better from one language to another. --- ### Similarity All The Way Up: Multilingual Generalization in LLMs Relies on Language-Level Similarity Structures Authors: Supantho Rakshit, Adele Goldberg, Henry Conklin Published: 2026-07-18 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2607.22699v1 As Large Language Models (LLMs) grow more capable across diverse tasks, their (in)ability to generalize remains difficult to quantify and poorly understood beyond limited domains. In particular, LLMs are known to struggle generalizing multilingually, to languages outside of English, and that are poorly attested in their training data. To understand why this may be, and what enables some models to perform better than others, we turn to a long history of work across the cognitive sciences, arguing that successful generalization derives from appropriate representations in similarity space. We look at how well LLMs' representations capture the hierarchical similarity structure between distinct languages. Strikingly, we show LLMs' latent representations largely recover the hierarchical structure of the Indo-European language family tree -- grouping languages that are members of the same subfamily closely together in representation space. Furthermore, we show that the degree to which models reflect the similarity structure of languages correlates with their performance on XNLI, a multilingual natural language inference benchmark. This extends classic work on similarity-driven generalization at scale, showing how models that represent similar languages similarly generalize better from one language to another. --- ### FogDrive: A Multi-Modal Synthetic Driving Dataset for Perception under Graded Fog Authors: Vansh Panwar Published: 2026-07-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.22698v1 Perception under adverse weather remains a critical bottleneck for reliable autonomous driving, yet existing benchmarks lack the systematic multi-modal alignments needed to evaluate robust sensor fusion. Real-world weather datasets suffer from uncontrolled collection and single-level, uncalibrated conditions, while synthetic alternatives either target camera-only restoration or lack the paired clean-and-foggy structure needed to benchmark "defog-then-detect" pipelines. We present FogDrive, a rigorously calibrated, multi-modal autonomous-driving dataset bridging data-centric engineering and robust machine learning. Built with the CARLA simulator, FogDrive contains 660 scenes (~133k fully annotated frames, 50:50 day/night) across four synchronized cameras (RGB, depth, semantic segmentation), a LiDAR and semantic-LiDAR pair, and front radar. Physically consistent fog is modeled independently on camera channels (Koschmieder model) and LiDAR channels (Beer-Lambert law) at three calibrated visibility densities (160m, 100m, 50m). Every scene ships in four matched variants (clean plus three graded fog levels) with cross-calibrated 2D and 3D bounding boxes. A semantic-segmentation-based quality audit over 8k images validates annotations at 95.1% precision and over 99% recall for vehicles within 40m. We establish baseline benchmarks with state-of-the-art architectures (TransFusion, BEVFusion, YOLOv8-m) across two paradigms: 3D multi-modal fusion and 2D image restoration. These yield critical data-centric insights: mixing multi-density fog during training tightens 3D bounding-box geometry without added data-scaling cost, while in 2D pipelines image-quality metrics (PSNR, SSIM) prove poor predictors of downstream detection performance. FogDrive will be fully open-sourced alongside our data-generation framework to accelerate robust, multi-modal research. --- ### Test-Time Coverage: Test-Conditioned Data Curation for Deployment-Aware Learning Authors: Nadine Chang, Maying Shen, Shizhe Diao, Jialiang Wang, Jingde Chen, Thomas Breuel, Pavlo Molchanov, Rafid Mahmood, Jose M. Alvarez Published: 2026-07-18 Categories: cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2607.22697v1 Deployed AI systems are often trained from broad candidate data pools, necessitating data curation towards the deployment test distribution. However, standard data curation methods score training-side criteria rather than directly optimizing deployment match. We introduce TTCov (Test-Time Coverage), a data-level test-conditioned curation method that uses test-side information before training instead of updating model weights at inference. TTCov decomposes deployment-conditioned curation into coverage and distribution. To represent coverage, it builds a task Atlas, a collection of LLM-based atomic propositions (APs) describing deployment-relevant concepts, seeded from open task knowledge and expanded with unmatched APs extracted from unlabeled deployment samples. To represent distribution, it instantiates the matched deployment APs with their frequencies, yielding a Knowledge Atlas (K-Atlas) that operationalizes the deployment distribution as a curation target. TTCov then selects a budgeted training set whose deployment APs distribution approximates this target. We apply TTCov towards autonomous driving (AD), keeping adaptation off the inference path while selecting data with greater deployment-relevant coverage, closer K-Atlas matching, and stronger downstream end-to-end driving performance than data-curation baselines, including seamless adaptability to novel domains via city-to-city expansion. --- ### Genesis: An Empirical Platform for Studying Open-Ended Evolution Without Fitness Functions Authors: Anushka Sharma Published: 2026-07-18 Categories: cs.NE arXiv: https://arxiv.org/abs/2607.21631v1 Biological evolution sustains complex dynamics without any fitness function, yet virtually all evolutionary algorithms depend on one. Genesis is an open-source platform designed to test, empirically, what an artificial system needs to sustain evolutionary dynamics after complete fitness removal. Evolution in Genesis is governed by physical constraints, relational dominance, and adaptive regulation - no scalar fitness, no designer-specified objectives. Across experiments totalling over one million evolutionary generations, Genesis has: (i) shown that constraint-driven selection can sustain evolutionary activity after complete fitness removal (7/12 runs; Wilson 95% CI (30.2%, 82.5%); p<0.01, Cohen's d=1.47 vs. baselines); (ii) produced a sham-controlled negative result demonstrating that niche construction alone does not break the complexity plateau; and (iii) provided preliminary evidence that speciation-protected niche construction initiates structural diversification that unprotected secretion cannot. These findings establish empirical boundaries for fitness-free evolution and open a new direction: meta-evolution of physics, in which the laws governing an evolutionary system are themselves evolved. --- ### Evolving Self-Organising Agents Without Fitness: Three Falsifiable Experiments from Constraint-Driven Selection to Developmental Encoding Authors: Anushka Sharma Published: 2026-07-18 Categories: cs.NE arXiv: https://arxiv.org/abs/2607.21630v1 Can evolutionary dynamics characteristic of biological development arise without a designer-specified fitness function? We present Genesis, a platform in which agents inhabit a Gray-Scott reaction-diffusion substrate and evolve under physical constraints alone. Three successive experimental cycles, each testing one falsifiable hypothesis, show: (1) constraint-driven selection sustains evolutionary activity after complete fitness removal but reaches a hard phenotypic complexity ceiling; (2) agent-mediated niche construction via chemical secretion is real but causally insufficient to break that ceiling; and (3) replacing the fixed-alphabet genome with a Compositional Pattern Producing Network (CPPN) indirect encoding, protected by NEAT-style speciation, produces the first evidence of progressive structural complexification in a fitness-free system. Null results are treated as precise, informative answers rather than failures, yielding reusable diagnostic tools and a sham-control protocol applicable to any open-ended evolution evaluation pipeline. --- ### CT-Merging: Consensus Directions and Task-Specific Scaling for LoRA Adapter Merging Authors: Keumseo Ryum, Joonhyuk Kang Published: 2026-07-18 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2607.20561v2 LoRA merging methods increasingly operate on the low-rank structure of task updates, yet how the common subspace is estimated and how coefficients are assigned after recomposition are rarely compared directly. We propose CT-Merging, which estimates common directions from averaged task subspace projectors and assigns a separate residual scale to each task. Projector averaging selects directions supported across task subspaces without weighting them by singular magnitude, while task-specific scaling removes component-wise magnitude variation and preserves scale differences across tasks. On the released KnOTS CLIP adapters, CT-Merging achieves the best average and worst-task normalized accuracy on both backbones, improving over the strongest baseline by up to 2.56 and 6.65 points, respectively. On the DC-Merge adapter benchmark, it achieves the best average normalized accuracy in eight of nine backbone and task-count settings. Ablations show that projector averaging outperforms summed-update SVD and that task-specific scaling improves worst-task accuracy over global isotropic scaling. --- ### CT-Merging: Consensus Directions and Task-Level Scaling for LoRA Adapter Merging Authors: Keumseo Ryum, Joonhyuk Kang Published: 2026-07-18 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2607.20561v1 LoRA adapters provide an efficient way to specialize a pretrained model for many downstream tasks, but deploying one adapter per task requires adapter storage and task selection at inference time. Model merging addresses this issue by combining independently trained adapters into one multi-task adapter. Recent SVD-based LoRA merging methods mainly focus on constructing shared or task specific directions, while the coefficients assigned to the final directions are often directly from the original task SVD. On a fixed merged basis, inherited coefficients preserve component order with high rank correlation, yet their magnitudes differ substantially from the coefficients induced by the task updates. To address this mismatch, we propose CT-Merging, a LoRA-aware merging algorithm that estimates consensus directions from average task subspace projectors and assigns task-level RMS coefficient scales in the final update. CT-Merging uses repeated support across task SVD subspaces to construct the common basis, while reducing reliance on rank wise SVD magnitudes after direction construction. On the DC-Merge CLIP adapter benchmark, CT-Merging achieves superior average normalized accuracy compared to state-of-the-art merging methods and further improves over DC-Merge by 2.56 points on ViT-B/32 and 1.51 points on ViT-L/14 KnoTS-trained checkpoints. --- ### Hazard or Anomaly? Evaluating VLMs for Understanding Dangers and Discrepancies Authors: Murali Indukuri, Mohammad Eskandari, Sree Nitya Kollu, Stephanie Lukin, Cynthia Matuszek Published: 2026-07-18 Categories: cs.CV, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2607.18325v1 Modern safety-critical systems increasingly rely on human-robot interaction to reduce disaster risk and support decision-making during emergencies. Vision-Language Models (VLMs) are promising for these settings because they can interpret complex scenes and communicate safety-relevant information, but they still require careful evaluation to ensure reliable safety reasoning. In particular, current evaluations often frame danger recognition as a binary decision (Safe/Unsafe), making it unclear whether a model is identifying true physical hazards or merely reacting to unusual scene elements. We address this limitation by introducing an explicit distinction between hazard and anomaly, and by separately recognizing hazardous and anomalous states. We evaluate several state-of-the-art VLMs across two datasets and multiple prompting strategies to test whether this distinction changes model behavior. Our results show that VLMs frequently misinterpret anomalousness as hazardousness, revealing an over-reliance on contextual irregularity as a proxy for danger. We further show that explicitly separating anomaly from hazard provides a more informative evaluation of VLM safety reasoning and exposes failure modes that binary safety judgments can obscure. Our public dataset is available on Roboflow https://app.roboflow.com/vlm-in-context-anomaly-and-hazard-detection/camera-ready-roman-ds. --- ### Cost Accounting for Reactive Computational Graphs: Exhaustive Sweeps, Sequential Mutation, and the Backward-Locality Gap Authors: Abdallah Khemais Published: 2026-07-18 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.18323v2 Exhaustive site-by-site interventions on a neural network's computational graph -- activation-patching sweeps, circuit-discovery searches, systematic ablation studies -- mutate the graph at every candidate site, and their cost is dominated by recomputation after each mutation. On a reactive graph engine whose invalidation provably touches exactly the downstream cone of a mutated node, we give a complete cost accounting for such workloads. First, the aggregate speedup of an exhaustive sweep over independent full recomputations is not a universal constant: if per-layer weight varies regularly with depth at Karamata index q, the ratio converges to (q+2)/(q+1) when weight concentrates near the output and to q+2 near the input, recovering 2 only in the depth-uniform case; a wall-clock corollary predicts a ceiling of about 1.79, below 2, until interpreter overhead is compiled away. Second, we prove the exact cost of a sequence of persistent mutations, never undone between insertions: the interleaved cost exceeds the isolated sum by an exact overcount summed over comparable site pairs, with closed-form extremes over insertion orders, while batched application is order-independent and sub-additive, costing exactly the union of the sites' cones plus the fresh nodes. Third, we prove the exact mirror of forward locality for the backward pass, showing it collapses the aggregate speedup to 1 under backpropagation on architectures without long skip connections. Every identity is validated on NeuroDSL, a reactive graph engine in Julia: measured sweep ratios converge to the predicted limits under four cost profiles; the training-mode ratio collapses to 1 at the predicted rate; and all 18 per-graft sequential costs and the batched total match the closed forms at zero tolerance across three insertion orders. --- ### Cost Accounting for Reactive Computational Graphs: Exhaustive Sweeps, Sequential Mutation, and the Backward-Locality Gap Authors: Abdallah Khemais Published: 2026-07-18 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.18323v1 Exhaustive site-by-site interventions on a neural network's computational graph -- activation-patching sweeps, circuit-discovery searches, systematic ablation studies -- mutate the graph at every candidate site, and their cost is dominated by recomputation after each mutation. On a reactive graph engine whose invalidation provably touches exactly the downstream cone of a mutated node, we give a complete cost accounting for such workloads. First, the aggregate speedup of an exhaustive sweep over independent full recomputations is not a universal constant: if per-layer weight varies regularly with depth at Karamata index q, the ratio converges to (q+2)/(q+1) when weight concentrates near the output and to q+2 near the input, recovering 2 only in the depth-uniform case; a wall-clock corollary predicts a ceiling of about 1.79, below 2, until interpreter overhead is compiled away. Second, we prove the exact cost of a sequence of persistent mutations, never undone between insertions: the interleaved cost exceeds the isolated sum by an exact overcount summed over comparable site pairs, with closed-form extremes over insertion orders, while batched application is order-independent and sub-additive, costing exactly the union of the sites' cones plus the fresh nodes. Third, we prove the exact mirror of forward locality for the backward pass, showing it collapses the aggregate speedup to 1 under backpropagation on architectures without long skip connections. Every identity is validated on NeuroDSL, a reactive graph engine in Julia: measured sweep ratios converge to the predicted limits under four cost profiles; the training-mode ratio collapses to 1 at the predicted rate; and all 18 per-graft sequential costs and the batched total match the closed forms at zero tolerance across three insertion orders. --- ### Increasing Line Outage Localization Performance with Ensemble Classifiers Authors: Daniel Flores, Yuanrui Sang, Michael P. McGarry Published: 2026-07-18 Categories: eess.SY, cs.LG arXiv: https://arxiv.org/abs/2607.17008v1 In many cases, the outage of one transmission line in a system can be localized by monitoring the power flow of another line, and machine learning methods can be used to distinguish the cases under uncertainty. In this study, we examine the improvements in line outage localization performance achieved by various ensemble classifiers compared to single-model methods. In the case studies, we compared the classification results with measurement data collected at observed transmission lines (OTLs) selected using three algorithms, i.e, greedy maximum coverage problem (MCP), high-eta, and random selection, based on two sensitivity factors, i.e., line outage distribution factors (LODFs) and line outage impact factors (LOIFs). We found that the OTLs selected by the greedy MCP algorithm yielded the highest F1 score and the ensemble classifiers significantly outperformed a base kNN classifier. The extra-trees bagging technique achieved the highest F1 score in many instances. All the findings were statistically significant. --- ### Scalable Causal Imitation Learning Authors: Eylam Tagor, Mingxuan Li, Elias Bareinboim Published: 2026-07-18 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.17003v1 Imitation learning enables learning a policy in an unknown environment with a latent reward signal using expert demonstrations, but it struggles when the imitator's and expert's observations are mismatched and unobserved confounders are present in expert demonstrations. By identifying appropriate adjustment sets via the sequential $π$-backdoor criterion, causal imitation learning (CIL) provides a framework for approximating the expert's policy from confounded data. However, existing CIL methods, Causal Behavioral Cloning (Causal BC) and Causal Generative Adversarial Imitation Learning (Causal GAIL), are designed for short-horizon, low-dimensional settings. When applied to continuous control tasks with long horizons and high-dimensional state-action spaces, these methods exhibit poor performance: Causal BC suffers from compounding errors, Causal GAIL is unstable and sample-inefficient, and sequential $π$-backdoor adjustment becomes impractical. We introduce Causal Soft Q Imitation Learning (SQIL) and Causal Inverse soft-Q Learning (IQ-Learn), two off-policy causal imitation learning algorithms that combine the causal adjustment framework with state-of-the-art inverse reinforcement learning objectives. Both algorithms operate on causally-adjusted state representations produced by an efficient approximation of the sequential $π$-backdoor criterion, exploiting the causal structure of continuous control environments to reduce the full-horizon adjustment to a fixed-size sliding window. We evaluate all methods in a suite of confounded environments and find that Causal SQIL and Causal IQ-Learn substantially outperform prior CIL algorithms on long-horizon tasks, sometimes surpassing the expert, whereas all causally unaware imitation methods fail to learn meaningful behavior. ---