# Research Papers — 2026-06 · bundle page 88 of 260 > Full-text markdown of papers #4351–4400 of 12993 published in 2026-06, newest first. AI-readable mirrors of the full 12993-paper 2026-06 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-06/page/88 [← Previous bundle](https://aipod.dev/papers/archive/2026-06/page/87.md) [Next bundle →](https://aipod.dev/papers/archive/2026-06/page/89.md) ### Embedding Linear Equality Constraints in Probabilistic Neural Networks for Dynamic Modelling Authors: Matthew Marsh, Benoit Chachuat, Antonio del Rio Chanona Published: 2026-06-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.21728v1 Machine learning models are increasingly used to model chemical process systems, yet they often lack principled uncertainty quantification and mechanisms to enforce physical constraints. We propose a probabilistic neural network framework that guarantees satisfaction of linear equality constraints within a given tolerance, while capturing aleatoric uncertainty. Compared to state-of-the-art methods, our formulation demonstrates improved predictive accuracy, uncertainty calibration, and adherence to constraints on reduced data. It also demonstrates competitive performance, but with significantly faster training times when evaluated on large data regimes. We evaluated this on two batch reactor case studies, enforcing mass balances. --- ### Entropy Objectives in Markov Decision Processes Authors: S. Akshay, Raghav Goyal, Aditya Neeraje, Piyush Srivastava Published: 2026-06-19 Categories: cs.AI arXiv: https://arxiv.org/abs/2606.21726v1 We consider the problem of synthesizing control policies that enforce a concentration property on the state distributions of a stochastic system. We present a formalization of this problem in terms of synthesizing strategies for maintaining an entropy-based objective in Markov Decision Processes (MDPs). We first show that even relaxed versions of this problem are complexity-theoretically hard. We then present a sound and (conditionally) relatively complete method to verify and synthesize strategies for such entropy objectives. The main challenge is the non-linear nature of such objectives, and our approach addresses this by exploiting and combining ideas from convex duality and invariant synthesis. We also investigate the role of memory and randomization in ensuring entropy objectives. Finally, we implement our ideas to evaluate our approach empirically on a few illustrative benchmarks. --- ### Physics-Informed Neural Networks for Computing the Morse Index of the Critical Catenoid Authors: Miraj Samarakkody Published: 2026-06-19 Categories: math.DG, cs.LG, math.NA, math.SP arXiv: https://arxiv.org/abs/2606.21725v1 The Morse index of a free boundary minimal surface is encoded in its Jacobi-Steklov spectrum, and we test how faithfully a physics-informed neural network (PINN) reproduces that spectrum on a problem whose answer is already known in closed form. The benchmark is the critical catenoid in the unit ball $\mathbb{B}^3$, where it is well known that the Morse index equals $4$ and the nullity equals $2$. Separating the angular variable reduces the eigenvalue problem to a family of one-dimensional Robin problems on $(-T,T)$, one for each Fourier mode. A network that enforces the parity of each mode by construction, and carries the eigenvalue as a trainable parameter, returns the three eigenvalues below the stability threshold to within $10^{-6}$ to $10^{-4}$ of their exact values, with PDE residuals of order $10^{-4}$; assembling them recovers the index $4$ and the nullity $2$. We then track the spectrum along a one-parameter homotopy joining a flat reference operator to the catenoid Jacobi operator and identify the crossings at which the index changes. Since the critical catenoid is rigid, a fact we prove, this homotopy deforms operators rather than surfaces. We close by explaining how the same pipeline, with its one-dimensional solver replaced by a two-dimensional one, is poised to address genuinely geometric families in ellipsoidal balls, where the boundary curvature is no longer constant, and the Morse index is not yet known. --- ### Denoising Iterative Self-Correction: Structured Verification Loops for Reliable LLM Reasoning Authors: Shen Yin, David Ken, Joel Stremmel Published: 2026-06-19 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2606.21724v1 Large language models produce fluent but often incorrect multi-step reasoning, and naive correction methods risk degrading already-correct answers. We introduce Denoising Iterative Self-Correction (DISC), a test-time procedure that treats verification question outputs as noisy measurements of where a solution may be corrupted. Using these signals, DISC progressively reduces errors across multiple verify-judge-correct passes, analogous to traditional iterative denoising. A binary judgment gate controls correction precision by blocking rewrites that would damage already-correct answers while the verifier and corrector together repair errors. We evaluate this trade-off using two paired diagnostics: an improvement-to-degradation ratio (precision) and a repair rate (recall). Across three benchmarks (BIG-Bench Mistake, HotpotQA, GPQA Diamond) and four models, DISC dominates Chain-of-Verification and Self-Refine on the precision-recall trade-off, reaching 81.6% accuracy with 13x more improvements per degradation than Chain-of-Verification and 5x more than Self-Refine on BIG-Bench Mistake (Sonnet~4.5). On GPQA Diamond, we identify a capability floor below which judges acknowledge contradictions in evidence but cannot translate that recognition into a correction. We further show that cross-model role allocation -- assigning verification and judgment to a model different from the generator -- mitigates self-confirmation bias. --- ### Leveraging LaBSE with Progressive Curriculum Learning for Multicultural Polarization Authors: Sachin Sundar, Sandeep Kumar, Mothish M Published: 2026-06-19 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.21718v1 Detecting online polarization remains a critical challenge, particularly in multilingual and multicultural contexts where intergroup hostility is prevalent. The problem is particularly challenging due to the data scarcity for these tasks in the low-resource languages. Identifying such phenomena has become an active area of research and is addressed in SemEval-2026 Task 9: Multilingual, Multicultural Online Polarization Detection. To address this problem we propose an architecture that leverages LaBSE embeddings - an unconventional choice typically reserved for retrieval tasks, to obtain strong cross-lingual learning which enhances scores in low-resource language by a score up to 0.2 macro F1. Furthermore, we provide a comprehensive ablation study evaluating the performance of diverse encoder models in the Qwen model family within a retrieval-based prompting framework. Our code will be soon available at https://github.com/carrycurious/PolarMind. --- ### Adaptive Beam Selection for Efficient Scanning Probe Tomography Authors: San Dinh, Zichao Wendy Di, Matt Menickelly Published: 2026-06-19 Categories: physics.med-ph, cs.CV arXiv: https://arxiv.org/abs/2606.21713v1 In X-ray tomography, reconstruction quality generally improves with larger numbers of projections. However, more projections increase experiment costs, acquisition time and the radiation dose imparted to the sample. One mitigation to these trade-offs is to adopt a sequential design of experiments, in which each subsequent measurement is determined as a function of previously acquired data in order to maximize information gain. In practice, a widely used heuristic to maximize information is to align beams with the edges of the sample. A key challenge, however, is that the true sample is unknown, so identifying edge-aligned beams typically requires reconstructing the sample based on available measurements. This work proposes a novel sequential design method that identifies edge-aligned measurements directly from the sinogram, bypassing any reconstruction, thereby improving computational efficiency and reducing the experimental design's susceptibility to reconstruction errors. Our method dynamically selects the next set of measurement beams by maximizing an acquisition function that balances exploration and exploitation over the domain of all possible measurements, improving reconstruction quality while reducing measurement redundancy. --- ### BatchGen: An Architecture for Scalable and Efficient Batch Inference Authors: Tairan Xu, Leyang Xue, Zhan Lu, Jinfu Deng, Hongyang Xiao, Yinsicheng Jiang, Congjie He, Matej Sandor, Le Xu, Luo Mai Published: 2026-06-19 Categories: cs.DC, cs.LG arXiv: https://arxiv.org/abs/2606.21712v1 Batch inference has become a central mode of AI computation, yet existing inference engines still rely on execution models designed for interactive serving. When scaled to millions of sequences, batch workloads reveal two fundamental requirements: the ability to handle extreme inter- and intra-sequence load variation that emerges only at runtime, and the ability to sustain high utilization across large fleets of GPUs. Existing systems fail to meet these requirements, losing substantial fractions of achievable throughput. We introduce a new architectural foundation for batch inference: the sequence coroutine compute model, which represents each sequence as a fine-grained, event-driven coroutine. This model exposes expressive primitives that allow the runtime to reorganize work dynamically, enabling larger expert-level batches, mitigating stragglers, reallocating work across devices, and maintaining utilization even on cost-effective or memory-constrained GPUs. Building on this abstraction, we implement BatchGen, a production-ready system that uses the coroutine model at cluster scale. On a 128-GPU cluster, BatchGen reduces batch completion time by up to $2.3\times$, and on memory-constrained accelerators it outperforms the strongest offloading baseline by up to $9.6\times$. We will open-source BatchGen at https://github.com/batchgen-project/batchgen --- ### PrivacyAlign: Contextual Privacy Alignment for LLM Agents Authors: Manveer Singh Tamber, Abhay Puri, Marc-Etienne Brunet, Perouz Taslakian, Jimmy Lin, Spandana Gella Published: 2026-06-19 Categories: cs.CL, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2606.21710v1 AI agents acting on behalf of users are constantly making decisions, and for users to trust their agents, those decisions must align with what they actually want. Privacy is an important alignment problem for agents: every message, post, or tool call an agent makes is a contextual judgment about what is appropriate to share, with whom, and under which conditions. Because such judgments depend on social expectations and norms, human judgment does not merely label privacy violations but also helps define them. While existing work relies on unreliable proxies for both training and evaluation, we place human judgment at the center of agentic privacy alignment. We introduce PrivacyAlign, a dataset of 1,350 samples with 3,516 detailed annotations from 599 unique annotators across diverse scenarios where current LLMs actually leak, and use it to ground both alignment training and automated evaluation in human privacy norms. Building on these annotations, we first show that conditioning LLM judges on human annotations and explanations for reference responses to the same prompt makes their judgments more reliable. We then introduce annotation-conditioned reward modeling, which uses these annotations to score new responses during RL, and show that small open-weight agents trained with this reward better align with human privacy norms, with strong gains on PrivacyAlign and existing privacy benchmarks for agents. --- ### Structural Assessment for Understanding and Guiding Dataset Distillation in Discrete Token Space Authors: Yue Cao, Jianyang Gu, Vyacheslav Kungurtsev, Yu Hu, Jozsef Hamari, Zheng Liu, Mohsen Zardadi Published: 2026-06-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.21705v1 Dataset distillation (DD) has proven to reduce training cost while preserving accuracy. While promising, the factors that make one distilled dataset more effective than another remain poorly understood. In this work, we investigate this question through the lens of discrete visual tokenizers. Whereas many prior DD efforts emphasize matching global data distributions, we suggest that the effectiveness depends on which semantic concepts are captured and how they are composed. Discrete visual tokenizers provide a finite vocabulary that enables direct statistical analysis of such compositional structure. Through quantitative analysis of token-level statistics, we introduce the structural score to measure the adequacy of token compositions. We observe that distilled datasets with balanced token composition yield higher validation performance. On the other hand, divergence from the original data does not necessarily harm performance. We further show that samples with high structural scores in the discrete token space can effectively guide diffusion-based DD. Our findings highlight the importance of token composition in dataset effectiveness, offering a principled complement to distributional similarity considerations in DD. --- ### When Compression Helps and When It Hurts: Condition-Aware Analysis of Chain-of-Thought Distillation Authors: Siyang Lyu, Xinghao Chen, Zhijing Sun, Tong Liu, Dawei Zhu, Xiaoyu Shen Published: 2026-06-19 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.21704v2 Chain-of-Thought (CoT) distillation transfers multi-step reasoning from large reasoning models to smaller students, but verbose teacher traces inflate both training and inference cost. Existing CoT compression methods fall into two families, selective pruning and generative rewriting, yet prior studies have left key factors entangled: granularity is confounded with importance criteria in pruning, restructuring level is rarely isolated in rewriting, and compression budgets are not systematically evaluated across domains or regimes. We recast CoT compression along three dimensions: importance criterion, restructuring level, and compression budget. Sweeping these across two model families, Math and General domains, and Long-/Short-CoT regimes, we find that (i) importance criterion utility is strictly governed by granularity: step-level criteria converge on a shared reasoning backbone, while token-level pruning requires symbol-aware signals to preserve the logical core; (ii) restructuring level inverts across domains: Math degrades monotonically with structural disruption, while aggressive rewriting acts as a denoiser on General tasks; (iii) training-time compression does not necessarily translate to inference-time savings: Long-CoT students retain verbose habits despite concise supervision, making the training ratio an optimistic lower bound on deployment cost. These findings yield condition-aware guidelines for matching compression to deployment context. --- ### When Compression Helps and When It Hurts: Condition-Aware Analysis of Chain-of-Thought Distillation Authors: Siyang Lyu, Zhijing Sun, Xinghao Chen, Tong Liu, Dawei Zhu, Xiaoyu Shen Published: 2026-06-19 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.21704v1 Chain-of-Thought (CoT) distillation transfers multi-step reasoning from large reasoning models to smaller students, but verbose teacher traces inflate both training and inference cost. Existing CoT compression methods fall into two families, selective pruning and generative rewriting, yet prior studies have left key factors entangled: granularity is confounded with importance criteria in pruning, restructuring level is rarely isolated in rewriting, and compression budgets are not systematically evaluated across domains or regimes. We recast CoT compression along three dimensions: importance criterion, restructuring level, and compression budget. Sweeping these across two model families, Math and General domains, and Long-/Short-CoT regimes, we find that (i) importance criterion utility is strictly governed by granularity: step-level criteria converge on a shared reasoning backbone, while token-level pruning requires symbol-aware signals to preserve the logical core; (ii) restructuring level inverts across domains: Math degrades monotonically with structural disruption, while aggressive rewriting acts as a denoiser on General tasks; (iii) training-time compression does not necessarily translate to inference-time savings: Long-CoT students retain verbose habits despite concise supervision, making the training ratio an optimistic lower bound on deployment cost. These findings yield condition-aware guidelines for matching compression to deployment context. --- ### ARCO-Mars: A Unified Cloud-Optimized Archive of Mars Atmosphere Reanalysis Authors: Ananyo Bhattacharya Published: 2026-06-19 Categories: astro-ph.EP, astro-ph.IM, cs.DB, cs.LG arXiv: https://arxiv.org/abs/2606.21701v1 Long-term records of the Martian atmosphere based on general circulation models and reanalysis of atmospheric state variables are important to understand the diurnal, seasonal, and climatological changes of the planet. Atmospheric dynamics of the Martian atmosphere are strongly influenced by the characterization of dust lifting, solar insolation, and spatial variations in topography. We present ARCO-Mars, a unified Analysis-Ready Cloud-Optimized dataset providing integrated access to three independent Mars atmospheric reanalysis products: EMARS, MACDA, and OpenMARS spanning over Mars Years 24-35. These reanalyses assimilate thermal infrared retrievals from the MGS/TES, ODY/THEMIS, and MRO/MCS instruments, providing both two and three-dimensional surface and atmospheric state variables, including temperature, winds, surface pressure, and dust optical depth. The dataset is stored in Zarr v3 format and hosted on HuggingFace, enabling efficient cloud-based access without requiring local storage of the full archive. We compare the state variables between the three reanalysis products to identify systematic differences, attributed to differences in data assimilation and general circulation models. ARCO-Mars provides a community resource for Mars atmospheric science, numerical weather prediction validation, and machine learning applications, including weather forecasting and data assimilation. --- ### VT-DUDA: Visual Token Conditioning for Diffusion-guided Unsupervised Domain Adaptation Authors: Xuan Qi, Daniele Berardini, Dario Serez, Vito Paolo Pastore, Vittorio Murino Published: 2026-06-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.21700v1 Unsupervised domain adaptation (UDA) aims to learn a target-domain classifier from labeled source data and unlabeled target data under distribution shift. Recent diffusion-based UDA methods approach this problem by synthesizing labeled target-style images and training on the resulting synthetic data. However, their performance depends heavily on the conditioning design: class prompts provide only coarse guidance, while domain adaptation modules mainly control appearance, which may leave target-style synthesis insufficiently specified. We propose VT-DUDA, a visual-token conditioning framework for diffusion-guided UDA. Instead of relying only on text prompts, VT-DUDA uses source images to provide additional instance-level visual context for target-style synthesis. Specifically, VT-DUDA maps each source image to a compact sequence of visual tokens and forms a hybrid conditioning context by concatenating these tokens with the corresponding text embeddings along the cross-attention context dimension of a latent diffusion model. This provides instance-dependent conditioning beyond text alone, while synthesis is performed with the target-domain adapter branch. Because guidance is represented explicitly as a token sequence, the same interface also permits inference-time manipulation of the conditioning signal through token selection and token-strength adjustment. The proposed method preserves the standard diffusion objective and can be integrated into existing adapter-based diffusion frameworks without modifying the backbone. Across Office-31, Office-Home, and VisDA-2017, VT-DUDA improves average target-domain accuracy over strong discriminative and diffusion-based UDA baselines. The results suggest that, in generation-based UDA, a stronger conditioning interface can improve the downstream usefulness of synthetic target-style data. --- ### A Hybrid, Multi-Layered Pipeline for Phishing and Threat Classification: Independently Validated URL and NLP Engines with a Calibrated Multi-Channel Fusion Stage Authors: Saifelden M. Ismail, Aser O. Ibrahim, Omar A. Mahmoud Published: 2026-06-19 Categories: cs.CR, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.21690v2 Phishing is a multi-modal threat. We present a hybrid pipeline that scores each modality with its own engine and fuses the results. Three engines are built, deployed, and independently benchmarked: a four-stage URL stack (Domain Guard, lexical model, threat intelligence, and an asymmetric L2 fusion sidecar); a generalization-hardened DistilBERT NLP classifier whose held-out real-phishing recall rises from 0.8% to 87.3%; and a threat-intelligence synchronizer with end-to-end OpenTelemetry instrumentation confirming 1:1 message conservation. A decision-level fusion stage, characterized on a 10,677-email whole-system benchmark, reaches F1=0.914 with a calibrated probabilistic-OR over URL, header, and phishing-probability channels while reducing held-out real-spam false positives to 3.6%. Because that benchmark uses proxy URL and header channels and an operating point still needing recalibration, we present it as a preliminary integrated result. For deployable detection, the limiting factor is how well a model generalizes, not how accurately it scores data drawn from its own training distribution. --- ### Clinical Term Extraction using Open-Source Small Language Models Authors: Noah Marchal, William E. Janes, Mihail Popescu, Xing Song Published: 2026-06-19 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2606.21689v1 Clinical information for amyotrophic lateral sclerosis (ALS) care documented in unstructured clinical notes limits downstream analysis without extraction into structured formats. Open-source small language models with few-shot prompting for detecting the presence of ALS-relevant clinical terms in patient documentation were evaluated without task-specific training data. The detection task targeted 17 categories spanning functional scores, respiratory measures, medications, and related clinical and non-clinical attributes. Clinical note content was normalized from JSON-encoded discharge summaries and processed with a prompt template having structured JSON outputs. We compared 26 open-source models using aggregate, label-level, and manual-validation multilabel classification metrics. Manual validation showed that a regex rule baseline had higher overall micro-F1 and lower Hamming loss than any single SLM or TF-IDF baseline, while Qwen3-4B-Instruct-2507 was the highest-performing SLM by micro-F1. Model rankings varied by metric and label category, with the TF-IDF baseline showing high recall but low precision, some SLMs showing higher precision but lower recall, and Hammer2.1-7b showing strong performance for ALSFRS-R subscore detection. These findings support targeted hybrid extraction workflows rather than replacement of existing rule-based methods. --- ### Expressivity Saturation: Reduced Affine Region Usage Under Increasing Task Complexity Authors: Xuan Qi, Yi Wei, Fanqi Yu, Manuel Lecha Published: 2026-06-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.21687v1 Piecewise-affine neural networks (e.g., with ReLU or LeakyReLU activations) implement continuous piecewise-affine maps, and the number of affine regions provides a natural proxy for expressive capacity. However, the gap between theoretical region capacity and the affine regions realized after training remains insufficiently understood. We study this gap from two complementary perspectives. First, we give a rigorous, architecture-dependent theorem for affine line-segment probes: for multilayer perceptrons with piecewise-affine activations, the number of affine pieces realized along an affine line-segment probe is upper bounded by an explicit product of layer-wise width terms (and activation breakpoint factors). This yields a neuron-threshold lower bound for representing target functions with prescribed one-dimensional piece complexity, formalizing the minimal region budget required for complex signals. Second, we exactly enumerate affine regions realized within bounded 2D and higher-dimensional domains under controlled task complexity. Under fixed architectures and training protocols, increasing input--label complexity yields trained solutions with markedly fewer realized regions in the evaluation domain, even though worst-case architectural capacity is unchanged; we call this reduced region usage expressivity saturation. Moreover, in the most challenging regimes, 2D visualizations show that region-usage collapse often coincides with degraded decision boundaries. Finally, we visualize the training dynamics of affine-region partitions and decision boundaries, revealing a consistent refinement process during optimization. --- ### TACO: Task-Aware Column Description Generation Using LLMs Authors: Ting Cai, Rakesh R. Menon, Yiru Chen, Zifan Liu, Yuan Tian, Fei Wu, Anudeep Chimakurthi, Prashanthi Ramamurthy, Sunav Choudhary, Kun Qian, Yunyao Li Published: 2026-06-19 Categories: cs.CL, cs.AI, cs.DB arXiv: https://arxiv.org/abs/2606.21685v1 Generating accurate and informative column descriptions (e.g. "membership status of customers" for the column name "cust_mem") is essential for a wide range of downstream NLP tasks on tabular data, including NL2SQL, table question answering, and entity linking. This problem arises in enterprises, domain sciences, government data portals, and so on. Despite its importance, most real-world datasets suffer from missing or cryptic documentation, often due to abbreviated column names or domain-specific jargon. Existing approaches largely rely on single-prompt large language models (LLMs), which struggle with three key issues: (i) inconsistent or incorrect handling of abbreviations, (ii) hallucinated or incomplete descriptions, and (iii) redundancy or vagueness that hinders downstream performance. We present TACO, a task-aware framework for automatic column description generation using LLMs. TACO introduces a three-step pipeline: (1) abbreviation expansion, which standardizes column names; (2) description generation, which produces initial semantic descriptions enriched with synonyms and search-oriented keywords; and (3) description revision, which refines these outputs using simulated downstream tasks. In addition, we investigate human-in-the-loop extensions and release new evaluation datasets for entity linking and schema enrichment. Extensive experiments across public and proprietary datasets show that TACO consistently outperforms existing methods, improving downstream task performance by up to 32%. --- ### Finite-Sample Performance of Gradient Descent in Logistic Regression with Gaussian Design Authors: Junren Chen, Arya Mazumdar Published: 2026-06-19 Categories: stat.ML, cs.IT, cs.LG arXiv: https://arxiv.org/abs/2606.21683v2 We consider the parameter estimation problem in logistic regression with Gaussian design: the estimation of a fixed unknown parameter $θ^*\in \mathbb{R}^d$ ($\|θ^*\|_2\ge 1$) from $n$ i.i.d. samples $\{(x_i,y_i)\}_{i=1}^n$, where $x_i\sim N(0,I_d)$ and $y_i|x_i \sim {\rm Bernoulli}(1/(1+\exp(-x_i^\top θ^*)))$. Our main aim is to characterize the finite-sample estimation performance and convergence behavior of gradient descent (GD) on the maximum likelihood objective (i.e., the logistic loss). Under small $O(1)$ stepsize and $0$ initialization, we show that GD linearly converges to a small neighborhood of $θ^*$ achieving an $\ell_2$ error of order $O(\sqrt{\|θ^*\|_2^5d/n})$. This substantially goes beyond existing theoretical results that lack non-asymptotic estimation error rate and exhibit much slower parameter convergence. We also establish a faster local linear convergence to the same statistical error under a large $Θ(\|θ^*\|_2)$ stepsize. The main technical component is to show that the gradient of the logistic loss satisfies a certain approximate invertibility condition (AIC). To that end, we uniformly control the deviation of the gradient from its population counterpart by covering and peeling arguments, and then show that the population GD is a contraction by a delicate analysis based on the eigenvalues of population Hessian matrices. Finally, we build upon the recent work Matsumoto and Mazumdar (2025) and devise a novel efficient estimator that attains a sharper rate in high dimensions. This indicates that the existing non-asymptotic guarantees exhibit sub-optimal dependence on $\|θ^*\|_2$, and that in many regimes $Θ(\sqrt{\|θ^*\|_2d/n})$ is the tight estimation error rate. Numerical examples are provided to corroborate our theoretical results. --- ### Finite-Sample Performance of Gradient Descent in Logistic Regression with Gaussian Design Authors: Junren Chen, Arya Mazumdar Published: 2026-06-19 Categories: stat.ML, cs.IT, cs.LG arXiv: https://arxiv.org/abs/2606.21683v1 We consider the parameter estimation problem in logistic regression with Gaussian design: the estimation of a fixed unknown parameter $θ^*\in \mathbb{R}^d$ ($\|θ^*\|_2\ge 1$) from $n$ i.i.d. samples $\{(x_i,y_i)\}_{i=1}^n$, where $x_i\sim N(0,I_d)$ and $y_i|x_i \sim {\rm Bernoulli}(1/(1+\exp(-x_i^\top θ^*)))$. Our main aim is to characterize the finite-sample estimation performance and convergence behavior of gradient descent (GD) on the maximum likelihood objective (i.e., the logistic loss). Under small $O(1)$ stepsize and $0$ initialization, we show that GD linearly converges to a small neighborhood of $θ^*$ achieving an $\ell_2$ error of order $O(\sqrt{\|θ^*\|_2^5d/n})$. This substantially goes beyond existing theoretical results that lack non-asymptotic estimation error rate and exhibit much slower parameter convergence. We also establish a faster local linear convergence to the same statistical error under a large $Θ(\|θ^*\|_2)$ stepsize. The main technical component is to show that the gradient of the logistic loss satisfies a certain approximate invertibility condition (AIC). To that end, we uniformly control the deviation of the gradient from its population counterpart by covering and peeling arguments, and then show that the population GD is a contraction by a delicate analysis based on the eigenvalues of population Hessian matrices. Finally, we build upon the recent work Matsumoto and Mazumdar (2025) and devise a novel efficient estimator that attains a sharper rate in high dimensions. This indicates that the existing non-asymptotic guarantees exhibit sub-optimal dependence on $\|θ^*\|_2$, and that in many regimes $Θ(\sqrt{\|θ^*\|_2d/n})$ is the tight estimation error rate. Numerical examples are provided to corroborate our theoretical results. --- ### Decodable but Not Faithful: Coupling Natural-Language Rationales to Programmatic Verifiers Authors: Vatsal Ananthula, Adarsh Kumarappan Published: 2026-06-19 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2606.21678v2 Language models can generate plausible rationales for their predictions, but these explanations may not faithfully represent the model's internal reasoning. We propose verifier-coupled reasoning, a framework that inserts inline claims into reasoning traces and trains an auxiliary consistency head to predict programmatic verifier outputs from rationale-span hidden states. The central finding is a gap between decodability and faithfulness: consistency training reliably makes verifier information decodable from rationale representations, but decodability does not guarantee faithful generation. In LeanCheck (formal theorem proving), rationale-only and proof-only pooling achieve perfect directional separation under counterfactual conflict. In KataGo (Go engine), commentary spans encode 10-way win-rate buckets at 81% accuracy. Yet in a code setting, the model achieves 98.6% coupling while its generated explanations remain unfaithful: fluent prose with correct structured claims, but describing unrelated algorithms; a controlled pretrained-vs-from-scratch comparison shows the gap is not capacity-driven. Synthetic activation patching confirms causal influence (73-89% vs. 31% baseline), FEVER reveals that evidence-only pooling isolates genuine evidence sensitivity at the cost of raw accuracy, and per-claim analysis shows that consistency loss disproportionately benefits fine-grained claims over binary ones. These results establish that consistency losses are effective diagnostics and representation-shaping tools, but not sufficient conditions for faithful reasoning. --- ### Decodable but Not Faithful: Coupling Natural-Language Rationales to Programmatic Verifiers Authors: Vatsal Ananthula, Adarsh Kumarappan Published: 2026-06-19 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2606.21678v1 Language models can generate plausible rationales for their predictions, but these explanations may not faithfully represent the model's internal reasoning. We propose verifier-coupled reasoning, a framework that inserts inline claims into reasoning traces and trains an auxiliary consistency head to predict programmatic verifier outputs from rationale-span hidden states. The central finding is a gap between decodability and faithfulness: consistency training reliably makes verifier information decodable from rationale representations, but decodability does not guarantee faithful generation. In LeanCheck (formal theorem proving), rationale-only and proof-only pooling achieve perfect directional separation under counterfactual conflict. In KataGo (Go engine), commentary spans encode 10-way win-rate buckets at 81% accuracy. Yet in a code setting, the model achieves 98.6% coupling while its generated explanations remain unfaithful: fluent prose with correct structured claims, but describing unrelated algorithms; a controlled pretrained-vs-from-scratch comparison shows the gap is not capacity-driven. Synthetic activation patching confirms causal influence (73-89% vs. 31% baseline), FEVER reveals that evidence-only pooling isolates genuine evidence sensitivity at the cost of raw accuracy, and per-claim analysis shows that consistency loss disproportionately benefits fine-grained claims over binary ones. These results establish that consistency losses are effective diagnostics and representation-shaping tools, but not sufficient conditions for faithful reasoning. --- ### Enlight: Fast Low-Light Image Enhancement via Multi-Objective Optimization and Shadow-Aware Refinement Authors: Nirjhor Datta, M. Sohel Rahman Published: 2026-06-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.21674v1 We present ENLIGHT, a fast and training free framework for low-light image enhancement based on direct optimization of a perceptual objective. Unlike deep learning approaches that require large scale training data and supervision, ENLIGHT operates in a zero-shot manner by optimizing image quality at inference time. The method employs a two stage global to local optimization strategy. In the first stage, ENLIGHT performs global illumination adjustment to improve visibility while maintaining structural consistency and avoiding excessive noise enhancement. In the second stage, a shadow aware refinement selectively improves low-intensity regions through masked local optimization, enhancing visibility without overexposure. To balance quality and efficiency, we introduce two modes: Fast, which uses a multi-objective formulation combining entropy, gradient preservation, and noise regularization, and Ultrafast, which reduces computational cost via a lightweight approximation of the same objective. The framework is optimizer agnostic and supports both evolutionary and lightweight local search methods. Experiments on BAID, Backlit300, LIME, MEF, NPE, and DICM demonstrate that ENLIGHT achieves competitive perceptual quality (MUSIQ, NIQE, BRISQUE) with significantly lower inference time. Qualitative results further show improved contrast, preserved structural details, and controlled noise amplification, making ENLIGHT a practical and interpretable alternative to learning based methods. --- ### Imitation from Heterogeneous Demonstrations using Grounded Latent-Action World Models Authors: Tianyou Wang, Anson Lei, Joe Watson, Ingmar Posner Published: 2026-06-19 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2606.21672v1 Imitation learning has emerged as a powerful paradigm for learning visuomotor policies, but its generalisation and stability are limited by the scale and quality of demonstration data needed. A promising direction is to leverage more abundant but heterogeneous data sources, which differ in action space and often lack action labels altogether. Existing co-training approaches that combine heterogeneous data sources rely on heuristic and hand-engineered alignment techniques. In contrast, we argue that action representations should be grounded in prediction: actions that produce the same effect on the environment should share the same representation, regardless of their sources. To this end, we instantiate this principle by using a grounded latent-action world model (GLAM), a pair of generative models with a shared latent action space across data sources that is grounded by predicting future observations consistently across sources. This latent action space is used to train downstream behavioural cloning (BC) policies which map observations to latent actions and decode them back to robot actions, providing a paradigm for learning from heterogeneous data. Empirically, we demonstrate that GLAM successfully learns an aligned latent action space that facilitates action transfer across data sources with and without action labels. Across five manipulation tasks in simulation and in the real world, GLAM-aligned policies significantly outperform BC baselines and prior latent-action methods, achieving an average of +48% improvement in task success rate with the same data-scarce setting. Videos and code are available at https://viccccciv.github.io/glam/. --- ### Improving Text-to-Music Generation with Human Preference Rewards Authors: Yonghyun Kim, Junwon Lee, Haiwen Xia, Yinghao Ma, Chris Donahue Published: 2026-06-19 Categories: cs.SD, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2606.21670v1 We describe our entry to the efficiency track of the Academic Text-to-Music (ATTM) Grand Challenge at ICME 2026. Beyond the challenge protocol's FAD-CLAP and CLAP score, we add a learned human-preference reward from TuneJury, a twin pairwise ranker trained over open music-preference datasets. The reward serves both as a training-time conditioning signal and as a sample-selection criterion. The pipeline combines five engineering decisions on a 120M-parameter FluxAudio-S backbone, four at training time and one at inference: (i) training-time reward conditioning that doubles as an inference-time CFG axis, (ii) a sweep over five score-conditioning architectures, where training and inference use different variants, (iii) expert iteration on the top decile, (iv) a short preference-tuning pass (CRPO) for audio-text alignment, and (v) inference post-processing via joint CFG, source separation, and loudness normalization. Per-stage decomposition on 100 Song Describer prompts shows training-time reward conditioning as a functional conditioning axis, expert iteration as the dominant contributor, the preference-tuning pass adding only noise-level gain, and the inference-time score scalar already saturated by the end of the chain. --- ### A Framework for Directed Acyclic Hypergraph Learning Authors: Zhiyuan Dong, Carlos Mundo-Levano, Wei Qian, Daniel Lau, Gonzalo R. Arce Published: 2026-06-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.21668v1 Continuous optimization methods for learning Directed Acyclic Graphs (DAGs) operate on weighted adjacency matrices and are therefore limited to pairwise causal relationships. We propose a framework for learning Directed Acyclic Hypergraphs (DAHGs) from observational data, capturing joint parental influences that pairwise models cannot represent. Our approach rests on three components: (i) a generalized linear structural equation model (SEM) with multiplicative interaction terms whose non-zero weights correspond one-to-one with directed hyperedges; (ii) a weighted adjacency tensor representation whose acyclicity is characterized via nilpotency under the tensor t-product; and (iii) a differentiable acyclicity constraint derived through the Fourier decomposition of the t-product, which reduces tensor nilpotency to slice-wise matrix nilpotency and enables least-squares learning via the augmented Lagrangian method. --- ### Hallucination as Context Drift: Synchronization Protocols for Multi-Agent LLM Systems Authors: Carson Rodrigues Published: 2026-06-19 Categories: cs.AI, cs.CL, cs.MA arXiv: https://arxiv.org/abs/2606.21666v1 Multi-agent LLM systems routinely produce hallucinated outputs that cannot be explained by model deficiencies alone. A significant class of these failures arises not from model incapacity but from context drift: the divergence of internal knowledge states between concurrent agents. When agents enter a collaborative task with mismatched or stale representations of shared world state, their joint reasoning produces contradictions that manifest as hallucination. We define the Context Divergence Score (CDS), a lightweight scalar metric quantifying knowledge-state discrepancy between agent pairs across spatial, temporal, and task dimensions, and propose the Shared State Verification Protocol (SSVP), which lets agents periodically exchange compressed state summaries and flag high-divergence conditions before joint reasoning. We evaluate SSVP across two domains (multi-agent travel and software project planning) using Claude Haiku. In controlled experiments (n=30 per condition, travel; n=10, software) across 8 scenarios, naive full-broadcast synchronization increases hallucination rate by 34% above the no-sync baseline (HR: 0.658 vs. 0.492, p=0.0022, d=1.18), a contamination effect from propagating erroneous agent states. SSVP avoids this failure mode while showing modest, consistent reduction (HR: 0.463, d=0.30) and achieves significantly lower hallucination than full-broadcast (p=0.0005, d=1.47) using 58% fewer API calls. The contamination effect does not replicate in the software domain, where all conditions converge to low HR (<0.2), confirming it is specific to tasks where one erroneous shared belief cascades across evaluation dimensions. Our results reframe hallucination mitigation as a distributed systems problem and establish context synchronization as a first-class primitive in multi-agent LLM design. --- ### UnityShots: Memory-Driven Multi-Shot Audio-Video Generation with Boundary-Aware Gating Authors: Jiehui Huang, Yuechen Zhang, Bin Xia, Jiahao Wang, Xu He, Zhenchao Tang, Meng Chu, Xin Tao, Pengfei Wan, Jiaya Jia Published: 2026-06-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.21661v1 Generating a coherent multi-shot video requires structured cross-shot memory. Subject appearance, scene context, and speaker identity must persist across cuts. Existing approaches either train end-to-end over fixed-length sequences and cannot scale, generate shot-by-shot with memory banks that grow linearly, or orchestrate pretrained generators under an LLM planner without a multi-shot-aware backbone. We present UnityShots, a memory-driven multi-shot audio-video generation system built on LTX-2.3, trained on annotated cinematic and music-video shots. The video stream maintains two fixed-size slots, a long-term memory (LTM) slot anchored to the opening shot and a short-term memory (STM) slot holding the immediately preceding tail, both updated at every cut by a boundary-conditioned gate that fuses visual cut probability and beat-tracker signals. The audio stream injects a reference speaker token at every shot to preserve vocal timbre without a sliding audio bank. A discrete cut-type prior, learned through AdaLN, becomes an inference-time control knob over transition strength. We release a benchmark of $200$ multi-cultural multi-shot sequences spanning six ethnic regions and ten or more languages, with per-shot reference identities, reference audio, and per-boundary transition labels. Evaluated across I2V, T2V, and R2V conditioning modes, UnityShots leads open-source baselines on every cross-shot coherence metric and matches the strongest closed-source system on the multi-shot axes. --- ### Chehre: An Emoji-Prompted Dataset to Explore Perceptual Flexibility in Video Language Models Authors: Bita Azari, Zoe Stanley, Avneet Batra, Poorvi Bhatia, Hali Kil, Manolis Savva, Angelica Lim Published: 2026-06-19 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2606.21657v2 Do people perceive the same facial expression in the same way? Should we expect vision models to be flexible in how they perceive facial expressions? Facial expressions are nonverbal social signals used in human interaction, but facial expression recognition datasets often focus on a single deterministic annotation per sample. We introduce Chehre, an emoji-prompted video dataset with a wide range of dynamic facial expressions for exploring perceptual variation. In Chehre, 203 participants were prompted to express and record 40 facial emojis. Later, their facial motions were transferred onto synthetic faces to preserve privacy. A separate group annotated the videos, resulting in 2,111 videos annotated by 1,242 perceivers, with ~30 annotators per video. Chehre enables us to define a new task: "distributional expression recognition", which tests whether a model can reproduce the variation observed across annotator responses. We test a selection of video language models on our task. Interestingly, we find that persona prompting can act as a controllable way to shift model perception while helping models better capture the variation observed across human annotators. The dataset and code are available at https://chehre-dataset.github.io/. --- ### Chehre: An Emoji-Prompted Video Dataset for Perceptually Diverse Facial Expression Recognition Authors: Bita Azari, Zoe Stanley, Avneet Batra, Poorvi Bhatia, Hali Kil, Manolis Savva, Angelica Lim Published: 2026-06-19 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2606.21657v1 Facial expressions are nonverbal social signals used in human interaction, but facial expression recognition datasets often focus on static images, basic emotion categories, or single deterministic annotations. We introduce Chehre, an emoji-prompted video dataset for analyzing dynamic facial expressions across a wide range of expressions for exploring inter-individual perceptual diversity. In Chehre, participants were prompted to express and record 40 facial emojis. Later, their facial motions were transferred onto synthetic faces to preserve privacy. A separate group of annotators analyzed the anonymized videos using emoji and label annotations, resulting in 2,111 high quality videos collected from 203 performers and validated by 902 annotators. We define two benchmark tasks: dominant expression recognition, which tests whether models recover the top human-rated labels, and distributional expression recognition, which tests whether models capture the diversity of human responses. We benchmark recent vision-language models using random sampling and persona prompting to generate multiple predictions per video. Results show that both tasks are challenging: among the models evaluated, the best-performing model achieves only 32.5% Top-1 accuracy on dominant expression recognition and a Spread Ratio well below the human reference on distributional recognition. Chehre provides a benchmark for evaluating diverse, dynamic, and distributional facial expression recognition --- ### PaaF: Raising the perceived quality of INR-Based Image Compression Authors: Lorenzo Catania, Dario Allegra Published: 2026-06-19 Categories: eess.IV, cs.CV, cs.MM arXiv: https://arxiv.org/abs/2606.21655v1 Implicit Neural Representations (INRs) have recently emerged as a promising paradigm for image compression, offering a fundamentally different approach from traditional and learned codecs. Nevertheless, INR-based methods for image compression suffer from long encoding times and a consistent performance gap in classic quality metrics such as PSNR. In this work, we explore the potential of purely INR-based compression methods and we propose PaaF (Picture as a Function), a novel INR-based image codec that introduces improved architectural design, adaptive quantization, and an efficient entropy coding scheme. These components are designed to enhance rate-distortion performance while preserving the simplicity and parallelizability of INR-based decoding. Experimental results demonstrate consistent improvements over existing INR-based methods in both quantitative metrics and perceptual quality. These findings highlight the potential of INR-based approaches and contribute to narrowing the gap between functional representations and more established compression paradigms. --- ### ChainWorld: Composing Long-Horizon Desktop Workloads from Atomic OSWorld Tasks Authors: Vincent Siu, Manasi Sharma, Dawn Song, Daniel Yue Zhang, Chenguang Wang, Ying Liu Published: 2026-06-19 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2606.21654v2 Computer use agents are evaluated almost exclusively on atomic desktop tasks, but realistic desktop work requires sustaining state across multiple objectives. We study this gap with ChainWorld, which composes atomic OSWorld tasks into long horizon desktop workloads through directional compatibility search while preserving the source evaluators. The resulting workload contains 347 chains of length two to four and compares two renderings of the same task sequence. In single turn evaluation, all tasks are presented together in one prompt. In multi turn evaluation, tasks are revealed one at a time. Across four current computer use agents, maximum chain completion is 31%. Multi turn evaluation improves completion for three models, but both protocols remain challenging. The two protocols also expose different failure profiles. Single turn failures concentrate on artifact precision, while multi turn failures more often reflect session management problems such as fragmented progress and later turn disengagement. --- ### ChainWorld: Composing Long-Horizon Desktop Workloads from Atomic OSWorld Tasks Authors: Vincent Siu, Manasi Sharma, Dawn Song, Daniel Yue Zhang, Chenguang Wang Published: 2026-06-19 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2606.21654v1 Computer use agents are evaluated almost exclusively on atomic desktop tasks, but realistic desktop work requires sustaining state across multiple objectives. We study this gap with ChainWorld, which composes atomic OSWorld tasks into long horizon desktop workloads through directional compatibility search while preserving the source evaluators. The resulting workload contains 347 chains of length two to four and compares two renderings of the same task sequence. In single turn evaluation, all tasks are presented together in one prompt. In multi turn evaluation, tasks are revealed one at a time. Across four current computer use agents, maximum chain completion is 31%. Multi turn evaluation improves completion for three models, but both protocols remain challenging. The two protocols also expose different failure profiles. Single turn failures concentrate on artifact precision, while multi turn failures more often reflect session management problems such as fragmented progress and later turn disengagement. --- ### EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory Authors: Chang Nie, Chaoyou Fu, Junlan Feng, Caifeng Shan Published: 2026-06-19 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.21649v2 Existing embedding models are inherently static: they encode text segments in isolation, ignoring their surrounding context and temporal order. This paper introduces EvoEmbedding, a novel embedding model that generates evolvable representations for retrieval. It is tailored for long-context scenarios, where information is dynamic, sequential, and requires continuous state tracking. Our design is simple: EvoEmbedding maintains a continuously updated latent memory as it sequentially processes inputs, and uses it alongside the raw content to jointly generate evolvable embeddings. Consequently, for the same query, our model adapts its representation to retrieve distinct targets based on the evolving context, going beyond static semantic search. To equip the model with this capability, we construct EvoTrain-180K, a diverse dataset for the joint optimization of latent memory and retrieval. Furthermore, we introduce a memory queue to prevent representation collapse during recurrent encoding, alongside segment-batching techniques that tackle significant length variance and accelerate training by 3.8$\times$. Extensive experiments show that our model not only outperforms larger-scale specialists (e.g., Qwen3-Embedding-8B and KaLM-Embedding-Gemma3-12B) across a range of long-context retrieval benchmarks, but also generalizes well to downstream tasks (e.g., personalization) with contexts 10$\times$ longer than its training window. Notably, EvoEmbedding seamlessly integrates into agentic workflows to boost performance. For instance, a naive RAG pipeline equipped with our model surpasses dedicated agentic memory systems. Project Page: https://clare-nie.github.io/EvoEmbedding/. --- ### Behavioral and Representational Evidence of Binomial Ordering Preferences in Large Language Models Authors: Zhiqing Yang, Yilun Liu, Yunpu Ma, Volker Tresp, Hinrich Schütze Published: 2026-06-19 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.21645v1 Large language models (LLMs) can readily reproduce conventional expressions, yet their ability to model gradient frequency distributions remains underexplored. We investigate this using linguistic binomials, such as men and women, where both word permutations are grammatically valid but exhibit distinct, cross-linguistic variations in conventionality. We formalize binomial ordering as a distributional alignment problem, and construct a multilingual dataset of 600 binomial pairs across 8 languages. With categorical and distributional metrics, we measure and compare the corpus-derived preferences with model-induced ordering probabilities of 6 open-weight LLMs. While models often behaviorally recover the dominant corpus-preferred order, particularly for strongly conventionalized pairs, they align less well with the exact corpus preference distributions. This suggests that apparent directional order overstates how faithfully LLMs capture the statistical nuances of language use. Sparse probing verifies that the concept of preference strength is partially encoded among middle-to-late layers, and steering along probe-derived directions alters model-induced ordering distributions, demonstrating that the statistical behavioral preference of LLMs can be mechanistically measured and manipulated via internal representations. --- ### When Is an LLM Worth It for Hyperparameter Optimization? A Budget-Matched Study on Tabular Data Finds the Warm-Start Is a Default Configuration, Not the Model Authors: Carson Rodrigues, Oysturn Vas, Isaiah Abner DCosta, Nithish Kumar Prabhakaran Published: 2026-06-19 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.21641v2 Large language models (LLMs) have been proposed as hyperparameter-optimization (HPO) advisors that "warm-start" search from prior knowledge, proposing strong configurations in very few evaluations. We test that claim under a budget-matched, multi-seed protocol on eight PMLB tabular benchmarks, comparing an LLM advisor (LLM-OptFlow) against four classical baselines (random search, Optuna-TPE, Gaussian-process Bayesian optimization, and successive halving) over one shared search space, with paired tests and bootstrap 95% CIs across 8 x 5 = 40 (task, seed) units. The finding is cautionary. The advisor's strong first point is not an LLM output at all: like prior LLM-HPO systems the loop is seeded with a fixed default configuration, evaluated before any model call, which alone reaches 88.7% mean best-CV, identical to within 0.01 pp across all seven advisor models tested. The LLM's own proposals add only +0.40 pp of cross-validation accuracy over that seed and nothing on held-out test (LLM-Default = -0.01 pp, p = 0.92). When the same seed is granted to classical search, the apparent lead collapses: against seeded random search it leads by +0.20 pp at 2 evaluations, is tied by 5, and is behind by 12 (-0.37 pp). Without the seed, classical search ties the advisor by 12 evaluations and beats it by 40 (+0.6 to +0.8 pp, p <= 1e-4). Two LLM-specific behaviors survive: a single-task exploration failure (vehicle), and a rule-based confidence filter that removes ~33% of wasted compute without changing accuracy. The recommendation is deflationary: on tabular HPO, seed classical search with a sensible default; an LLM advisor adds no measurable generalization benefit and is overtaken within a handful of evaluations. We release the harness and a script that reproduces every statistic. --- ### A new classification method based on Minimum Spanning Trees Authors: Julio González-Díaz, Beatriz Pateiro-López, Iria Rodríguez-Acevedo Published: 2026-06-19 Categories: cs.LG, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2606.21639v2 Minimum Spanning Trees have been used in unsupervised learning, particularly in clustering tasks, due to their ability to recognize clusters by removing edges that are considered inconsistent in defining those clusters. This paper aims to study the use of Minimum Spanning Trees in supervised learning. Specifically, we propose a classification algorithm based on Minimum Spanning Trees. To improve its performance, we introduce a robust version of the method that is also computationally more efficient. We evaluate the effectiveness of our proposed method through an extensive simulation study. We also apply the proposed methodology to a real-world case study involving aircraft trajectories. --- ### Toward Open Weight Models Without Risks: Separating Public and Private Capabilities in LLMs Authors: Charbel El Feghali, Arkil Patel, Nicholas Meade, Spandana Gella, Verna Dankers, Siva Reddy Published: 2026-06-19 Categories: cs.CR, cs.CL arXiv: https://arxiv.org/abs/2606.21638v1 Open-weight Large Language Models (LLMs) enable scientific progress and broad deployment. However, they make it difficult to control access to sensitive capabilities. Current practice either suppresses dangerous capabilities before release or mediates access through closed services that use specialized model variants, input/output monitors, and API permissions. The former is susceptible to jailbreaks while sacrificing capability for all users to mitigate the risks posed by a few, and the latter is fundamentally incompatible with open-weight release. In this paper, we propose Tiered Language Models (TLMs), where a single set of released weights supports multiple capability levels. In its default public configuration, a TLM behaves as a conventional LLM. A compact secret key specifies a permutation over a small parameter subset, inducing an alternative computation graph over the same weights that exposes additional capabilities. We develop a training protocol that jointly pretrains both configurations from scratch, then fine-tunes the keyed configuration on private data with regularization to preserve the public model's behavior. We pretrain 180M- and 650M-parameter TLMs and demonstrate that the keyed configuration can acquire a new language, gain instruction-following ability, and memorize private factual knowledge, whereas the public configuration exhibits none of these capabilities. Moreover, we show that our approach extends naturally to multiple hierarchical tiers. Because authorization operates on the model's weight structure rather than in the input space, the mechanism resists fine-tuning-based extraction and partial key compromise. In general, TLMs take a step toward reconciling open-weight release with selective capability control. --- ### Time-Frequency Weighted Losses for Phoneme Reconstruction in DNN-Based Speech Enhancement Authors: Nasser-Eddine Monir, Paul Magron, Romain Serizel Published: 2026-06-19 Categories: cs.SD, cs.CL arXiv: https://arxiv.org/abs/2606.21635v1 Conventional training losses for speech enhancement based on the signal-to-distortion ratio (SDR) treat all time-frequency (TF) regions uniformly, overlooking the fine-grained spectral cues that are relevant to specific phoneme intelligibility. We propose a TF weighting framework that modulates the SDR objective based on local speech presence, speech-to-interference ratio (SIR), and spectral flux. By integrating these factors into a differentiable objective, the framework emphasizes TF bins with high speech-noise competition while also accounting for transient cues such as consonant bursts. Experimental results show that our approach improves objective frequency-weighted enhancement metrics, as well as phoneme recognition accuracy, particularly for consonants. Spectral analysis shows better reconstruction of mid-frequency structures at less adverse SIR. --- ### HERALD: High-Throughput Block Diffusion LLM Serving via CPU-GPU Cooperative KV Cache Retrieval Authors: Omin Kwon, Doyeon Kim, Jongseok Park, Seung Yul Lee, Ion Stoica, Jae W. Lee Published: 2026-06-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.21633v2 The KV cache dominates GPU memory in long-context LLM serving, crowding out batch capacity and leaving GPU compute idle. Offloading the cache to CPU DRAM restores capacity, but the limited PCIe bandwidth forces state-of-the-art offloading systems to pair it with sparse attention, fetching only a small critical subset of the cache to the GPU. These systems, however, follow the KV access pattern of autoregressive decoding, in which the critical set changes at every token: selection and fetching recur at every decoding step, and throughput remains capped by PCIe bandwidth rather than by either processor. Block diffusion LLMs(block dLLMs), which decode a block of B tokens over T denoising steps, exhibit a different KV access pattern that opens a new opportunity for offloading. Recent sparse block dLLM methods have shown that sparse inference separates into a selection phase that scans the full KV cache once per block and a denoising phase that reuses the selected small subset T times. This asymmetry aligns with the compute and memory asymmetry of a CPU-GPU system, making it advantageous to run selection on the CPU and denoising on the GPU: the critical KV cache then crosses PCIe only once per block, removing the interconnect as the bottleneck. We present HERALD, to our knowledge the first KV offloading system designed for block dLLMs. HERALD resolves the two obstacles of this mapping, the serialized dependency between the phases and the compute-bound B-query selection on the CPU, by overlapping the phases with a draft block, reducing the selection cost with a single (MASK) query, and executing both as a dual-stream pipeline over double-buffered sparse KV pools. On two production block dLLMs, HERALD sustains near-lossless accuracy at a 5% KV budget and reaches up to 2.28x the decode throughput of GPU-only serving, with gains that widen with context length. --- ### CuratorKIT : Data Curation and Synthetic Data Generation for LLM Post-Training Authors: Soham Bhattacharjee, Karun Sharma, Vinay Kumar Sankarapu, Pratinav Seth Published: 2026-06-19 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.21631v1 Data curation is a critical part of post-training pipelines for large language models, yet existing tools often treat ingestion, deduplication, synthetic generation, and quality filtering as separate stages. This fragmentation makes it difficult to audit pipeline decisions or understand why individual samples are rejected. CuratorKIT is an open-source Python library that covers this full lifecycle in a single configurable pipeline. The framework is composed of six source format readers and automatic schema detection, a pre-generation data hygiene layer for credentials, PII, and toxic content, eight LLM-powered generation tasks, three complementary quality gates with provenance-exact hallucination verification, structured adaptive recovery, and five training-ready export formats compatible with TRL, Unsloth, and AlignTune. Every pipeline decision is recorded in an append-only per-sample provenance chain, and rejected samples carry structured failure reasons rather than being silently discarded. CuratorKIT supports 100+ LLM providers through LiteLLM, exposes both a Python API and a YAML-driven CLI, and is designed for practitioners who need reproducible, auditable data pipelines at scale . --- ### Counsel: A Meta-Evaluation Dataset for Agentic Tasks Authors: Sashank Pisupati, Henry Broomfield, Eujeong Choi, Antonia Calvi, Charlie Wang, Roman Engeler, Max Bartolo, Patrick Lewis Published: 2026-06-19 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2606.21627v1 As agentic systems tackle increasingly complex multi-step tasks, evaluating their trajectories presents a major bottleneck - human annotation of a single trajectory on popular agentic benchmarks can take hours, making it difficult to scale evaluations for measuring performance or curating training data. This has driven widespread reliance on automated approaches such as LLM-as-a-judge (LLMJ) to critique agents at the process and outcome-levels at scale, however, the soundness of LLMJ critiques often goes unmeasured. Here, we introduce Counsel, the first public dataset of meta-evaluations for agentic tasks. Counsel consists of process-level critiques from open-weight LLMJs on two agent benchmarks: tau-bench (customer support agents) and DA-Code (coding agents), and human meta-evaluations of these critiques. Human annotators label critiques on each flagged error as "spot on", "correct location but poor reasoning", or "should not have flagged", achieving reliable inter-annotator agreement (Krippendorff's alpha of 0.78). The resulting dataset stratifies LLMJ critiques by human alignment across both error location within a trajectory and reasoning quality, serving as valuable data to calibrate, improve, or train LLMJs for agents. Comparing open-weight judges, we find that more capable judge models and more reasoning effort both enabled improved human agreement, with the strongest judge reaching ~88% agreement on location and ~65% on reasoning. Counsel is generated using open-weight models and is permissively licensed for broad community use, which we hope will enable rigorous study and improved alignment of LLM-based evaluators for agentic systems. --- ### A DVDrive Approach for doScenes Instructed Driving Challenge Authors: Zijian Fu, Xiangyang Chu, Mengshi Qi, Huadong Ma, Guanghao Zhang, Wei Li Published: 2026-06-19 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2606.21623v1 Instruction-conditioned trajectory prediction is an emerging problem in autonomous driving, where a model predicts the future ego trajectory not only from visual scene context and historical motion, but also from a natural-language maneuver instruction. This paper presents our submission to the doScenes Instructed Driving Challenge, built upon OmniDrive, a vision-language-action driving agent with 3D perception, reasoning, and planning capabilities. We adapt OmniDrive to the doScenes setting by training it on instruction-annotated nuScenes scenes and generating a 6-second ego trajectory represented by 12 future waypoints. To improve multi-view visual grounding, we further introduce a DVPE-style divided-view perception module into the OmniDrive perception head. Instead of attending globally to all camera features, the proposed module groups query features and image tokens into divided local view spaces and performs visibility-aware cross-attention within each view. This design reduces irrelevant cross-view interference and helps the model better align language instructions with local driving-relevant visual evidence. The code is publicly available at: https://github.com/feel12348/doscenes-omnidrive. --- ### Evaluating Document-Tuned Transformer Representations for Person-level Mental Health Assessment Authors: Aaron Marker, Oscar Kjell, Vasudha Varadarajan, H. Andrew Schwartz Published: 2026-06-19 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.21622v1 Person-level psychological assessment requires aggregating meaning across many messages from the same individual, a task that document-level training objectives were not explicitly designed for. We present a systematic, empirical comparison between architecturally matched traditional (a) base-transformers and (b) document-tuned-transformers (further contrastively fine-tuned at the document-level, sometimes referred to as "sentence transformers") under otherwise identical conditions. Comparing layer-wise and overall performance across two longitudinal mental health and psychological datasets, we find document-tuned models demonstrated a consistent improvement over base representations (increase in Pearson r of 13.4%, p=.015). Robustness analyses revealed document-tuned models remained more accurate under perturbations to word deletion, synonym replacement, typo injection, and back translation. Further, hedged language (e.g., `usually') was more characteristic of outcomes in document-tuned embeddings while abundance (e.g., `lot') was more characteristic of base-transformers, suggesting document-tuned models may better capture uncertainty. These results suggest representation choice impacts mental health prediction, document-tuned models often being more adept. --- ### The Alignment Problem in Constrained Code Generation Authors: Matteo Biagiola, Jahrim Gabriele Cesario, Luca Di Grazia, George Zakhour, Guido Salvaneschi Published: 2026-06-19 Categories: cs.SE, cs.LG, cs.PL arXiv: https://arxiv.org/abs/2606.21619v1 Large Language Models (LLMs) have demonstrated strong capabilities in code generation, but their outputs frequently contain syntax or type errors that result in compilation failures. Constrained decoding has been proposed as a solution to mitigate compilation errors by construction, improving functional correctness as a byproduct. However, previous works overlook a critical aspect of constrained decoding: the alignment between constrainer (e.g., types), language model and the target specification language (e.g., TypeScript). Misalignment is caused by the constrainer being incomplete--rejecting programs that belong to the target--or unsound--allowing programs that are not part of the target. The bias created by incompleteness distorts the language model distribution, and can be detrimental for code generation. We evaluate this hypothesis using seven language models, two target languages, two constrainers, enforcing types and syntax during decoding, and we study how language models react to varying levels of incompleteness. On three benchmarks, when the constrainer is incomplete, unconstrained decoding significantly outperforms constrained decoding in terms of functional correctness. Incompleteness pushes the model into low-probability regions of the program space, causing the generation to frequently time out, and reducing functional correctness by up to 97%. These contributions make the community aware of the negative effects of misalignment in constrained decoding, and provide quantitative insights on how to design constrainers that are beneficial for code generation systems with formal guarantees. --- ### CulMind: Benchmarking Multimodal Understanding and Reasoning in Chinese Cultural Heritage Authors: Zhangwei Cao, Shuhan Fan, Yuting Wei, Jiajun Zhang, Yihang Peng, Qi Meng, Yangfu Zhu, Liangbin Yang Published: 2026-06-19 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.21618v1 Evaluating Multimodal Large Language Models (MLLMs) in Chinese Cultural Heritage (CCH) requires fine-grained reasoning over visual, textual, stylistic, and historical clues. However, existing CCH benchmarks mainly emphasize final-answer accuracy, while the accuracy and completeness of reasoning processes remain underexplored. To address this gap, we introduce CulMind and CulMind-R: a high-quality benchmark for multimodal CCH covering 50 tasks from collections of more than 100 museums, and a 24-task reasoning subset that adaptively defines task-specific dimensions for reasoning process evaluation. To evaluate reasoning quality, we propose ReaScore, a task-adaptive metric that evaluates reasoning by automatically weighting task-relevant dimensions. Experiments on 14 leading MLLMs reveal a substantial gap between answers and reasoning, especially on challenging tasks. Further analysis shows that task-adaptive dimension selection and weighting better align evaluation results with expert judgments. Overall, our benchmark and metric support a more expert-aligned assessment of CCH understanding and offer a transferable reference for broader evaluations of cultural heritage. We publicly release the data, code, and evaluation scripts at https://github.com/ZevTsao/CulMind to facilitate reproducible research. --- ### LLM and Human Modes of Representation Authors: Shalom Lappin Published: 2026-06-19 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.21616v1 Much work on the cognitive foundations of AI has focussed on comparisons between the ways in which Large Language Models (LLMs) and humans process information and represent it. One aspect of this comparison involves determining the extent to which LLMs can achieve or surpass human performance on a variety of cognitively interesting tasks. A second explores points of convergence and divergence between LLM and human systems for processing information. Here, I consider some recent research that has addressed both issues in two informational domains. The first is the representation of linguistic knowledge. The second is real world reasoning and planning. While LLMs frequently achieve impressive levels of performance and fluency on linguistic applications, they tend to handle linguistic content in ways that are distinct from human processing. They are also, for the most part, less efficient than humans in learning and generalisation for reasoning tasks. --- ### Cross-Modal Corroboration for Annotation-Free Wildlife Monitoring Authors: Bharath Pillai, Varun Viswapriyan, Christopher Stewart, Tanya Berger-Wolf, Jenna Kline Published: 2026-06-19 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2606.21613v1 Scaling wildlife monitoring for real-world conservation deployments requires automated analysis of smart sensors that operate under severe annotation scarcity. We propose leveraging expert knowledge of species activity patterns as an annotation-free validation signal for multimodal monitoring pipelines. We operationalize agreement as the alignment of independently derived hourly activity curves both with each other and with published behavioral priors-a three-way convergence that rules out shared-data confounds and dataset-internal correlation as alternative explanations. Our vision pipeline combines zero-shot species detection via BioCLIP 2, sliced inference to handle deployment-constrained camera positioning, and geometry-based geographic localization from camera trap imagery. Our acoustic pipeline detects species vocalizations via a fine-tuned classifier. We validate the pipeline on a breeding herd of Milu deer and demonstrate that both modalities independently recover activity patterns consistent with known deer behavioral ecology with minimal manual annotation. The framework applies to species detectable in both visual and acoustic modalities for which behavioral priors are documented in the literature, suggesting a practical path toward self-validating wildlife-monitoring pipelines at conservation scale. --- ### The Two-Hump Problem: Bridging the Difficulty Gap in Mathematical Reinforcement Learning Authors: Lucas Fagan, Michele Tarquini, Ali Shehper, Maksymilian Manko, Angus Gruen, Coco Huang, Giorgi Butbaia, Davide Passaro, Sergei Gukov Published: 2026-06-19 Categories: cs.LG, cs.AI, math.GR, math.GT arXiv: https://arxiv.org/abs/2606.21611v1 Mathematical search problems present a unique challenge for Reinforcement Learning (RL) due to vast search spaces and sparse rewards. In previous works, the Andrews-Curtis (AC) conjecture was established as an illustrative example of such problems. In this work, we identify a critical structural barrier in the AC landscape: a "Two-Hump" distribution, where problem instances are either trivially solvable or effectively impossible, with a scarcity of intermediate "hard-but-solvable" instances required for effective learning. We tackle this challenge through two primary avenues: novel data generation techniques to populate the difficulty gap, and significant algorithmic enhancements including the introduction of supermoves and Transformer-based architectures. We demonstrate substantial performance improvements over previous baselines, and release new comprehensive benchmark datasets including AC-19 (125,192 AC-trivial presentations of varying difficulty with length at most 19) and AC-1M (1,136,154 hard AC-trivial presentations of length at most 30), the first large-scale, publicly available datasets of this kind. --- ### CurvSegFlow: Time-Conditioned Flow Matching for Robust Segmentation of Curvilinear Structures in Noisy Biomedical Images Authors: Sidi Mohamed Sid'El Moctar, Achraf Ait Laydi, Alexandre Beber, Marcus Braun, Zdenek Lansky, Yousef El Mourabit, Helene Bouvrais Published: 2026-06-19 Categories: cs.CV, q-bio.QM arXiv: https://arxiv.org/abs/2606.21608v1 Accurate segmentation of curvilinear structures remains challenging in biomedical imaging due to their thin geometry, complex topology, and sensitivity to noise. This is particularly critical for microscopy images of cytoskeletal network, where low signal-to-noise ratios and dense filament crossings often lead to fragmented or inaccurate segmentation. In this work, we propose CurvSegFlow, a segmentation framework based on time-conditioned flow matching. Instead of predicting a segmentation mask in a single pass, the method models segmentation as a dynamic process that progressively refines a noisy initialization into the target structure through a learned velocity field. The proposed model combines a U-Net backbone with triple-term loss function and temporal embeddings to guide the refinement process across reconstruction stages. This formulation enables gradual error correction and improves the continuity of thin structures. CurvSegFlow is evaluated on multiple synthetic and real microtubule datasets, as well as on public benchmarks of retinal vessels, corneal nerves and coronary arteries. Across datasets, the method achieves competitive or superior performance compared to established segmentation models, with consistent improvements in precision and structural continuity, particularly under low signal-to-noise conditions. These results show that flow-based iterative refinement provides a robust and general framework for curvilinear structure segmentation. Overall, the proposed approach improves segmentation quality in challenging imaging conditions and generalizes effectively across modalities without architectural changes. --- ### T-MOR: Learning Motion-Aware Skeleton Representations for Human Action Recognition Authors: Di Yang, Mahmoud Ali, Quan Kong, Gianpiero Francesca, Francois Bremond Published: 2026-06-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.21607v1 Vision-language models such as CLIP have recently achieved strong performance on a wide range of visual understanding tasks. However, most existing models rely primarily on appearance-level supervision from images or videos, and do not explicitly model human motion, which is essential for fine-grained and human-centric action recognition task as actions are defined by temporally structured and physically grounded body movements. To address this problem, we propose Transferable skeleton MOtion Representation (T-MOR), a motion-aware framework that learns transferable action representations from skeleton sequences with the aid of video and language supervision during training. T-MOR adopts a multi-modal contrastive learning scheme that aligns skeleton motion with visual and textual representations, while performing inference using only lightweight skeleton inputs. To support large-scale pre-training, we construct PoseCap-1M, a new dataset that contains over one million synchronized video, skeleton, and text triplets covering diverse human activities. We evaluate T-MOR on a range of human-centric action recognition benchmarks, including action classification and frame-wise temporal detection. Experimental results show that T-MOR consistently improves performance across multiple datasets, such as Toyota Smarthome, Penn Action, UAV-Human, TSU, and Charades. In addition, T-MOR demonstrates strong generalization ability in few-shot and zero-shot settings, highlighting the effectiveness of motion-centric and embodied representations for transferable action understanding. ---