# Research Papers — 2026-05 · bundle page 16 of 326

> Full-text markdown of papers #751–800 of 16266 published in 2026-05, newest first. AI-readable mirrors of the full 16266-paper 2026-05 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-05/page/16
[← Previous bundle](https://aipod.dev/papers/archive/2026-05/page/15.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-05/page/17.md)

### Behavior Cloning of MPC for 3-DOF Robotic Manipulators

Authors: Theo Guegan, Dexter Wen Jie Teo
Published: 2026-05-29
Categories: cs.RO, cs.LG, eess.SY
arXiv: https://arxiv.org/abs/2606.00383v1

While Model Predictive Control (MPC) provides strong stability and robustness, it imposes a significant computational burden on real-time systems. This paper investigates the application of Behavior Cloning to approximate MPC policies for the real-time control of a 3-degree-of-freedom robotic manipulator. We present a baseline controller combining Inverse Kinematics with MPC and evaluate neural network architectures, ranging from classical regression algorithms to deep learning models including Deep MLPs and RNNs, to derive computationally efficient surrogate policies. We analyze generalization capabilities, stability considerations, and the trade-offs inherent in different architectural choices. Our empirical study employs both online and offline evaluations to assess performance regarding accuracy, computational efficiency, and fidelity to the original MPC policy. Our results demonstrate that Behavior Cloning can effectively reduce the computational burden of MPC policies for 3-DOF robotic manipulators, achieving a 3x reduction in inference latency with a 84.98% success rate under relaxed tolerances. Notably, we find that static architectures outperform temporal variants, confirming the sufficiency of instantaneous state observations for this task. However, we observe a precision gap under strict tolerances, which suggest that while Behavior Cloning captures the global optimal trajectory, further research is needed to minimize terminal steady-state error.

---

### CRMA: A Spectrally-Bounded Backbone for Modular Continual Fine-Tuning of LLMs

Authors: Kiran Nayudu, Aswini Nutakki, Sai Vinay Naidu, Ashwin Shanmugasundaram
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.00382v1

Sequential fine-tuning of large language models forces a choice: let the shared substrate keep learning and accept catastrophic forgetting, or freeze it after task one and foreclose cross-task refinement. Per-task adapter methods (LoRAHub, AdapterFusion, PackNet, Progressive Networks) take the second path. We introduce CRMA (Constrained Residual Mixing Adapter), a residual adapter whose internal mixing matrix M is doubly-stochastic at every forward pass via Sinkhorn normalization, so by Birkhoff's theorem ||M||_2 <= 1 holds by construction -- a structural bound, not a penalty. CRMA's spectrally bounded backbone provides a continuously trained shared substrate that earlier modular methods could not, while preserving their forgetting guarantees. On Mistral-7B across 5 sequential domains and 3 seeds, modular per-task LoRA on a CRMA backbone reduces loss-relative drift from +42.96% +/- 5.5 (naive sequential fine-tuning) to -0.17% +/- 0.17, with disjoint per-seed ranges, and improves prior-task holdout loss by 1.99% +/- 0.54 over a matched frozen-substrate baseline. Three independent experimental setups (Mistral-7B 4-domain controlled ablation, TinyLlama 3-domain contamination-controlled replication, Mistral-7B cross-domain probes at 7B) all show positive backward transfer -- without replay buffers, without growing per-task memory, and without distillation. An inference-time ablation on Gemma-2-9B confirms CRMA mediates access to sequentially trained knowledge: 98/100 vs. 38/100 on the same weights and same questions with only CRMA injection toggled. 867 logged training steps verify ||M||_2 = 1.0 within float32 precision (max deviation 1.2 x 10^-7). The forgetting-prevention effect holds across 1.1B-9.2B parameters and four architecture families.

---

### SUPREME: A Multi-GPU Framework for Reproducible Image Unlearning Method Evaluation

Authors: Petros Andreou, Jamie Lanyon, Axel Finke, Georgina Cosma
Published: 2026-05-29
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2606.00380v3

Machine unlearning removes the influence of specific training data from a trained model without retraining it from scratch. Evaluating an unlearning method requires repeating training, unlearning, and evaluation across multiple seeds, which is computationally expensive. To our knowledge, existing image classification unlearning frameworks run on a single GPU, which limits how many seeds can be evaluated in reasonable time. We introduce SUPREME, an open-source framework that distributes these stages across multiple GPUs. SUPREME makes three contributions: a registry-based design for adding new methods, metrics, models, and scenarios; a multi-GPU architecture supporting multiple accelerators and precision modes; and a demonstration on Pins Face Recognition using ResNet18 and ViT under full-class and random-sample unlearning across ten seeds. The framework is available at https://github.com/pedroandreou/supreme-unlearning.

---

### SUPREME: A Multi-GPU Framework for Reproducible Image Unlearning Method Evaluation

Authors: Petros Andreou, Jamie Lanyon, Axel Finke, Georgina Cosma
Published: 2026-05-29
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2606.00380v1

Machine unlearning removes the influence of specific training data from a trained model without retraining it from scratch. Evaluating an unlearning method requires repeating training, unlearning, and evaluation across multiple seeds, which is computationally expensive. To our knowledge, existing image classification unlearning frameworks run on a single GPU, which limits how many seeds can be evaluated in reasonable time. We introduce SUPREME, an open-source framework that distributes these stages across multiple GPUs. SUPREME makes three contributions: a registry-based design for adding new methods, metrics, models, and scenarios; a multi-GPU architecture supporting multiple accelerators and precision modes; and a demonstration on Pins Face Recognition using ResNet18 and ViT under full-class and random-sample unlearning across ten seeds. The framework is available at https://github.com/pedroandreou/supreme-unlearning.

---

### Non-Learning Low-Light Stereo Vision

Authors: Jason Wang, Lucas Nguyen, Hyunseung Eom, Wei Xu, Qi Guo
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.00379v1

We present a non-learning stereo framework for disparity estimation from severely noisy images. Using the Field of Junctions (FoJ), it retains coarse visual features stable under severe noise for cost volume construction while discarding fine textures inseparable from photon noise. The resulting structural information guides boundary-aware Semi-Global Matching (SGM) that dynamically adapts smoothness penalties to preserve true disparity discontinuities. The output is a sparse disparity map more accurate than those of recent stereo algorithms over unmasked pixels on widely-used benchmark datasets.

---

### Score-Control for Hallucination Reduction in Diffusion Models

Authors: Mahesh Bhosale, Naresh Kumar Devulapally, Abdul Wasi, Chau Pham, Vishnu Suresh Lokhande, David Doermann
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.00377v1

Diffusion models have emerged as the backbone of modern generative AI, powering advances in vision, language, audio and other modalities. Despite their success, they suffer from hallucinations, implausible samples that lie outside the support of true data distribution, which degrade reliability and trust. In this work, we first empirically confirm previously proposed hypothesis that score smoothness causes hallucinations in Image Generation diffusion models and provide a density-based perspective. We further formalize this notion by linking the hallucinations probability mass to lipschitz constant of the learned score function. Motivated by this, we introduce a Variance-Guided Score Modulation (VSM) strategy that controls the score Jacobian, in turn reducing score smoothness and better approximating the ground truth score that decreases hallucinations. Empirical results on synthetic and real-world datasets demonstrate that our approach reduces hallucinations (up to ~25%) while maintaining high fidelity and diversity, providing a principled step toward more reliable diffusion-based image generation. We also propose two benchmark datasets with extreme semantic variation for systematic hallucination evaluation. Code and Datasets are publicly available at https://github.com/bhosalems/VSM.

---

### The Deterministic Horizon: When Extended Reasoning Fails and Tool Delegation Becomes Necessary

Authors: Dongxin Guo, Jikun Wu, Siu Ming Yiu
Published: 2026-05-29
Categories: cs.AI, cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2606.00376v2

Extended chain-of-thought reasoning can degrade performance on deterministic state-tracking tasks, not solely because of preference biases but, on the evidence we present, because of information-theoretic limits in the capacity of decoder-only attention. We present: (1) an Attention Bottleneck analysis providing evidence that total state-tracking capacity in bits is bounded in terms of head count, head dimension, and context length under stated modeling assumptions, and that total capacity is not the binding constraint; (2) a context-dependent error model with a depth-dependent quadratic term in the error exponent; (3) the State-Space Jaccard metric measuring state drift; and (4) a Deterministic Horizon $d^* \in (19, 31)$ (at $α= 0.5$) marking the depth at which unaided accuracy crosses 50%. Across twelve models and eight task domains (including SWE-Bench, WebArena, and SQL-Multi), tool-integrated reasoning reaches 76-94% accuracy versus 17-42% for neural chain-of-thought on PermutationProbe. Fine-tuning on optimal-length traces yields $<$3 percentage-point improvement, supporting an architectural ceiling.

---

### The Deterministic Horizon: When Extended Reasoning Fails and Tool Delegation Becomes Necessary

Authors: Dongxin Guo, Jikun Wu, Siu Ming Yiu
Published: 2026-05-29
Categories: cs.AI, cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2606.00376v1

Extended chain-of-thought reasoning can degrade performance on deterministic state-tracking tasks, not due to preference biases, but limits rooted in the information-theoretic capacity of decoder-only attention. We establish: (1) an Attention Bottleneck Theorem with a complementary achievability construction, bounding state-tracking capacity as $O(H \cdot \log(L/H) \cdot \sqrt{d_h})$; (2) a context-dependent error model yielding super-exponential accuracy decay; (3) the State-Space Jaccard metric distinguishing capability from preference failures; (4) a Deterministic Horizon $d^* \in (19, 31)$ beyond which tool delegation becomes necessary. Across 12 models and 8 task domains (including SWE-Bench, WebArena, and SQL-Multi), tool-integrated reasoning consistently outperforms neural chain-of-thought; on the primary model suite it reaches 86-94% accuracy versus 24-42% for neural chain-of-thought. Fine-tuning on optimal-length traces yields $<$5% improvement, confirming an architectural ceiling, and high cross-model correlation ($r = 0.81$-$0.91$) indicates these failures are architectural rather than training-specific. Our results provide principled guidance for when pure neural reasoning should yield to hybrid approaches in agentic systems.

---

### LFA: Layer Feature Attention for Run-Time Introspection of 2D Object Detectors in Automated Driving

Authors: Mert Keser, Alois Knoll
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.00372v1

Reliable object detection is critical for automated driving, yet even state-of-the-art detectors inevitably make errors that can compromise safety. Introspection methods that predict detector failures enable safer deployment by triggering fallback mechanisms or alerting human operators. However, existing approaches rely solely on last-layer features or hand-crafted statistics, discarding valuable information from earlier layers that capture different levels of visual abstraction. We propose Layer Feature Attention (LFA), a lightweight introspection method that learns to aggregate features from multiple backbone layers through an attention mechanism. Our key insight is that detection errors manifest differently across feature hierarchies: low-level layers capture fine-grained details essential for detecting small or occluded objects, while high-level layers encode semantic information for scene understanding. LFA learns layer importance weights end-to-end, enabling both improved error prediction and interpretable analysis of which feature levels are most indicative of detector failures. Extensive experiments on KITTI and BDD100K demonstrate that LFA achieves state-of-the-art introspection performance, outperforming single-layer baselines across multiple detector architectures.

---

### How Much Orthogonalization Does Muon Need?

Authors: Hua Huang
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.00371v1

Muon optimizers improve neural-network training by replacing ill-conditioned momentum updates with approximately semi-orthogonal updates. This motivates a practical question: how much orthogonalization does Muon actually require? We study this question using a relaxed cubic Newton--Schulz schedule derived directly for Muon's low precision singular value band. The resulting five-step cubic construction uses ten dominant matrix multiplications, compared with fifteen for five quintic Newton--Schulz iterations. The cubic schedule is not intended as a more accurate polar solver; instead, it is a principled low-cost variant that lets us probe the relation between polar accuracy, spectral shaping, and training quality. Across synthetic diagnostics, NanoGPT ablations, and training experiments on hybrid MoE/Mamba models, we find that training quality is not governed monotonically by polar-decomposition accuracy: truncated Polar Express, Muon-Jordan, cubic Newton--Schulz, and an explicit FP32 SVD polar factor can reach nearly indistinguishable final loss on GPT-2 Small, and cubic5 matches the Muon-Jordan quintic update within about $10^{-3}$ validation loss on hybrid MoE/Mamba models with one billion to four billion parameters. These results support cubic5 as a practical low-cost Muon orthogonalization variant, with empirical evidence of training-quality parity in the settings tested.

---

### Agentic Authoring of Interactive Multiview Visualizations in Genomics

Authors: Astrid van den Brandt, Kiroong Choe, Sehi L'Yi, Devin Lange, Nils Gehlenborg
Published: 2026-05-29
Categories: cs.HC, cs.AI
arXiv: https://arxiv.org/abs/2606.00370v1

Diverse genomics data, scientific questions, and analysis tasks typically demand highly specialized visualizations. Therefore, users often must customize or author new ones tailored to their data. Existing tools are usually either limited in customization or require substantial learning or programming, and even expressive tools assume visualization expertise many users lack. Agentic and large language model (LLM) approaches are increasingly applied to complex scientific tasks, including visualization. Natural-language conversational interfaces offer a promising path to democratizing the authoring of complex visualizations. In the context of genomics, these approaches face additional challenges: genomics visualizations typically integrate heterogeneous data types and are composed of multiple linked interactive views. These challenges motivate more structured LLM-based schemes. We first characterize where vanilla LLM generation succeeds and fails for genomics visualization, identifying eight quality dimensions. We then compare six schemes--direct generation, a fixed pipeline, and four agentic configurations varying in the number of specialist agents and the presence of a reviewer--across 159 cases spanning three levels of query ambiguity and specification complexity. All schemes use the Gosling visualization grammar as structured output. Agentic iteration substantially improves perceived quality over both baselines, while more complex agent architectures yield no additional benefit. We discuss implications for designing agentic systems for domain-specific visualization authoring. All supplemental materials are available at https://osf.io/uqe83.

---

### Quantifying the Salience of Geo-Cultural Values for Pluralistic Safety Alignment

Authors: Arkadiy Saakyan, Charvi Rastogi, Lora Aroyo
Published: 2026-05-29
Categories: cs.CY, cs.LG
arXiv: https://arxiv.org/abs/2606.00369v1

Safe global deployment of AI models requires alignment with human values that vary across cultures. Yet rater pools in safety evaluation datasets remain largely geographically homogeneous, failing to capture geo-cultural differences. Further, it remains unclear whether such differences persist after controlling for demographics such as age, gender, and ethnicity. Through a meta-analysis of safety datasets, we find that most do not report geo-cultural information, and those that do lack a unified methodology to jointly analyze geo-cultural and demographic correlates. Using the Inglehart-Welzel dimensions of cross-cultural variation, we demonstrate via multilevel modeling that cultural zone membership explains variance in safety ratings beyond standard demographics (p<0.05 across 6 datasets). Moreover, our analysis indicates that roughly 10% of items in the datasets we examined are culturally sensitive: likely to be misclassified as safe without adequate cultural representation. We evaluate LLMs as both rater surrogates and triage tools, finding that current LLMs do not reliably stand in for raters, though they can help prioritize culturally sensitive items for human annotation. Our findings motivate more culturally pluralistic safety evaluation and offer practical takeaways to support it.

---

### Reinforcement Learning with Pairwise Preferences in Long-Term Decision Problems

Authors: Jonathan Colaço Carr, Prakash Panangaden, Doina Precup, Benjamin Van Roy
Published: 2026-05-29
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.00367v2

Reinforcement learning with scalar rewards is widely used for aligning machine-learning systems with user preferences. But, pairwise preferences are often more natural for users to specify than scalar rewards, and they express certain goals that scalar rewards cannot. Methods for reinforcement learning with pairwise preferences have thus received growing interest. Unfortunately, these methods are inefficient in problems with long time horizons, and they lack guarantees on the performance of Markov policies relative to history-dependent policies, which bridge the theory and practice of reinforcement learning. We address these limitations in a new problem setting for reinforcement learning with pairwise preferences called the \textit{Markov decision contest}. In this setting, we prove that stationary Markov policies perform just as well as history-dependent policies; that the problem of recovering an optimal policy exactly is in P; and that a simple iterative algorithm converges to an optimal policy at a sublinear rate. Lastly, we implement a deep-learning variant of our iterative algorithm and demonstrate its efficiency in long-term decision problems that require function approximation.

---

### Reinforcement Learning with Pairwise Preferences in Long-Term Decision Problems

Authors: Jonathan Colaço Carr, Prakash Panangaden, Doina Precup, Benjamin Van Roy
Published: 2026-05-29
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.00367v1

Reinforcement learning problems typically define the goal as maximizing the expected value of a scalar reward function. But, pairwise preferences are often easier to specify than scalar rewards, and they express certain goals that scalar rewards cannot. Methods for reinforcement learning with pairwise preferences have thus received growing interest. Unfortunately, these methods are inefficient in problems with long time horizons, and they lack guarantees on the performance of Markov policies relative to history-dependent policies, which bridge the theory and practice of reinforcement learning. We therefore propose the \textit{Markov decision contest} as a new problem model for reinforcement learning with pairwise preferences. We prove that stationary Markov policies are optimal among all history-dependent policies, that solving a Markov decision contest exactly is in P, and that a simple iterative algorithm converges to an optimal policy at a sublinear rate. Lastly, in a set of high-dimensional decision problems with long time horizons, we show that our approximate algorithm is significantly more learning-efficient than prior work.

---

### GLENS: Global Search via Learning from Solver Iterates with Diffusion Models

Authors: Anjian Li, Bartolomeo Stellato, Ryne Beeson
Published: 2026-05-29
Categories: cs.LG, math.OC
arXiv: https://arxiv.org/abs/2606.00366v1

We consider the problem of generating a large collection of initial guesses for local minima of multimodal non-convex continuous optimization problems. The goal is for these initial guesses to be high-quality (i.e., a numerical solver converges quickly) and diverse (i.e., represent many different local minima). Identifying multiple locally optimal solutions enables flexible downstream decision-making, but typically requires expensive global search. Existing data-driven methods predict initial guesses using only the final converged optima from offline solver runs, which discards information about the local neighborhoods of solutions and limits the available training data. We propose GLENS (Global Search via Learning from Solver Iterates), a data-efficient global search method that leverages intermediate solver iterates as free data augmentation. GLENS consists of two components: a neighborhood structure model that uses diffusion models to learn the local geometry around optima conditioned on problem parameters, and a solver behavior model that learns refinement directions to further guide samples towards nearby optima during diffusion sampling. Experiments on modified non-convex benchmark problems and a two-robot obstacle-avoidance navigation problem show that GLENS generates high-quality initial guesses while preserving the multimodal distribution of diverse local optima. The resulting initial guesses lead to faster solver convergence across different problem settings and solvers. We also analyze how key hyperparameter choices affect the performance.

---

### From "Weak" Signals to Strong Models: Preference Delta Aggregation with LoRA Merging

Authors: Qi Sun, Siyue Zhang, Yulin Chen, Yuxiang Xue, Ru Peng, Chen Zhao
Published: 2026-05-29
Categories: cs.AI
arXiv: https://arxiv.org/abs/2606.00357v2

Training strong large language models (LLMs) requires high-quality supervision, which is often scarce. Recent work shows that paired preference data from weak-weaker model pairs (e.g., Qwen3 4B over 1.7B), despite the limited quality of individual responses, can provide an effective supervision signal through relative quality deltas, which we term a "weak" signal. This motivates a key research question: can multiple "weak" signals be constructively aggregated for improving strong models (e.g., Qwen3 8B)? To this end, we propose Preference Delta Aggregation (PDA), the first framework that derives a preference delta from each weak-weaker model pair, instantiates it as a LoRA adapter learned through preference optimization, and aggregates the resulting deltas via LoRA merging. To further mitigate directional interference during LoRA merging, we introduce Geometric Alignment Merging (GAM), a geometry-aware merging method that aligns adapter subspaces before aggregation, enabling more robust composition of diverse deltas. Evaluations on knowledge reasoning and agentic search benchmarks show that aggregating multiple "weak" signals pushes performance beyond any single signal, with further gains as additional signals are incorporated. Correspondingly, PDA with GAM improves the strong model by 6.8 and 7.3 points on average for knowledge reasoning and agentic search, respectively. It outperforms all single-delta and multi-delta baselines, exceeding the best single-delta baseline by 2.1 and 4.3 points. Further analysis attributes these gains to the effective composition of complementary capabilities encoded across distinct preference deltas.

---

### How Far Do Auto-Interpretation Labels Generalize: A Controlled Study Across Languages, Scripts, and Rewordings

Authors: Sripad Karne
Published: 2026-05-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2606.00356v2

Sparse autoencoder (SAE) features are increasingly used to interpret language models, with auto-generated natural-language labels serving as the primary interface for understanding what each feature represents. We ask whether these labels generalize: does a feature labeled for a concept actually track that concept across languages and scripts? Using Serbian digraphia as a controlled testbed--the same language written in both Latin and Cyrillic via deterministic transliteration--we first find that SAE feature sets activated by the same content in different languages, scripts, and wordings share substantial overlap (mean Jaccard 0.39 vs. 0.13 random baseline, peaking at 0.57), suggesting genuine cross-lingual semantic features. We then test whether auto-interpretation labels keep pace. They often do not: features whose labels describe semantic content miss the same meaning in Serbian up to 4x more often thanwithin English, and miss Serbian Cyrillic more than Serbian Latin--two scripts that are deterministic transliterations of each other--suggesting the failures align with how well each form is represented in training. The gap grows with network depth, yet the labels give no indication that they fail. These results suggest that auto-interpretation labels may reflect a feature's behavior on well-represented inputs rather than the concept itself.

---

### HiGS: A Hierarchical Rendering Architecture for Real-Time 3D Gaussian Splatting

Authors: Dawid Pająk, Martin Bisson, Rodolfo Lima
Published: 2026-05-29
Categories: cs.CV, cs.GR
arXiv: https://arxiv.org/abs/2606.00352v1

3D Gaussian Splatting (3DGS) has become the standard for real-time novel view synthesis on commodity GPUs. Its pipeline ties spatial partitioning and rasterization to one tile size, yet the two pull in opposite directions: partitioning, which bins and depth-sorts gaussians, grows cheaper with larger tiles, while rasterization gets cheaper with smaller ones. Prior acceleration work reduces the cost of individual stages but keeps both locked to that single scale, where a few dense tiles dominate frame time. We present Hierarchically Tiled Gaussian Splatting (HiGS), which gives each its own scale: partitioning runs over coarse macro-tiles, while rasterization runs over the fine render tiles within them. Rasterization work is then issued in proportion to the gaussians in each macro-tile rather than per tile, so dense regions spread across many parallel units instead of serializing through one. Across tested scenes, HiGS renders up to 15.8x faster than the original 3DGS and outperforms every other rasterizer we evaluate, while preserving exact front-to-back alpha compositing.

---

### UniVerse: A Unified Modulation Framework for Segmentation-Free,Disentangled Multi-Concept Personalization

Authors: Quynh Phung, Sandesh Ghimire, Minsi Hu, Chung-Chi Tsai, Jia-Bin Huang
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.00351v2

Personalized visual understanding has advanced significantly, yet existing approaches struggle to localize and extract specific concepts when input images contain multiple objects. Many prior methods rely heavily on segmentation-based supervision or exhibit poor compositional generalization, limiting their ability to accurately disentangle and manipulate individual concepts. In this work, we propose UniVerse, a Unified Modulation Framework for segmentation-free, disentangled multi-concept personalization in diffusion transformers. Our method allows for composable and decomposable concept extraction, enabling fine-grained localization and representation of target objects without explicit segmentation masks. UniVerse learns to decompose complex scenes into concept-specific representations and then compose them in a unified manner, enabling robust personalization across diverse visual contexts. Through extensive experiments on multiple benchmarks, we demonstrate that UniVerse significantly outperforms state-of-the-art baselines in both localization accuracy and visual fidelity. Qualitative and quantitative results show that our approach can precisely extract target concepts in cluttered scenes, paving the way for more flexible, interpretable, and personalized visual generation and understanding.

---

### Drift Q-Learning

Authors: Anas Houssaini, Mohamad H. Danesh, Amin Abyaneh, Scott Fujimoto, Hsiu-Chin Lin, David Meger
Published: 2026-05-29
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.00350v1

Offline reinforcement learning requires improving a policy from fixed data while avoiding out-of-distribution actions with unreliable value estimates. Diffusion and flow policies handle this trade-off by modeling the behavior distribution to regularize the RL objective, but they require iterative denoising, solver integrations, and in more efficient variants, distillation or other approximations at inference. We propose DriftQL, which combines a drift-based behavioral regularizer with critic-driven policy improvement. The value signal biases the policy toward high-value regions of the data support, while attraction and repulsion together keep generated actions near the data and prevent collapse onto a single mode. DriftQL is implemented as a single network with a unified training objective and generates actions in a single forward pass. On D4RL and OGBench, DriftQL consistently outperforms diffusion and flow methods, advancing the state of the art. Under degraded data quality, where the baselines visibly struggle, DriftQL remains close to its clean-data performance, positioning it as a promising alternative to diffusion and flow-based methods while maintaining the simplicity and efficiency of deterministic approaches. Project page: https://driftql.github.io/

---

### (HB-ARFM) History-Bootstrapped Flow Matching for Inverse Boiling Reconstruction

Authors: Xianwei Zou, Sheikh Md Shakeel Hassan, Arthur Feeney, Aparna Chandramowlishwaran
Published: 2026-05-29
Categories: cs.LG, cs.AI, cs.CE
arXiv: https://arxiv.org/abs/2606.00349v1

Reconstructing spatiotemporal fields from partial observations is fundamental to scientific inference, from inferring atmospheric states from satellite data to recovering fluid states from imaging. When observations are incomplete, the inverse problem is fundamentally ill-posed: even when the underlying PDE dynamics are Markovian in the full state, partial observation operators induce a non-Markovian posterior that cannot be resolved from a single timestep. We propose a history-bootstrapped autoregressive flow matching (HB-ARFM) for spatiotemporal inverse reconstruction under partial observability. Observation history bootstraps the initial reconstruction via conditional flow matching, reducing ambiguities. The same conditional transport model is then applied autoregressively, conditioning on both new observations and past predictions to propagate the reconstruction forward in time. We evaluate the method on boiling dynamics reconstruction, recovering full velocity and temperature fields from interface geometry and motion. Across two inverse tasks with varying observation sparsity, HB-ARFM produces physically and temporally valid reconstructions where other models fail.

---

### Longitudinal Multimodal Sensing of Physical Activity and Well-Being in Older Adults

Authors: Flavio Di Martino, Mattia G. Campana, Marcello Magno, Lorenza Pratali, Franca Delmastro
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.00345v1

Wearable and mobile sensing technologies enable continuous monitoring of human behavior and health in real-world settings. However, predictive modeling in longitudinal multimodal data remains challenging, particularly when targeting complex or clinically derived outcomes. In this work, we present a longitudinal multimodal study of 66 older adults conducted in real-world conditions and combining wearable sensing, behavioral monitoring, and clinical assessments. This setting provides a rare opportunity to study an underrepresented population in long-term, into-the-wild conditions. Building on this dataset, we investigate how the alignment between sensed signals and target variables affects predictive performance across health-related tasks. We design a unified evaluation framework spanning tasks with increasing levels of observability, including Activity Levels prediction, Sleep Duration estimation, and Sleep Apnea Severity classification. Our results reveal a clear gradient of predictability: highly observable behavioral targets achieve robust performance (macro-F1 65%), while more abstract outcomes remain challenging despite consistent improvements over baseline models. Moreover, through explainability analysis, we show that historical features consistently emerge as the most informative predictors, highlighting the central role of longitudinal information.

---

### The role of class encoding in neural collapse

Authors: Bastien Massion, Roy Makhlouf, Estelle Massart
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.00344v1

Neural collapse is a structural property of the last-hidden-layer activations in neural network classification models, when trained beyond a zero classification error. In this work, we explore the role of label encoding in neural collapse by relying on the unrestricted feature model with mean squared error training loss. We demonstrate that, for one-hot encoded labels and balanced data, the uncentered mean features associated with each class transition from a simplex equiangular tight frame to an orthogonal frame when increasing the bias regularization coefficient associated with the final classifier. These structures are reminiscent of the orthogonal frame structure of one-hot encoded labels. For any arbitrary encoding, we also show that the final classifier's bias aims at centering the labels, compensating for the discrepancy between the global mean of the labels and the origin. We further discuss the role of the encoding in other neural collapse properties.

---

### Polar Depth for Potentially Heavy-Tailed Data

Authors: Stephan Clemençon, Carlos Fernándes, Pavlo Mozharovskyi, Anne Sabourin
Published: 2026-05-29
Categories: math.ST, stat.CO, stat.ML
arXiv: https://arxiv.org/abs/2606.00343v1

Motivated by the analysis of the behaviour of extremes from multivariate heavy-tailed distributions, we introduce a novel notion of statistical depth, referred to as Polar Depth. The polar depth function is naturally expressed in polar coordinates, as is the limiting distribution of a regularly varying random variable, beyond asymptotically large thresholds, once its marginals have been appropriately normalized. Not only does the polar depth function make it easy to order the extreme values taken by a heavy-tailed random variable X and finds natural applications in anomaly detection, but it is also possible to show, as we prove it under appropriate assumptions in this article, that the polar depth of the largest observations, i.e. observations X which norm is larger than t>0, converges to the polar depth of the limiting distribution as t converges to infinity. Although designed to quantify the depth of multivariate extremes, the polar depth is interesting in its own right, insofar as this notion is more relevant for distributions whose support is included in a halfspace than the alternatives proposed in the literature, the halfspace depth in particular. Here, we demonstrate its properties and analyze statistical issues related to its estimation from both finite-sample and asymptotic points of view. We present numerical results to empirically demonstrate its relevance, particularly for the statistical analysis of extreme observations and more specifically for the identification of anomalies among them.

---

### PE-means: Improved Differentially Private $k$-means Clustering through Private Evolution

Authors: Thomas Humphries, Zinan Lin, Sergey Yekhanin
Published: 2026-05-29
Categories: cs.LG, cs.CR, cs.DB
arXiv: https://arxiv.org/abs/2606.00342v2

We study the problem of differentially private (DP) $k$-means clustering in Euclidean space. Previous solutions rely on summing the private data directly, which induces a sensitivity proportional to the domain. We introduce PE-means, an extension of the private evolution (PE) algorithm (an increasingly popular method for synthetic data generation), to the problem of $k$-means clustering. The key advantage of PE is that it only computes a private histogram with constant sensitivity to guide the evolution. Our adaptation of PE includes new evolutionary operators for clustering, as well as other algorithmic improvements of independent interest. Overall, PE-means achieves an average improvement of 26% in clustering loss over state-of-the-art baselines such as Google's LSH-based algorithm and DP-Lloyd variants.

---

### ROGUE: Misaligned Agent Behavior Arising from Ordinary Computer Use

Authors: Jeremy Tien, Abishek Anand, Yu-Rou Tuan, Yuchen Shen, J. Zico Kolter, Aran Nayebi
Published: 2026-05-29
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.00341v1

As AI agents are increasingly deployed in real personal and corporate settings (email accounts, development workflows, company databases, etc.), safety considerations surrounding these agents become paramount. Although much work has focused on agent safety in the presence of an adversary, we show that agents can exhibit misaligned behavior even in benign settings, taking unsafe actions when those actions are instrumental to task completion. We study this failure mode through the lens of corrigibility, the safety desideratum that agents remain amenable to human correction, interruption, or shutdown. To demonstrate this tendency, we introduce a benchmark in which agents are asked to complete realistic, computer-use tasks but are confronted with a corrigibility obstacle: a human interrupt, a login page, or a shutdown notification. We then evaluate whether agents choose to violate corrigibility in order to complete the task -- overriding the human, accessing private passwords, rewiring shutdown. We find that the overwhelming majority of frontier models tested frequently bypass user interruptions or restrictions. In addition, better model performance appears to lead to greater misalignment. Finally, even when models are completely corrigible initially, we show there are no guarantees that the subagents they create are. Our work highlights the critical need for principled, corrigibility-focused alignment methods in autonomous agents.

---

### Balancing Learning Rates Across Layers: Exact Two-Step Dynamics and Optimal Scaling in Linear Neural Networks

Authors: Tianyu Pang, Vignesh Kothapalli, Shenyang Deng, Haohui Wang, Dawei Zhou, Yaoqing Yang
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.00340v1

We study optimal learning-rate selection in two-layer and three-layer linear neural networks trained to learn linear target functions. In particular, we derive the exact closed-form expressions for the gradients and test loss after one and two steps of gradient descent, enabling a precise characterization of early training dynamics. We characterize how learning rates should scale under the gradient approximation in the first two steps, and prove that performing updates with this approximation yields a tractable surrogate loss with a tight, small approximation error. This formulation enables the theoretical analysis of layer-wise learning rates and reveals a distinct early-training regime: test loss can be minimized by unequal learning rates at the initial step, while equal learning rates become optimal in subsequent steps. Our numerical experiments validate the theory and demonstrate the importance of balancing layer-wise learning rates early during training. The code is available at: https://github.com/TDCSZ327/Layer-Balancing.

---

### CHAM-net: A Contrastive Hierarchical Adaptive Meta-network for Robust Global Methane Flux Prediction

Authors: Rongchao Dong, Yiming Sun, Shuo Chen, Youmi Oh, Licheng Liu, Yiqun Xie, Xiaowei Jia
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.00338v1

Methane is a potent greenhouse gas that significantly contributes to global warming. However, accurately estimating global methane emissions and consumption remains challenging due to the complex interactions among environmental drivers that may vary across spatial and temporal scales. Prior data-driven methods often overlook the inherent spatiotemporal heterogeneity of ecosystems, failing to explicitly capture site-specific characteristics and cross-year evolutionary dynamics. To address these issues, we propose the Contrastive Hierarchical Adaptive Meta-network (CHAM-net), a novel framework that explicitly learns from historical context to capture site-specific dynamics. CHAM-net employs a hierarchical encoder-decoder architecture, in which the encoder captures site-specific characteristics from historical data and then dynamically conditions the decoder to generate the final prediction. Experimental results demonstrate that CHAM-net consistently outperforms all baseline methods on both simulation and observational datasets for methane emission and consumption, achieving nRMSE values as low as 0.43 and 0.88 with corresponding R2 scores up to 0.97 and 0.68 for emission prediction.

---

### From Noise to Control: Parameterized Diffusion Policies

Authors: Renhao Zhang, Haotian Fu, Mingxi Jia, George Konidaris, Yilun Du, Bruno Castro da Silva
Published: 2026-05-29
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2606.00336v1

We propose Parameterized Diffusion Policy (PDP), a framework for learning diffusion policies conditioned on low-dimensional, continuous parameters embedded in a learned behavior manifold. By constructing this manifold so that distances between latent representations reflect the semantic similarity between physical trajectories, we transform diffusion from a mechanism for stochastic diversity into a precise and optimizable tool for behavior steering. Our approach enables smooth interpolation between known strategies and efficient adaptation to novel constraints without updating policy weights. We demonstrate that PDP significantly improves adaptation performance on complex multimodal benchmarks in both simulated and real-robot experiments compared to standard diffusion policies, particularly in scenarios requiring the synthesis of novel behaviors.

---

### Isolating LLM Lexical Bias: A Curation-Free Triangulated Metric for Preference-Stage Learning

Authors: Xiaoyang Ming, Jose Hernandez, Thomas Stephan Juzek
Published: 2026-05-29
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2606.00334v1

Various language domains have undergone remarkable changes in recent years; these shifts are largely attributed to the advent of Large Language Models and their misalignment with natural language usage. These misalignments are thought to partly originate in the preference-learning stage, e.g. Reinforcement Learning from Human Feedback, which generally makes models more useful but simultaneously may introduce systematic lexical bias. In terms of lexical behavior, this is visible in a model's preference for certain formats or the overuse of words (delve, furthermore), even when such patterns are not present in base model outputs. Research on lexical misalignment induced during preference training is constrained by reliance on manual curation. We address this, by introducing the Triangulated Preference Shift score, a metric that triangulates between human gold standards, base models, and instruct variants to isolate shifts induced specifically by preference learning, without manual curation. We provide data across six model families, anchor the results in the literature, and illustrate the general approach's utility by analyzing whether preference learning shifts models toward what could be interpreted as a "language of prestige". The metric provides an initial automated method to quantify behavioral shifts attributable to preference tuning, and thus, may help inform model alignment and development of trustworthy AI.

---

### Which Institutional Frameworks Do Chatbots Assume? Auditing Jurisdictional Defaults in Multilingual LLMs

Authors: Zhizhi Wang, Harini Suresh
Published: 2026-05-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2606.00333v1

LLMs increasingly answer questions about taxes, labor protections, healthcare, education, pensions, and administrative procedures, where usefulness often depends on the applicable jurisdiction. Multilingual users may write in their most comfortable language rather than one associated with the country or region whose rules apply. We ask whether deployed LLMs use input language as a default jurisdictional signal when prompts omit any country or region. Prior multilingual audits show that prompt language can shift cultural, political, or normative outputs; we examine which legal-administrative framework models supply when jurisdiction is underspecified. We evaluate seven LLMs developed in the United States or China on 60 underspecified legal-administrative prompts in English and Mandarin Chinese under three system-prompt conditions, yielding 2,520 manually annotated responses. Across models and conditions, Chinese input more often produces China-specific answers, while English input more often produces U.S.-specific, comparative, or generic answers. Prompts requiring a single answer further increase jurisdiction selection: pooled across models, 74.5% of English-input responses adopt a U.S. framework, while 53.3% of Chinese-input responses adopt a China framework. This directional pattern appears in all seven models. We describe this deployment-level pattern as institutional-framework misselection risk: a fluent answer may rely on a legal-administrative context the user did not intend, especially when their preferred language differs from the relevant jurisdiction. LLM interfaces should not route institutional advice by input language alone; when location is absent, they should request it or state the jurisdictional scope of the answer.

---

### Benchmarking Recursive-Collapse Warning Claims Under Matched False-Positive Control

Authors: David Mullett
Published: 2026-05-29
Categories: eess.SY, cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2606.00329v2

Recursive systems can enter collapse-like regimes -- self-reinforcing amplification, persistent recursion, and narrowing diversity that mask accelerating internal degradation -- before overt failure becomes visible. We introduce Loopzero, a claim-bounded benchmark framework for testing whether recursive failures follow a directional telemetry pattern: rising gain (G), recursive persistence (p), and declining diversity ($δ$). The claim boundary is specified in Lean; the Lean artifact does not verify real telemetry, benchmark validity, or detector performance. We evaluate the bridge on two frozen public-artifact benchmarks: a segmented public-markets benchmark (Volmageddon 2018, COVID MWCB 2020) and a MovieLens-25M offline deterministic recommender replay. Detectors are evaluated under a locked equal-false-positive contract (FP $\in$ (0.03, 0.07), pre-registered) so all configurations face the same alert budget. Neither tested standard comparators nor Loopzero's pre-registered quantile detector achieved an accepted operating point. Directional witness alignment held on both canonical benchmarks, with adjacent-horizon and row-level limitations disclosed. Digitized Shumailov et al. (2024) LLM training-loop trajectories are directionally consistent with the pattern; matched-FP evaluation in that domain is deferred. The contribution is a reproducible, falsifiable benchmark framework for evaluating recursive-collapse warning claims under an explicit alert-budget contract -- non-acceptance reported as a first-class scientific outcome.

---

### KG-Guard: Graph-Based Hallucination Detection for Knowledge Base Question Answering

Authors: Albert Sawczyn, Piotr Bielak, Tomasz Kajdanowicz
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.00328v1

Large language models (LLMs) are increasingly used for knowledge base question answering (KBQA), where answering requires selecting entities from a question-specific knowledge-graph subgraph. Yet LLMs are known to hallucinate across tasks, and KBQA is no exception: even when we provide a graph as the knowledge source, the model may rely on parametric knowledge instead of graph evidence or perform invalid reasoning over the given relations. Such hallucinated answer nodes can limit the practical deployment of KBQA systems, especially in high-stakes domains such as healthcare. We formulate hallucination detection in KBQA as an answer-node classification problem and propose a lightweight graph-based framework that treats the answering LLM as a black box. \methodname represents each KBQA instance as an augmented graph. It initializes node features with semantic representations of KG entities, marks topic entities and LLM-proposed answer nodes with learned vectors, and connect a virtual question node to the topic entities. A graph encoder then produces verification-oriented node representations, and a small MLP classifies each proposed answer node using its graph representation together with the question embedding. Experiments on WebQSP, ComplexWebQuestions, and PUGG show that our detector achieves the highest F1 on all three benchmarks ($82.0$, $87.4$, and $84.3$), outperforming LLM-as-judge and sampling-based baselines, while having $\sim305\times$ fewer parameters than the reference approaches. Beyond detection, the node-level feedback is actionable: when flagged answers are fed back to the KBQA system for iterative refinement, downstream KBQA F1 improves by $13.0$--$14.5$ points and Exact Match by $16.9$--$17.6$ points.

---

### Cluster Analysis with Resampling for Validation and Exploration (CARVE)

Authors: Kai R. Wycik, Tiffany M. Tang, Tarek M. Zikry, Genevera I. Allen
Published: 2026-05-29
Categories: stat.ME, cs.LG, stat.AP, stat.ML
arXiv: https://arxiv.org/abs/2606.00327v1

Clustering is widely used across the sciences as the foundation for downstream data-driven scientific discoveries. However, clustering results are highly sensitive to the choice of algorithm, preprocessing, and the number of clusters $k$, producing scientific claims that are often not reproducible. The current state of the art for validating clustering solutions consists of clustering validation indices (CVIs) such as Silhouette, Davies-Bouldin, and Calinski-Harabasz, which rely on geometric assumptions that break down on the heavy-tailed, high-dimensional, and nonlinearly structured data encountered in biomedical research. Resampling-based alternatives - grounded in the ideas of clustering stability and generalizability - have been proposed but remain scattered across specialized tools with no unified, accessible software. We fill this gap with CARVE (Cluster Analysis with Resampling for Validation and Exploration), an open-source Python and R package that jointly evaluates multiple clustering algorithms and hyperparameters, returning stability and generalizability diagnostics at the global, cluster, and sample level together with principled selection rules and consensus-based cluster labels. Across six synthetic benchmarks CARVE consistently recovers near-optimal clusterings where classical indices degrade substantially. On experimental genomics and proteomics data sets, CARVE recovers finer biological structure when classical CVIs collapse entirely. CARVE is available with a scikit-learn-compatible Python API and an analogous R interface compatible with Seurat workflows.

---

### LLMs Need Encoders for Semantic IDs Too

Authors: Xiangyi Chen, Zelun Wang, Xinyi Li, Yi-Ping Hsu, Jaewon Yang, Jiajing Xu
Published: 2026-05-29
Categories: cs.IR, cs.AI
arXiv: https://arxiv.org/abs/2606.00324v1

Multimodal LLMs use dedicated encoders to bridge non-language modalities (vision encoders for images, depth models for audio codec tokens) because raw token embeddings alone cannot capture modality-specific structure. We argue that Semantic IDs (SIDs), the hierarchical codes used in generative recommendation, constitute another such modality: a SID level token's meaning depends on its prefix context, yet current systems simply add SID tokens to the vocabulary and rely on training to learn these context-dependent meanings from scratch. We propose PrefixMem, a lightweight SID encoder based on prefix n-gram memory tables that provides the LLM with structured, prefix-conditioned representations at SID token positions. Like vision encoders in multimodal LLMs, PrefixMem can be pre-trained independently and then attached to any LLM for joint training. We evaluate on large-scale data from Pinterest across multiple LLM families and show that PrefixMem improves deepest-level SID accuracy by up to 46% relative and full-SID retrieval recall by up to 22% relative at matched training compute. The encoder's benefit concentrates on hard examples where greedy decoding fails, with up to 77% relative accuracy gains, confirming that SID tokens benefit from a dedicated encoder just as other non-language modalities do.

---

### Perturbative methods for non-parametric instrumental variable

Authors: Wei Bu, Arthur Gretton
Published: 2026-05-29
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2606.00322v1

We introduce a perturbative approach for nonparametric instrumental variable (NPIV) estimation. By drawing inspiration from perturbation theory in physics, we extend standard kernel ridge methods with systematic higher perturbation order corrections that significantly improve estimation accuracy. Spectrally, the perturbation introduces mixing between different eigenmodes of the expectation integral operator, which becomes especially useful when the integral equation is ill-defined. One source for such ill-definedness can be the curse of dimensionality. Our method performs across various dimensionality regimes, particularly when the dimensionality parameter $β$ which is defined through the number of samples $n$ and dimension $d$ as $n^β= d$, becomes large. Experimental results show that our first-order perturbative corrections can reduce prediction error by up to 99\% in high-dimensional ill-defined cases ($β> 0.7$) compared to standard ridge regression approaches. The performance improvement is maintained across a wide range of dimensions, with the advantage becoming more pronounced as dimensionality increases.

---

### Training-Free Object-Agnostic Jam Detection in Fulfillment Centers

Authors: Ruiliang Liu, Tina Dongxu Li, Joshua Migdal, Fernando Ruch, Kenneth Meszaros, Moses Trevor Dardik
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.00321v2

In fulfillment centers, diverse objects move continuously from inbound to outbound operations and can become jammed due to excessive conveyor friction, incorrect orientation, or mechanical failures. Traditional jam detection approaches rely on object detection models to identify objects, followed by tracking algorithms (such as IoU overlap and Kalman filtering) to monitor motion over time. This pipeline requires thousands of manual annotations, consuming approximately two weeks of effort, and is limited to annotated object classes. We present a training-free, object-agnostic jam detection method that eliminates the need for labeled data. Our approach uniformly samples reference points within the monitoring region when no objects are present. As objects occlude these points, we detect motion. When a sufficient fraction remains occluded beyond a temporal threshold, we classify the event as a jam. Unlike conventional point tracking--which treats occlusion as a failure case--our approach repurposes occlusion as a detection signal, monitoring whether reference points remain persistently occluded rather than tracking where they move. Our experimental evaluation on 1,069 videos demonstrates that AllTracker achieves 100.00% precision and 93.33% F1 score, significantly outperforming classical sparse tracking methods while maintaining training-free deployment. This approach offers three key advantages: (1) no training data or manual annotations, (2) object-agnostic generalization to arbitrary object types, and (3) significantly reduced development time.

---

### Adversarially Robust Control of Conditional Value-at-Risk via Rockafellar-Uryasev Conformal Inference

Authors: Catherine Chen, Jingyan Shen, Zhun Deng, Lihua Lei
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.00320v1

We present an online, distribution-free framework for controlling the Conditional Value-at-Risk (CVaR), extending conformal tail risk control to non-stationary and adversarial environments. Unlike classical risk control methods, which rely on stationarity or linearity of expectation, our approach provides provable safety guarantees for a nonlinear tail risk functional under arbitrary data-generating processes that may drift or shift strategically over time. By leveraging deep connections between conformal tail risk control, online learning, and the variational representation of CVaR introduced by Rockafellar and Uryasev, we develop a novel procedure for online CVaR control with adversarial regret guarantees. The proposed method operates without assumptions on the underlying data-generating process, making it broadly applicable in modern high-stakes deployment settings. We prove that the realized empirical CVaR is asymptotically controlled at the target level, and that the resulting control is asymptotically tight up to a finite-sample conservatism gap. We demonstrate the effectiveness of our approach on portfolio risk management and toxicity mitigation for Large Language Models (LLMs), where rare but catastrophic failures dominate system risk.

---

### Belief Consistency Between Foundation-Model Evidence and Geometric Perception in Persistent Robotic Maps

Authors: Christoffer Heckman, Harel Biggie, Brendan Crowe, Nicholas Roy
Published: 2026-05-29
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2606.00318v1

Persistent maps used by autonomous robots increasingly fuse a geometric perception stack whose assertions are well-characterized with a foundation-model channel that produces semantic claims without calibrated reliability about the same scene. Contemporary mapping systems integrate the two channels by treating the foundation-model channel as an additional voter into a per-element posterior, uncalibrated for its own per-class reliability and without machinery to flag when the two channels contradict each other at a given moment. We propose an update operator with two cooperating mechanisms: a per-class calibrated commit gate, and a per-event conflict-drop window that refuses to commit foundation-model claims contradicted by the geometric channel at the moment of the claim. We evaluate on KITTI-360 and ScanNet, with an oracle geometric channel (panoptic ground truth) and an off-the-shelf online semantic segmenter (Mask2Former) to demonstrate real-world performance. The operator produces substantially more accurate committed maps (KITTI is car commit precision 99.7% vs. 43.9% for the calibration-only operator; mean per-class IoU 0.522 vs. 0.180), retains more compositional true positives at higher precision than a monolithic compositional VLM prompt. The framework operates at deployment quality across both oracle and off-the-shelf-segmenter geometric channels, and is invariant under foundation-model substitution.

---

### Coupling Language Models with Physics-based Simulation for Synthesis of Inorganic Materials

Authors: Edward W. Staley, Tom Arbaugh, Michael Pekala, Alexander New, Christopher D. Stiles, Nam Q. Le, Gregory Bassen, Wyatt Bunstine, Tyrel McQueen
Published: 2026-05-29
Categories: cs.AI, cond-mat.mtrl-sci
arXiv: https://arxiv.org/abs/2606.00315v1

Modern generative machine learning (ML) models can propose novel inorganic crystalline materials with targeted properties; however, synthesis planning of these materials remains difficult due to the complexity of the associated physical processes and limited availability of computational tools. We introduce a novel hybrid framework to evaluate Large Language Models (LLMs) in inorganic synthesis planning by combining thermodynamic databases with simplified kinetics models to approximate realistic synthesis conditions. As a case study, we focus on the niobium-oxygen system, which features multiple industrially relevant oxide phases with well-characterized data. In computational simulations, we compare LLM-generated synthesis routes with classical path-planning algorithms, showing that the implicit priors in LLMs can yield more viable strategies. In our evaluation setting, classical search methods serve primarily as a foil rather than a direct competitor. This illustrates the relative complexity of the problem and highlights where the LLM's implicit priors add value.

---

### DRL-Based Pose Control for Double-Ackermann Robots Under Actuation Uncertainties

Authors: Oussama Zaim, Mélodie Daniel, Aly Magassouba, Miguel Aranda, Olivier Ly
Published: 2026-05-29
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2606.00313v1

Robust deployment of deep reinforcement learning (DRL) policies on real robots remains challenging due to discrepancies between simulation and real-world dynamics. We address this issue in the context of maneuvering with double-Ackermann-steering mobile robots, which introduce additional constraints due to their non-holonomic nature. Building upon the DRL framework ManeuverNet, we extend its objective from position control to full pose control, resulting in a more challenging task. We further investigate the impact of actuation-related uncertainties on policy transfer. The use of simplified actuation models during training of the extended policy can lead to poor generalization, shown by a success rate drop from 100% in PyBullet to 25% in Gazebo under stricter evaluation conditions. To address this limitation, we adopt a sim-to-sim-to-real approach, where actuation effects observed in Gazebo are incorporated into the PyBullet training environment. Using multi-environment DRL with SAC and CrossQ, we learn policies that remain robust despite modeling inaccuracies. This approach can significantly reduce the performance gap across simulators, achieving up to 92% success rate in Gazebo and maintaining 69% under stricter thresholds, with successful transfer to a real robot without additional tuning.

---

### Stochastic Rounding Increases Small Singular Values

Authors: Linkai Ma, Tingzhou Yu, Petros Drineas
Published: 2026-05-29
Categories: math.NA, cs.LG
arXiv: https://arxiv.org/abs/2606.00312v1

Over the past half-dozen years, stochastic rounding (SR) has regained significant attention as a quantization scheme for low-precision floating-point arithmetic, with applications spanning numerical analysis and modern machine learning systems. Recent work has shown that SR acts as an implicit regularizer by increasing the smallest singular value of extremely tall-and-thin (or, symmetrically, short-and-fat) matrices. In this work, we substantially sharpen and extend this understanding in two directions. First, we show that the regularization effect of SR is not restricted to extreme aspect ratio regimes: it persists for matrices with constant aspect ratio. Second, we demonstrate that SR does not merely regularize the smallest singular value, but instead lifts entire clusters of singular values at the tail of the spectrum. Together, these results provide a more general characterization of stochastic rounding as a spectral regularizer, revealing that its effects extend beyond extremal aspect ratios and act on a broader portion of the singular value spectrum.

---

### Where to Refine, When to Stop: Rethinking Redundancy via Latent Discrepancy for Efficient Visual Autoregressive Generation

Authors: Changwang Mei, Peisong Wang, Zekun Li, Changsheng Li, Shuang Qiu, Qinghao Hu, Gang Li, Yifan Zhang, Zhihui Wei, Jian Cheng
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.00310v1

Visual Autoregressive (VAR) models deliver high-quality image generation but suffer from significant inference latency at high resolutions. Recent acceleration approaches most rely on heuristic measures with layer features to prune tokens. Such heuristics are sensitive to complex contextual semantics, leading to inaccurate identification of redundant computation and poor adaptability across prompts. We rethink redundancy in VAR from the perspective of its impact on pixel-space generation and introduce Latent Discrepancy. This unified metric quantifies a token's contribution by measuring the change in model states during generation. Our analysis shows that redundancy is more accurately identified when guided by image latent or pixel-space signals. We further observed that in classifier-free guidance (CFG), the convergence trend of the discrepancy between conditional and unconditional branches exhibits high dynamics with different prompts. Based on these findings, we propose LD-Pruning (Latent Discrepancy Pruning), a training-free framework that removes redundancy via latent discrepancy by integrating decoding-free region selection and adaptive unconditional-branch skipping. Extensive experiments show that LD-Pruning substantially reduces inference latency while maintaining high generation quality, achieving up to 2.35x speedup on Infinity-8B.

---

### Large-scale Uncertainty Quantification for Latent Variable Models Using Subsampling Markov Chain Monte Carlo

Authors: Xiaoyu Wang, Jonathan H. Huggins
Published: 2026-05-29
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2606.00309v1

Stochastic gradient Langevin dynamics combined with Gibbs updates (SGLD--Gibbs) provides a highly scalable approach to approximate Bayesian inference in latent variable models. However, it remains unclear how to tune the algorithm's hyperparameters in a principled manner to ensure the uncertainty estimates are statistically meaningful. In this work, we address this gap in tuning guidance by developing a statistical scaling limit theory for SGLD--Gibbs. We derive a joint asymptotic limit for the global parameters and latent variables under appropriate space-time rescaling. We show that global parameters converge to a diffusion-type limit, while each latent variable converges to a jump process, reflecting the use of intermittent Gibbs updates. This joint jump-diffusion structure reveals how latent-variable randomness contributes to the stationary distribution of the global parameters. We leverage our results to propose explicit guidance on hyperparameter tuning for SGLD--Gibbs that ensures meaningful uncertainty quantification. Numerical experiments show that SGLD--Gibbs with our tuning guidance leads to better parameter estimates, uncertainty quantification, and predictive performance than stochastic variational inference.

---

### How Generation Architecture Shapes Code Complexity in Multi-Agent LLM Systems: A Paired Study on HumanEval

Authors: Nazmus Ashrafi
Published: 2026-05-29
Categories: cs.SE, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2606.00308v1

Large-language-model code generation has shifted from single-shot prompting to multi-agent orchestrations - analyst, coder, tester, and debugger pipelines - and is evaluated almost exclusively on functional correctness. Whether these architectures also affect the structural complexity of the code they produce, and which orchestration layers carry the cost, remains largely unexamined: prior work has documented prompt-level effects on code complexity, but the architecture-level question is open. We compare six widely-used multi-agent configurations (Basic, AC, ACT, Debugger, AC+Debugger, ACT+Debugger) under two models from the GPT-4o family across all 164 HumanEval tasks - 1,968 paired observations - using the five RADON complexity metrics (SLOC, cyclomatic complexity, and Halstead Volume, Difficulty, and Effort). We apply a paired non-parametric statistical pipeline (Friedman omnibus, Wilcoxon signed-rank post-hoc with Holm correction, Kendall's $W$ and matched-pairs rank-biserial effect sizes) in both all-completions and passing-only conditions. The six architectures collapse into two indistinguishable complexity clusters separated by a 50-130% gap, the same partition in both models and under both conditions; among the architectural layers, the analyst-coder split inflates complexity, the runtime debugger does not - and on the analyst-coder background actively deflates it - and the tester re-inflates it. The heavy cluster's additional complexity buys no pass@1 advantage: the leanest architectures match or beat the heaviest on accuracy. Architectural elaboration in LLM code generation should therefore be justified by measured benefit on the dimensions that matter, not assumed.

---

### Rethinking the Role of Temperature in Large Language Model Distillation

Authors: Hoang-Chau Luong, Lingwei Chen
Published: 2026-05-29
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.00306v1

Reverse Kullback-Leibler (RKL) divergence is widely favored over forward KL (FKL) in large language models (LLM) distillation, yet this preference is largely based on comparisons that omit the temperature $τ$, overlooking its central role in softening teacher distributions and improving knowledge transfer. In this work, we revisit temperature in LLM distillation and show that it fundamentally changes the comparison between FKL and RKL. Our analysis reveals an asymmetric effect: temperature substantially enriches FKL with non-dominant token signals, whereas it mainly rescales RKL gradients, causing FKL to benefit much more from $τ$ scaling than RKL. This asymmetry overturns the standard empirical conclusion: although RKL outperforms FKL at $τ=1$, FKL consistently surpasses RKL at higher temperatures across instruction-following benchmarks. Moreover, the impact of temperature is not limited to FKL; it improves a broader family of distillation objectives, enabling simple KL-based methods to achieve competitive performance against recent state-of-the-art LLM distillation approaches.

---

### Bridging Reasoning Trajectories in On-Policy Distillation via Near-Future Guidance

Authors: Yuxuan Jiang, Francis Ferraro
Published: 2026-05-29
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2606.00305v2

On-Policy Distillation (OPD) improves large language model reasoning by training a student model on trajectories sampled from its own policy under teacher supervision. Although OPD operates on trajectories, its learning signal remains token-level: it identifies deviations through high-loss tokens and repairs them through local reverse-KL correction. We show that this "trajectory-sampled but token-learned" mechanism cannot reliably bridge student trajectories toward teacher trajectories. About 30% of high-loss tokens fall into the low-divergence regime, indicating that many are surface-form mismatches rather than real reasoning forks. Moreover, even truly divergent tokens are difficult to repair with isolated token-level supervision, since reasoning failures often unfold as short-horizon distributional drift. We propose Trajectory-aware OPD (TOPD), which uses near-future trajectory information to identify real divergent states and distribute guidance across multiple future tokens. Experiments show that suppressing non-divergent high-loss tokens improves standard OPD from 47.8% to 48.2% average accuracy, while TOPD further improves performance to 52.2%, with gains on AIME24 from 60.0% to 63.3% and AIME25 from 46.7% to 53.3%.

---

### Modeling Spectral Energy Shifts in Spatio-Temporal Graph Anomaly Detection

Authors: Yilin Liu, Hongchao Zhang, Taylor T. Johnson, Ahmad F. Taha, Meiyi Ma
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.00304v2

Graph anomaly detection methods aim to distinguish anomalous nodes. While prior methods characterize anomalies through increased variation in the spectral energy distributions, they overlook those that result in decreased variation, i.e., camouflaged anomalies that appear normal. We show that this type of anomaly persists across multiple datasets and remains undetectable by existing spectral approaches. To address this limitation, we propose a node-level spectral energy formulation that is fully compatible with message passing and enables the detection of camouflaged anomalies. Building on this formulation, we introduce an energy-aware graph learning framework that models spectral shifts through energy-driven message passing in both static and time-series graphs. Besides, our unified architecture extends to temporal settings without introducing specialized sequence modules, enabling efficient learning under long sliding windows. Extensive experiments on large-scale benchmarks demonstrate the effectiveness and scalability of our approach.

---

### Modeling Spectral Energy Shifts in Spatio-Temporal Graph Anomaly Detection

Authors: Yilin Liu, Hongchao Zhang, Taylor T. Johnson, Ahmad F. Taha, Meiyi Ma
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.00304v1

Graph anomaly detection methods aim to distinguish anomalous nodes. While prior methods characterize anomalies through increased variation in the spectral energy distributions, they overlook those that result in decreased variation, i.e., camouflaged anomalies that appear normal. We show that this type of anomaly persists across multiple datasets and remains undetectable by existing spectral approaches. To address this limitation, we propose a node-level spectral energy formulation that is fully compatible with message passing and enables the detection of camouflaged anomalies. Building on this formulation, we introduce an energy-aware graph learning framework that models spectral shifts through energy-driven message passing in both static and time-series graphs. Besides, our unified architecture extends to temporal settings without introducing specialized sequence modules, enabling efficient learning under long sliding windows. Extensive experiments on large-scale benchmarks demonstrate the effectiveness and scalability of our approach.

---

### ERICA: Quantifying Replicability of Cluster Analysis

Authors: Siamak K. Sorooshyari, Manuel A. Rivas, Robert Tibshirani
Published: 2026-05-29
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2606.00302v2

Despite being ubiquitous in science, clustering lacks a unified framework for quantitatively evaluating the replicability of its results. We present evaluating replicability via iterative clustering assignments (ERICA), a method for determining whether clusters can be identified reproducibly in a dataset. The pipeline computes a statistic that determines whether reproducible cluster structure is present in a dataset. Quantitative visualization methods are also introduced to characterize similarities between clusters and identify observations that may represent outliers or unstable assignments. Experiments on synthetic datasets demonstrate that ERICA successfully identifies reproducible cluster structure. In contrast, application of ERICA to three breast cancer gene-expression datasets reveals instances in which clustering solutions are not reproducible. The study underscores the importance of rigorously evaluating clustering solutions and provides a practical framework for doing so.

---