# Research Papers — 2026-05 · bundle page 310 of 326

> Full-text markdown of papers #15451–15500 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/310
[← Previous bundle](https://aipod.dev/papers/archive/2026-05/page/309.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-05/page/311.md)

### Enhanced LLM Reasoning by Optimizing Reward Functions with Search-Driven Reinforcement Learning

Authors: Arash Ahmadi, Sarah Sharif, Yaser, Banad
Published: 2026-05-03
Categories: cs.CL
arXiv: https://arxiv.org/abs/2605.02073v2

Mathematical reasoning is a key benchmark for large language models. Reinforcement learning is a standard post-training mechanism for improving the reasoning capabilities of large language models, yet performance remains sensitive to the design of the reward function that drives policy optimization. This paper introduces a search-driven framework that treats the reward specification itself as an object of optimization. The setting of interest is one in which the base model is held fixed and the reward specification is the primary remaining design lever. Candidate reward functions are generated by a frontier language model, validated automatically, screened through 500-step Group Relative Policy Optimization (GRPO) training runs on a Llama-3.2-3B-Instruct base model with Low-Rank Adaptation (LoRA), and ranked by F1 on the GSM8K test set. Ranked summaries from prior rounds are then fed back into the next round of generation. Over five rounds, the search produces 50 candidate rewards. The mean F1 rises from 0.596 in Round 1 to 0.632 in Round 5, and the top individual reward reaches F1 = 0.787. Seven ensemble configurations of top-ranked rewards are evaluated. The best ensemble achieves F1 = 0.795 (95% bootstrap CI (0.756, 0.832)) and accuracy 0.660 (0.635, 0.686), a 0.19 absolute F1 gain over a base-rewards-only GRPO baseline (F1 = 0.609). Pairwise McNemar tests with Bonferroni correction show all five-or-more-reward configurations are statistically indistinguishable at α = 0.05/21. A three-seed re-training of the best ensemble yields F1 of 0.785. A randomly drawn 5-reward control collapses to F1 = 0.047, which shows that the ranked-feedback loop, not the additive signal of having more rewards, drives the gain.

---

### Weight Clipping for Robust Conformal Inference under Unbounded Covariate Shifts

Authors: James Wang, Surbhi Goel
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.02072v1

Conformal prediction (CP) provides powerful, distribution-free prediction sets, but its guarantees rely on the exchangeability of training and test data, which is often violated in practice due to covariate shifts. While weighted conformal prediction (WCP) is designed to handle such shifts, it can suffer from significant undercoverage when the density ratio between the distributions is unbounded and/or must be learned. This is because of both overfitting in learning the density ratio, and high variance in estimating the nonconformity score threshold. To address this, we introduce clipped least-squares importance fitting (CLISF) as a reduced-variance method for density ratio estimation. Specifically, we show that density ratios learned using CLISF, when plugged into WCP, have bounded expected undercoverage. Furthermore, we show that the undercoverage can be corrected by running WCP with a slightly inflated coverage target; crucially, we are able to estimate the required level of inflation from the data. We provide the first theoretical guarantees for weight clipping in conformal inference, achieving dataset-conditional coverage with a sample complexity that does not blow up with the higher moments of the true density ratio -- a key limitation of prior work. We verify our results on real-world benchmarks and synthetic data.

---

### Pair2Score: Pairwise-to-Absolute Transfer for LLM-Based Essay Scoring

Authors: İbrahim Rıza Hallaç, Hasan Oğul
Published: 2026-05-03
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2605.02069v1

Many scoring applications require absolute predictions, while pairwise comparisons can provide a simpler learning objective. We present Pair2Score, a two-stage learning framework that transfers pairwise comparisons into absolute scoring with parameter-efficient LLaMA adaptation. Stage 1 trains a directional Siamese ranker on pairwise comparisons derived from absolute trait labels; Stage 2 trains an absolute predictor using configurable transfer strategies (warm-start and embedding-fusion variants). We evaluate on rubric-aligned Automated Essay Scoring (AES) traits (grammar, vocabulary, syntax) under a five-fold protocol that co-rotates held-out fold and random seed. At the trait level, the best-performing transfer variant improves quadratic weighted kappa (QWK) over an absolute-only baseline for all three traits. However, not all transfer configurations help: a one-epoch pairwise stage transfers more reliably than extended pairwise training, and transfer configuration -- not just the inclusion of a pairwise stage -- determines whether downstream scoring benefits.

---

### Coopetition-Gym v1: A Formally Grounded Platform for Mixed-Motive Multi-Agent Reinforcement Learning under Strategic Coopetition

Authors: Vik Pant, Eric Yu
Published: 2026-05-03
Categories: cs.MA, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2605.02063v1

We present Coopetition-Gym v1, a benchmark platform for mixed-motive multi-agent reinforcement learning under strategic coopetition. The platform comprises twenty environments organized into four mechanism classes that correspond to four foundational technical reports: interdependence and complementarity (arXiv:2510.18802), trust and reputation dynamics (arXiv:2510.24909), collective action and loyalty (arXiv:2601.16237), and sequential interaction and reciprocity (arXiv:2604.01240). Each environment carries a closed-form payoff structure and a calibrated interdependence matrix derived from the corresponding report. Every environment exposes a parameterized reward layer configurable across three structurally distinct modes (private, integrated, cooperative). This separation of payoff from reward enables reward-type ablation, the platform's principal methodological apparatus. Four of the twenty environments are calibrated against historically documented coopetitive relationships and reproduce their outcomes at 98.3, 81.7, 86.7, and 87.3 percent on the validation rubric (Samsung-Sony LCD, Renault-Nissan Alliance, Apache HTTP Server, Apple iOS App Store). The platform exposes Gymnasium, PettingZoo Parallel, and PettingZoo AEC interfaces and ships 126 reference algorithms: 16 learning algorithms, 7 game-theoretic oracles, 2 heuristic baselines, and 101 constant-action policies. A reference experimental study trained the 16 learning algorithms on every environment under every reward configuration with seven random seeds, producing a 25,708-run training corpus and a 1,116-run behavioral audit corpus, both released under CC-BY-4.0 with Croissant 1.0 metadata. Coopetition-Gym v1 is the first platform to combine continuous-action mixed-motive environments, parameterized reward mutuality, calibrated interdependence coefficients, game-theoretic oracle baselines, and validated case studies.

---

### DR-SNE: Density-Regularized Stochastic Neighbor Embedding

Authors: Maksim Kazanskii
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.02060v2

Dimensionality-reduction methods such as t-SNE preserve local neighborhood structure but can substantially distort the local distribution of data. We introduce Density-Regularized Stochastic Neighbor Embedding (DR-SNE), which augments stochastic neighbor embedding with a regularizer that directly aligns normalized inverse-neighborhood-scale profiles between the original and embedding spaces. The resulting objective preserves relative local concentration while remaining invariant to global rescaling of the embedding. Across real and synthetic datasets, DR-SNE generally improves preservation of the targeted concentration profile while maintaining controlled levels of neighborhood fidelity. Experiments reveal a consistent empirical trade-off between concentration preservation and neighborhood fidelity as the regularization strength varies. These results position DR-SNE as a simple extension of SNE for settings in which relative variation in local concentration is an important property of the representation.

---

### DR-SNE: Density-Regularized Stochastic Neighbor Embedding

Authors: Maksim Kazanskii
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.02060v1

Dimensionality reduction methods such as t-SNE are designed to preserve local neighborhood structure but do not explicitly account for how probability mass is distributed, often leading to distortions of data density. We reformulate dimensionality reduction as the joint alignment of two components: (i) conditional structure, capturing local relationships, and (ii) relative density structure, captured via local density statistics. Based on this perspective, we introduce Density-Regularized SNE (DR-SNE), which augments the stochastic neighbor embedding objective with a density regularization term derived from normalized log-density estimates. Unlike prior approaches such as DensMAP and DenSNE, which rely on local scale consistency, DR-SNE directly aligns normalized density estimates, providing a simple and scale-invariant mechanism for preserving relative density variations. Empirically, DR-SNE improves density preservation while maintaining competitive neighborhood fidelity, and yields gains on density-sensitive tasks such as anomaly detection across multiple datasets. These results suggest that incorporating density information complements geometry-focused objectives in dimensionality reduction.

---

### Observability Conditions and Filter Design for Visual Pose Estimation via Dual Quaternions

Authors: Nicholas B. Andrews, Kristi A. Morgansen
Published: 2026-05-03
Categories: eess.SY, cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2605.02054v1

This paper presents a dual quaternion framework for 6-DOF visual target tracking that addresses key limitations of perspective-n-point (P$n$P) solvers: sensitivity to noise and outliers, and inability to propagate estimates through measurement dropouts. A nonlinear observability analysis is performed using a Lie algebraic approach, deriving sufficient conditions for local observability under two sensing modalities: relative position vector and unit vector measurements. For the unit vector case, the classical collinear feature point degeneracy of the perspective-three-point problem is recovered through rank analysis of the observability codistribution matrix, providing a control-theoretic interpretation of a previously geometric result. A dual quaternion Lie group unscented Kalman filter is then developed, directly modeling relative dynamics without assumptions about cooperative measurements or slowly-varying motion. Simulations demonstrate improved pose estimation accuracy and robustness to occlusions compared to an off-the-shelf P$n$P solver. Results are broadly applicable to visual-inertial navigation, simultaneous localization and mapping, and P$n$P solver development.

---

### Methods, Data, and Conceptual Change: Reflections from Two Quantitative Diachronic Case Studies

Authors: Catherine Wong, Bach Phan-Tat, Susan Fitzmaurice
Published: 2026-05-03
Categories: cs.CL
arXiv: https://arxiv.org/abs/2605.02052v1

This discussion paper reflects on how quantitative approaches to historical linguistics interact with dataset properties. Drawing on two worked examples, we examine English data using quad-based concept modelling of Early Modern English discourse in EEBO-TCP (c. 1470s-1690s; 765M words) alongside SynFlow analysis of scientific writing in Royal Society Corpus 6.0.4 (1750-1799; drawn from a 78.6M-token open corpus). Through parallel comparison, the paper explores how each approach operationalises concepts, the data assumptions they entail, and the diachronic interpretations they support. We argue that comparative methodological reflection clarifies the limits of purely lexical, frequency-based approaches and highlights how dataset structure shapes the kinds of semantic change that quantitative methods can reliably detect.

---

### Principles and Guidelines for Randomized Controlled Trials in AI Evaluation

Authors: Christopher Kelly, Angelica Chowdhury, Alexandra Campili, Bimpe Ayoola, Devin Barbour, Thomas Chen Dawson, Ze Shen Chin, Rokas Gipiškis
Published: 2026-05-03
Categories: cs.CY, cs.AI, cs.HC, cs.LG
arXiv: https://arxiv.org/abs/2605.02050v2

This work establishes a framework for standardizing AI evaluation RCTs (sometimes called human uplift studies). Drawing on established practices from disciplines with established RCT traditions, including software engineering, economics, clinical and health sciences, and psychology, we synthesize five principles drawn from established validity frameworks and open-science standards on transparency, repeatability, and verification, which together serve as the conceptual foundation for 33 actionable guidelines adapted for AI evaluation RCT contexts, expressed as requirements with rationales, implementation instructions, and evidence bases. We position the principles and guidelines as serving three key roles for AI evaluation RCTs: a design tool for planning studies, an evaluation rubric for assessing existing work, and a blueprint for standard setting as the field converges on norms. AI evaluation research currently lacks common standards and shared vocabulary for producing cumulative, comparable, policy-ready evidence. This framework is a contribution toward that foundation, providing evaluative criteria and a shared conceptual language alongside actionable guidelines.

---

### Optimization of CV-QKD Under Practical Constraints

Authors: Svitlana Matsenko, Amirhossein Ghazisaeidi, Marcin Jarzyna, Konrad Banaszek, Darko Zibar
Published: 2026-05-03
Categories: cs.IT, cs.AI, quant-ph
arXiv: https://arxiv.org/abs/2605.02045v1

Using reinforcement learning, we optimize for practical hardware constraints, including limited FIR filter taps at the transmitter and receiver, mean photon number and finite DAC/ADC resolution. Under these realistic conditions, the proposed approach achieves significant performance improvements.

---

### NeuroViz: Real-time Interactive Visualization of Forward and Backward Passes in Neural Network Training

Authors: Tanvi Sharma, Reza Rawassizadeh
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.02044v2

Training neural networks is difficult to interpret, particularly for newcomers. We introduce NeuroViz, an interactive visualization tool that supports real-time exploration of fully connected neural network training. Users can configure network architecture, activation functions, learning rates, and datasets, then observe activations, weight updates, and loss progression. NeuroViz visualizes weight changes in direct correspondence with activation signals in both forward and backward passes, enabling users to distinguish pre- and post-update states within individual epochs and view dynamically updating per-neuron equations. We conduct a comparative user study with 31 participants against six established visualization tools and we achieved the highest usability score (SUS 80.97, in the 'excellent' range), with mean rankings of 2.47 for clarity and 2.23 for usefulness (lower is better). Over 70% of participants reported that the visualizations substantially increased their perception of neural network training transparency. The implemented instance is accessible at https://neuroviz.org.

---

### Bringing Order to Asynchronous SGD: Towards Optimality under Data-Dependent Delays with Momentum

Authors: Tehila Dahan, Roie Reshef, Sharon Goldstein, Kfir Y. Levy
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.02043v2

Asynchronous stochastic gradient descent (SGD) enables scalable distributed training but suffers from gradient staleness. Existing mitigation strategies, such as delay-adaptive learning rates and staleness-aware filtering, typically attenuate or discard delayed gradients, introducing systematic bias: updates from simpler or faster-to-process samples are overrepresented, while gradients from more complex samples are delayed or suppressed. In contrast, prior approaches to data-dependent delays rely on a Lipschitz assumption that yields suboptimal rates or leave the smooth, convex case unaddressed. We propose a momentum-based asynchronous framework designed to preserve information from delayed gradients while mitigating the effects of staleness. We establish the first optimal convergence rates for data-dependent delays in both convex and non-convex smooth setups, providing a new result for asynchronous optimization under standard assumptions. Additionally, we derive robust learning-rate schedules that simplify hyperparameter tuning in practice.

---

### TIJERE: A Novel Threat Intelligence Joint Extraction Model Based on Analyst Expert Knowledge

Authors: Inoussa Mouiche, Sherif Saad
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.02041v1

The extraction of entities and relationships from threat intelligence reports into structured formats, such as cybersecurity knowledge graphs, is essential for automated threat analysis, detection, and mitigation. However, existing joint extraction methods struggle with feature confusion, language ambiguity, noise propagation, and overlapping relations, resulting in low accuracy and poor model performance. This paper presents TIJERE, an innovative joint entity and relation extraction framework that formulates joint extraction as a multisequence labeling representation (MSLR) problem. Specifically, separate sequences are generated for each entity pair. Unlike prior tagging schemes, MSLR integrates expert domain features to enrich positional, contextual, and semantic representations of entities, thereby enhancing feature distinction and classification accuracy. Additionally, TIJERE reduces language ambiguity and enhances domain-specific generalization by leveraging SecureBERT+, a contextual language model fine-tuned on cybersecurity text. This improves both named entity recognition (NER) and relation extraction (RE). This paper also introduces DNRTI-JE, the first publicly available jointly labeled dataset for cybersecurity entity and RE, filling a crucial gap in cyber threat intelligence automation. Empirical evaluations on the curated DNRTI-JE dataset demonstrate that TIJERE achieves state-of-the-art performance, with F1-scores exceeding 0.93 for NER and 0.98 for RE, outperforming existing methods. Together, TIJERE and the standardized benchmarking DNRTI-JE dataset enable high-performance cybersecurity intelligence extraction, with transferable applications in healthcare, finance, and bioinformatics.

---

### What Single-Prompt Accuracy Misses: A Multi-Variant Reliability Audit of Language Models

Authors: Ranit Karmakar, Jayita Chatterjee
Published: 2026-05-03
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2605.02038v1

Single-prompt accuracy is the dominant way to benchmark language models, but it can miss reliability failures that matter. We evaluate a 15-model open-weight corpus, with the main reliability analyses focused on 10 instruct models across five classification and reasoning benchmarks under five prompt variants each, measuring accuracy, token-probability calibration, verbal-confidence calibration, verbal parse rate, and prompt-perturbation spread for every (model x dataset x variant) cell. We find three broad results. First, evaluation design can materially change the conclusion. Switching Expected Calibration Error (ECE) token from a raw to a label-set-normalised definition changes per-cell calibration by a mean absolute 0.149. More strikingly, pairing a chain-of-thought prompt with a first-character evaluator on ARC-Challenge reduces apparent accuracy by 72-88% across all five primary models; two independent repair procedures recover 93.8% and 102.7% of the lost performance, indicating an evaluator-side rather than model-side failure. Second, confidence signals are fragile. On MMLU-Pro, every primary model verbally reports confidence substantially above both its accuracy and its token-probability confidence on the same rows, and verbal parse rate can collapse for a single model on a single prompt variant. Third, prompt robustness does not track parameter count reliably. Across 10 instruct models, the correlation between model size and prompt-perturbation spread ranges from -0.244 to 0.474 across benchmarks. Taken together, these results show that reliability conclusions for small language models depend not only on the model being evaluated, but also on the evaluation pipeline used to measure it. We argue that calibration definitions, evaluator logic, verbal parseability, and prompt robustness should be reported explicitly when making reliability claims.

---

### VILAS: A VLA-Integrated Low-cost Architecture with Soft Grasping for Robotic Manipulation

Authors: Zijian An, Hadi Khezam, Bill Cai, Ran Yang, Shijie Geng, Yiming Feng, Yue Zheng, Lifeng Zhou
Published: 2026-05-03
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2605.02037v2

We present VILAS, a fully low-cost, modular robotic manipulation platform designed to support end-to-end vision-language-action (VLA) policy learning and deployment on accessible hardware. The system integrates a Fairino FR5 collaborative arm, a Jodell RG52-50 electric gripper, and a dual-camera perception module, unified through a ZMQ-based communication architecture that seamlessly coordinates teleoperation, data collection, and policy deployment within a single framework. To enable safe manipulation of fragile objects without relying on explicit force sensing, we design a kirigami-based soft compliant gripper extension that induces predictable deformation under compressive loading, providing gentle and repeatable contact with delicate targets. We deploy and evaluate three state-of-the-art VLA models on the VILAS platform: pi_0, pi_0.5, and GR00T N1.6. All models are fine-tuned from publicly released pretrained checkpoints using an identical demonstration dataset collected via our teleoperation pipeline. Experiments on a grape grasping task validate the effectiveness of the proposed system, confirming that capable manipulation policies can be successfully trained and deployed on low-cost modular hardware. Our results further provide practical insights into the deployment characteristics of current VLA models in real-world settings.

---

### VIDA: A dataset for Visually Dependent Ambiguity in Multimodal Machine Translation

Authors: Jingheng Pan, Xintong Wang, Longyue Wang, Liang Ding, Weihua Luo, Chris Biemann
Published: 2026-05-03
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2605.02035v2

Ambiguity resolution is a key challenge in multimodal machine translation (MMT), where models must genuinely leverage visual input to map an ambiguous expression to its intended meaning. Although prior work has proposed disambiguation-oriented benchmarks probing the role of vision, we observe that existing benchmarks remain limited by task-format mismatch, narrow ambiguity coverage, or insufficient visual-dependency validation. Moreover, existing ambiguity evaluations are not well suited to diverse ambiguity types in open-ended translation. To address these limitations, we present VIDA (Visually-Dependent Ambiguity), a dataset of 2,500 carefully curated instances in which resolving an annotated source span requires visual evidence. We further propose Disambiguation-Centric Metrics that use an LLM-as-a-judge classifier to verify whether annotated ambiguous expressions are resolved correctly at the span level. Experiments with two state-of-the-art LVLMs show that supervised fine-tuning (SFT) improves overall translation quality, while chain-of-thought SFT (CoT-SFT) yields stronger out-of-distribution disambiguation, suggesting that explicit disambiguation guidance improves generalization to diverse ambiguity types.

---

### Conventional Commit Classification using Large Language Models and Prompt Engineering

Authors: H. M. Sazzad Quadir, Sakib Al Hasan, Md. Nurul Ahad Tawhid
Published: 2026-05-03
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2605.02033v1

Conventional commits provide a structured format for writing commit messages, which improves readability, software maintenance, and enables automation tools such as changelog generators and semantic versioning systems. Existing approaches to conventional commit classification typically rely on ML/DL models trained on large labeled datasets. In this paper, we investigated a training-free alternative by leveraging large language models (LLMs) through prompt engineering. Rather than building a task-specific classifier, we evaluate three prompting strategies, such as zero-shot, few-shot, and chain-of-thought, across three open-source LLMs of varying scale: Mistral-7B-Instruct, LLaMA-3-8B, and DeepSeek-R1-32B. Classification is performed directly on code diffs extracted from a balanced dataset of 3,200 commits mined from the InfluxDB repository, without any model fine-tuning. Our results show that few-shot prompting consistently achieves the highest accuracy, while chain-of-thought prompting does not yield additional gains for this classification task. Among the evaluated models, DeepSeek-R1-32B achieves the strongest overall performance, suggesting that model scale plays a meaningful role in conventional commit classification. These findings provide practical guidance for researchers and practitioners seeking to automate commit classification without the overhead of curating and maintaining labeled training data.

---

### Language models fail at extended rule following

Authors: Tianxiang Dai, Jonathan Fan
Published: 2026-05-03
Categories: cs.CL
arXiv: https://arxiv.org/abs/2605.02028v2

Large language models are highly capable of answering difficult questions by retrieving, recombining, and attending to information in long contexts. For agentic tasks, an additional capability is required: the preservation of an exact state while repeatedly applying rules. We find that this reliability is absent across language models. To demonstrate, we query 126 leading model variants with the task of counting a long string of repeated characters, and we find they all cannot accurately count above a model-dependent, syntax-sensitive counting capacity threshold. Failures are abrupt and persist even with increasing model size, inference time computation, and external tool. Mechanistic probing indicates that models use a finite number of internal states to mimic counting as a rule and fail once these states are exhausted. Furthermore, such states are the basis for performing complex tasks beyond counting. These results indicate that fundamentally new model architectures are required for autonomous agents to achieve truly reliable rule following capabilities.

---

### Large margin classifier with graph-based adaptive regularization

Authors: Vítor M. Hanriot, Turíbio T. Salis, Luiz C. B. Torres, Frederico Coelho, Antonio P. Braga
Published: 2026-05-03
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2605.02027v1

This paper introduces the use of per-class regularization hyperparameters in Gabriel graph-based binary classifiers. We demonstrate how the quality index used for regularization behaves both in the margin region and in the presence of outliers, and how incorporating this regularization flexibility can lead to solutions that effectively eliminate outliers while training the classifier. We also show how it can address class imbalance by generating higher and lower thresholds for the majority and minority classes, respectively. Thus, rather than having a single solution based on fixed thresholds, flexible thresholds expand the solution space and can be optimized through hyperparameter tuning algorithms. Friedman test shows that flexible thresholds are capable of improving Gabriel graph-based classifiers.

---

### Towards Systematic Generalization for Power Grid Optimization Problems

Authors: Zeeshan Memon, Yijiang Li, Hongwei Jin, Kibaek Kim, Liang Zhao
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.02026v1

AC Optimal Power Flow (ACOPF) and Security-Constrained Unit Commitment (SCUC) are fundamental optimization problems in power system operations. ACOPF serves as the physical backbone of grid simulation and real-time operation, enforcing nonlinear power flow feasibility and network limits, while SCUC represents a core market-level decision process that schedules generation under operational and security constraints. Although these problems share the same underlying transmission network and physical laws, they differ in decision variables and temporal coupling, and prior learning-based approaches address them in isolation, resulting in disjoint models and representations.We propose a learning framework that jointly models ACOPF and SCUC through a shared graph-based backbone that captures grid topology and physical interactions, coupled with task-specific decoders for static and temporal decision-making. Training includes solver supervision with physics-informed objectives to enforce AC feasibility and inter-temporal operational constraints. To evaluate generalization, we assess cross-case transfer on unseen grid topologies for ACOPF and SCUC without retraining, and systematic generalization on the UC-ACOPF problem using unsupervised, physics-based objectives and a power-dispatch consensus mechanism. Experiments across multiple grid scales demonstrate improved performance and transferability relative to existing learning-based baselines, indicating that the model can support learning across heterogeneous power system optimization problems.

---

### Tenability and Weak Semantics: Modeling Non-uniform Defense -- Extended Version

Authors: Uri Andrews, Luca San Mauro, John Spoerl
Published: 2026-05-03
Categories: cs.AI
arXiv: https://arxiv.org/abs/2605.02024v1

In Dung-style abstract argumentation, various semantics capture notions of acceptability of arguments. The admissibility semantics capture the notion that an argument can be consistently defended from any potential counterargument. Weak semantics often relax the demands of admissibility by restricting which counterarguments must be taken seriously (e.g., discounting self-defeating or otherwise incoherent attacks). Many prominent proposals for weak semantics remain extension-based in a stronger sense. While these semantics discount attacks from arguments which are considered unreasonable, they still require a uniform defense against all reasonable arguments, even if they are collectively inconsistent. This uniformity can be too demanding when defensibility is inherently strategic, and thus the appropriate reply depends on the opponent's line of attack. We introduce tenability, a family of dialogue-based semantics that formalize when a designated argument (or a set of arguments) can be maintained in debate by a proponent against any conflict-free attack which the opponent may present. The approach is motivated by three natural benchmark patterns: self-defeating attack, floating assignment, and disjunctive reinstatement, on which tenability behaves differently from all weak semantics previously considered in the literature. We define three variants -- static tenability, tenability, and strong tenability -- via monotone commitment games over finite conflict-free moves, differing in the obligations imposed on the disputants. We establish the relative strength of these notions, prove implications and separations with previously studied weak semantics, and we analyze computational complexity on finite frameworks: deciding static tenability is $Π^P_2$-complete, while deciding tenability and strong tenability is PSPACE-complete.

---

### Robust and Explainable Divide-and-Conquer Learning for Intrusion Detection

Authors: Yan Zhou, Kevin Hamlen, Michael De Lucia, Murat Kantarcioglu, Latifur Khan, Sharad Mehrotra, Ananthram Swami, Bhavani Thuraisingham
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.02015v1

Machine learning-based intrusion detection requires complex models to capture patterns in high-dimensional, noisy, and class-imbalanced raw network traffic, yet deploying such models remains impractical on resource-constrained devices with limited processing power and memory. In this paper, we present a correlation-aware divide-and-conquer learning technique that decomposes a complex learning problem into smaller, more manageable subproblems. This enables lightweight models as simple as decision trees to be trained on focused subtasks, yielding up to 43.3% higher local accuracy and up to 257 times reduction in model size on real-world network intrusion detection datasets, while also improving adversarial robustness and explainability.

---

### MIRA: A Score for Conditional Distribution Accuracy and Model Comparison

Authors: Sammy Sharief, Justine Zeghal, Gabriel Missael Barco, Pablo Lemos, Yashar Hezaveh, Laurence Perreault-Levasseur
Published: 2026-05-03
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2605.02014v1

We introduce Mira, a sample-based score for assessing the accuracy of a candidate conditional distribution using only joint samples from the true data-generating process. Relying on the principle that distributions coincide if they assign equal probability mass to all regions, we derive an analytic expression for the Mira statistic, whose average defines the Mira score. This formulation further allows us to compute theoretical reference values and uncertainty estimates when the candidate distribution matches the true one. This framework enables model comparison by quantifying the alignment between the conditional distribution of a candidate model and the true data generating process. Consequently, Mira enables Bayesian model comparison through direct posterior validation, bypassing the challenging evidence computation. We demonstrate its effectiveness across several toy problems and Bayesian inference tasks.

---

### Enhancing Judgment Document Generation via Agentic Legal Information Collection and Rubric-Guided Optimization

Authors: Weihang Su, Xuanyi Chen, Yueyue Wu, Qingyao Ai, Yiqun Liu
Published: 2026-05-03
Categories: cs.CL, cs.AI, cs.IR
arXiv: https://arxiv.org/abs/2605.02011v1

Automating the drafting of judgment documents is pivotal to judicial efficiency, yet it remains challenging due to the dual requirements of comprehensive retrieval of legal information and rigorous logical reasoning. Existing approaches, typically relying on standard Retrieval-Augmented Generation and Supervised Fine-Tuning, often suffer from insufficient evidence recall, hallucinated statutory references, and logically flawed legal reasoning. To bridge this gap, we propose Judge-R1, a unified framework designed to enhance LLM-based judgment document generation by jointly improving legal information collection and judgment document generation. First, we introduce Agentic Legal Information Collection, which employs a dynamic planning agent to retrieve precise statutes and precedents from multiple sources. Second, we implement Rubric-Guided Optimization, a reinforcement learning phase utilizing Group Relative Policy Optimization (GRPO) with a comprehensive legal reward function to enforce adherence to judicial standards and reasoning logic. Extensive experiments on the JuDGE benchmark demonstrate that Judge-R1 significantly outperforms state-of-the-art baselines in both legal accuracy and generation quality.

---

### Reliable AI Needs to Externalize Implicit Knowledge: A Human-AI Collaboration Perspective

Authors: Hengyu Liu, Tianyi Li, Zhihong Cui, Yushuai Li, Zhangkai Wu, Torben Bach Pedersen, Kristian Torp, Christian S. Jensen
Published: 2026-05-03
Categories: cs.AI
arXiv: https://arxiv.org/abs/2605.02010v2

This position paper argues that reliable AI requires infrastructure for human validation of implicit knowledge. AI learns from both explicit knowledge (papers, documentation, structured databases) and implicit knowledge (reasoning patterns, debugging processes, intermediate steps). Implicit knowledge remains unexternalized because documentation cost exceeds perceived value -- yet AI learns from it indiscriminately, acquiring both beneficial patterns and harmful biases. Current reliability methods can only verify explicit knowledge against sources, creating a fundamental gap: the most valuable AI capabilities (reasoning, judgment, intuition) are precisely those we cannot verify. We propose Knowledge Objects (KOs) -- structured artifacts that externalize implicit knowledge into forms humans can inspect, verify, and endorse. KOs transform verification economics: what was previously too costly to verify becomes feasible, enabling accumulated human validation to improve reliability over time.

---

### Benchmarking Wireless Representations: High-Dimensional vs. Compressed Embeddings for Efficiency and Robustness

Authors: Murilo Batista, Shirin Salehi, Saeed Mashdour, Paul Zheng, Rodrigo C. de Lamare, Anke Schmeink
Published: 2026-05-03
Categories: eess.SP, cs.IT, cs.LG
arXiv: https://arxiv.org/abs/2605.02009v1

Building on recent advances in representation learning for wireless channels, this work investigates the cost-benefit trade-offs of high-dimensional channel embeddings in practical systems. We benchmark multiple wireless representations: high-dimensional learned embeddings from a wireless foundation model, compact autoencoder-based representations with significantly lower dimensionality, and raw data baselines, evaluating their performance across diverse downstream tasks. We then systematically analyze data efficiency, noise robustness, and computational complexity, explicitly characterizing the resource overhead associated with high-dimensional embeddings. Beyond standard tasks such as line-of-sight/non-line-of-sight (LoS/NLoS) classification and beam selection, we introduce power allocation as a new downstream task. Our results reveal clear trade-offs: while high-dimensional embeddings can perform well in few-shot regimes for certain tasks, they incur substantial latency and parameter overhead. In contrast, compressed latent representations learned by autoencoders demonstrate improved noise robustness and more stable performance across tasks, while significantly reducing computational and transmission costs.

---

### How Can One Choose the Best CAM-Based Explainability Method for a CNN Model?

Authors: Daniel da Silva Costa, Pedro Nuno de Souza Moura, Adriana C. F. Alvim
Published: 2026-05-03
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2605.02007v1

In recent years, several advances have been observed in Deep Learning with surprising results. Models in this area have been increasingly used in numerous applications, including those sensitive to human life, which require clear explanations and justifications. Various explainability methods have been proposed, but not many metrics to evaluate these methods. The most commonly used metric is the Intersection over Union (IoU). However, due to the characteristics of the results of the explainability methods, called saliency maps, which do not have a known shape, we hypothesise that there must be a better metric that allows one to find an explainability method that produces results that best resemble the human perception. We propose using different metrics to assess the similarity between human perception and the explanation saliency maps to find a better metric. An investigation was conducted employing a subset of the Chihuahuas images from ImageNet dataset. Several CAM-based explainability methods were used to generate saliency maps for each chihuahua image. Alignment was measured by applying distance metrics between the bounding box of human annotations and the saliency maps produced by each explainability method. Rankings of the best saliency maps were created using the results of the distance metrics and compared to the ranking obtained using people's choice, collected through crowdsourcing, of the best explanation saliency maps for each selected image. Comparison between rankings was performed using the Rank-Biased Overlap (RBO) metric. The results indicate the feasibility of our method to find the explainability method that best resembles human perception. In our experiments, the two metrics that best resemble human perception corresponded to Manhattan and Correlation. Besides, the best explainability methods regarding human perception were LayerCAM, Score-CAM, and IS-CAM.

---

### Personalized Digital Health Modeling with Adaptive Support Users

Authors: Zhongqi Yang, Mahkameh Rasouli, Neda Mohseni, Yong Huang, Iman Azimi, Amir M. Rahmani
Published: 2026-05-03
Categories: cs.AI
arXiv: https://arxiv.org/abs/2605.02004v2

Personalized models are essential in digital health because individuals exhibit substantial physiological and behavioral heterogeneity. Yet personalization is limited by scarce and noisy user-specific data. Most existing methods rely on population pretraining or data from similar users only, which can lead to biased transfer and weak generalization. We propose a unified personalization framework that trains a personal model using adaptively weighted support users, including both similar and dissimilar individuals. The objective integrates personal loss, similarity-weighted transfer from similar users, and contrastive regularization from dissimilar users to suppress misleading correlations. An iterative optimization algorithm jointly updates model parameters and user similarity weights. Experiments on six tasks across four real-world digital health datasets show consistent improvements over population and personalized baselines. The method achieves up to 10% lower RMSE on large-scale datasets and approximately 25% lower RMSE in low-data settings. The learned adaptive weights improve data efficiency and provide interpretable guidance for targeted data selection.

---

### RamanBench: A Large-Scale Benchmark for Machine Learning on Raman Spectroscopy

Authors: Mario Koddenbrock, Christoph Lange, Robin Legner, Martin Jäger, Martin Kögler, Mariano N. Cruz Bournazou, Peter Neubauer, Felix Biessmann, Erik Rodner
Published: 2026-05-03
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2605.02003v2

Machine Learning (ML) has transformed many scientific fields, yet key applications still lack standardized benchmarks. Raman spectroscopy, a widely used technique for non-invasive molecular analysis, is one such field where progress is limited by fragmented datasets, inconsistent evaluation, and models that fail to capture the structure of spectral data. We introduce RamanBench, the first large-scale, fully reproducible benchmark for ML on Raman spectroscopy, consisting of streamlined data access, evaluation protocols and code, as well as a live leaderboard. It unifies 74 datasets (including 16 first released with this benchmark) across four domains, comprising 325,668 spectra and spanning classification and regression tasks under diverse experimental conditions. We benchmark 28 models under a standardized protocol, including classical methods (e.g., PLS), Raman-specific (e.g., RamanNet), Tabular Foundation Model (TFM) (e.g., TabPFN), and time-series approaches (e.g., ROCKET). TFM consistently outperform domain-specific and gradient boosting baselines, while time-series models remain competitive. However, no method generalizes across datasets, revealing a fundamental gap. Therefore, we invite the community to contribute new approaches to our living benchmark, with the potential to accelerate advances in critical applications such as medical diagnostics, biological research, and materials science.

---

### TumorXAI: Self-Supervised Deep Learning Framework for Explainable Brain MRI Tumor Classification

Authors: Abrar Hossain Zahin, Amit Kumar Saha, Tanvir Mridha, Saifur Rahman, Jannatul Ferdous Prome, Raima Husna, Israt Jahan, Ahmed Wasif Reza
Published: 2026-05-03
Categories: cs.AI
arXiv: https://arxiv.org/abs/2605.01999v1

Classifying brain tumors using magnetic resonance imaging (MRI) is crucial for early diagnosis and treatment; however, tumor heterogeneity and a dearth of annotated datasets restrict the use of supervised deep learning approaches. In this work, we use self-supervised learning (SSL) to study multi-class brain tumor classification. Using a ResNet-50 backbone, we evaluate four SSL frameworks including SimCLR, BYOL, DINO, and Moco v3 on a publicly available dataset of 4,448 MRIs with 17 distinct tumor types. On the dataset, SimCLR achieved 99.64% accuracy, 99.64% precision, 99.64% recall, and 99.64% F1-score. The workflow includes preprocessing, fine-tuning, linear evaluation, and SSL pretraining with data augmentations. Results show that, when labels are limited, SSL-pretrained models outperform supervised baselines in terms of F1-score, recall, accuracy, and precision. Additionally, by providing visual insights into model decisions, Explainable AI techniques (Grad-CAM, Grad-CAM++, EigenCAM) enhance interpretability. These results demonstrate SSL's scalability and dependability in diagnosing brain tumors from unlabeled medical data.

---

### From Concept to Capability: Evaluating 3D Gaussian Splatting for Synthetic Scene Editing in Autonomous Driving

Authors: Ali Nouri, Yifei Zhang, Yifan Zhang, Tayssir Bouraffa, Zhennan Fei, Zijian Han, Håkan Sivencrona, Anders Heyden
Published: 2026-05-03
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.01995v1

The perception of an Autonomous Driving System (ADS) critically depends on relevant, comprehensive, and diverse datasets to ensure its safety while operating in the environment. Field data collection lacks completeness with respect to the list of rare but still possible safety-related scenarios needed for the development, verification, and validation of the ADS. 3D Gaussian Splatting (3DGS) has shown promising capabilities for the reconstruction and editing of scenes based on data collected by cameras and LiDAR sensors. However, the industrial fidelity evaluation of reconstructions is underexplored, which is crucial when employing such methods in safety-related systems, especially for ADS. This becomes more challenging as ADS operates in a dynamic, uncontrolled environment with limited viewpoints and often partially occluded objects. This paper addresses this gap by proposing and implementing a framework (Fig. 1) to systematically analyze the capabilities and limitations of 3DGS for use in the reconstruction of safety-related scenes. It focuses on the quality of reconstruction for vehicles and pedestrians, which are the two most critical object classes for ADS. Our findings provide industry insights into the fidelity degradation of reconstructions from multiple novel viewpoints, both lateral and longitudinal, enabling the integration of these methods into real-world industrial AD software development and testing pipelines.

---

### Real-Time Text Transmission via LLM-Based Entropy Coding over Fixed-Rate Channels

Authors: Vishnu Teja Kunde, Jean-Francois Chamberland, Krishna R. Narayanan, Jamison Ebert
Published: 2026-05-03
Categories: cs.IT, cs.LG
arXiv: https://arxiv.org/abs/2605.01991v1

Learning, prediction, and compression are intimately connected: a model that accurately predicts the next symbol in a sequence can be coupled with a source coder to compress that sequence near its information-theoretic limit. When tokenized characters arriving at a fixed reading pace are encoded into variable-length codewords and streamed over a fixed-rate channel, a queue forms whose per-token delay depends on the mean and variance of the bit lengths and on the coder's algorithmic latency. This paper investigates the compression--delay tradeoff that arises when a causal language model serves as the sequential predictor within a predict-then-code architecture for real-time text transmission. Several coding schemes are compared: Shannon (ideal), Huffman, arithmetic coding, rANS at various block sizes, and gzip. The analysis separates algorithmic delay, inherent to the coder, from computational delay, which shrinks as hardware improves. Huffman is the practical choice for over-provisioned channels, with zero algorithmic delay and modest compression overhead. Arithmetic coding achieves near-optimal compression at the cost of decodability delay. Findings are validated across two scales: GPT-2 (124M) and Llama~3.2 (3B), a twenty-five-fold parameter range. This scaling yields an approximately 38\% reduction in bits per character, effectively over-provisioning the channel and thereby changing which coder is optimal.

---

### DBLP: Phase-Aware Bounded-Loss Transport for Burst-Resilient Distributed ML Training

Authors: Zechen Ma, Zixi Qu, Jinyan Yi, David Lin, Yashar Ganjali
Published: 2026-05-03
Categories: cs.LG, cs.NI
arXiv: https://arxiv.org/abs/2605.01989v2

Distributed machine learning (ML) training has become a necessity with the prevalence of billion to trillion-parameter-scale models. While prior work has improved training efficiency from the ML perspective at the application layer, it often fails to address transient congestion events at the network layer that introduce severe tail latency and training-time variability, thereby undermining the quality of service of distributed ML training systems. Existing network optimizations treat all gradients equally and thus fail to integrate sufficient model-training insights into communication protocol design. In this paper, we present Dynamic Bounded-Loss Protocol (DBLP), a burst-resilient, training-phase-aware, and hardware-agnostic transport protocol that incorporates model-level tolerance properties into gradient communication. By dynamically adjusting gradient loss tolerance across training phases, DBLP reduces overall training time and mitigates tail-latency collapse during transient high-loss events (i.e., microbursts). Compared to the current state-of-the-art solution (baseline), DBLP tolerates significantly higher loss while achieving comparable test accuracy, and reduces end-to-end training time by an average of 24.4% and a maximum of 33.9%. At microburst events, DBLP achieves up to 5.88x single-round communication latency speedups over the baseline, preventing burst-induced tail-latency spikes and maintaining stable training performance.

---

### Misclassification Rate and Privacy-Utility Trade-offs in Graph Convolutional Networks via Subsampling Stability

Authors: Yexin Zhang, Zhongtian Ma, Qiaosheng Zhang, Zhen Wang
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.01987v1

We study differential privacy (DP) in Graph Convolutional Networks (GCNs) through the framework of \textit{subsampling stability}. We derive upper bounds on the misclassification rate that depend explicitly on the subsampling probability $p_s$. Furthermore, we characterize the \textit{privacy--utility trade-off} by identifying feasible ranges of $p_s$; if $p_s$ is too large, the stability-based privacy condition becomes difficult to satisfy, yielding vacuous guarantees, whereas if it is too small, accuracy deteriorates. Our results provide the first rigorous theoretical framework for understanding subsampling stability in GCNs under DP.

---

### 12 Angry AI Agents: Evaluating Multi-Agent LLM Decision-Making Through Cinematic Jury Deliberation

Authors: Ahmet Bahaddin Ersoz
Published: 2026-05-03
Categories: cs.AI
arXiv: https://arxiv.org/abs/2605.01986v1

What if the twelve jurors of Sidney Lumet's 12 Angry Men (1957) were not men, but large language models? Would the one juror who disagrees still be able to change everyone's mind? This paper instantiates that scenario as a multi-agent benchmark for LLM deliberation: twelve agents, each conditioned on a film-faithful persona, debate the film's murder case using multi-agent framework. Two models representing opposite ends of the RLHF spectrum are tested: GPT-4o (closed-source, heavy alignment) and Llama-4-Scout (open-weight, lighter alignment), across three conditions (baseline, open-minded prompt, no initial vote), with N = 3 replications per cell (18 runs total). Three findings emerge. (i) Seventeen of eighteen runs end in a hung jury (a state where the jury fails to reach a unanimous verdict); the film's central event, gradual minority-to-majority persuasion, almost never occurs, indicating that anchoring is the dominant failure mode of current LLMs in this setting. (ii) The two models exhibit sharply different internal dynamics: GPT-4o produces a mean of 1.0 vote changes per run across all conditions, while Llama-4-Scout ranges from 2.0 (baseline) to 6.0 (open-minded prompt), and is the only model to reach a NOT\_GUILTY verdict (1 of 3 runs in the no-initial-vote condition). The same ``open-minded'' instruction is internalized by Llama and ignored by GPT-4o. (iii) This asymmetry suggests that the intensity of RLHF alignment training, not model capability, is the primary determinant of deliberative flexibility in multi-agent settings. Flexibility, not capability, tracks human deliberation. The work is framed as an exploratory study and discusses implications for jury-of-LLMs evaluation and multi-agent debate.

---

### Learn-To-Learn on Arbitrary Textual Conditioning: A Hypernetwork-Driven Meta-Gated LLM

Authors: Luo Ji, Qi Qin, Ningyuan Xi, Teng Chen, Qingqing Gu, Hongyan Li
Published: 2026-05-03
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2605.01973v3

Conventional LLMs may suffer from corpus heterogeneity and subtle condition changes. While finetuning can create the catastrophe forgetting issue, application of meta-learning on LLMs is also limited due to its complexity and scalability. In this paper, we activate the meta-signal of $β$ within the SwiGLU blocks, resulting in a meta-gating mechanism that adaptively adjusts the nonlinearity of FFN. A hypernetwork is employed which dynamically produces $β$ on textual conditions, providing meta-controllability on LLMs. By testing on different condition types such as task, domain, persona, and style, our method outperforms finetuning and meta-learning baselines, and can generalize reasonably on unseen tasks, condition types, or instructions. Our code can be found in https://github.com/AaronJi/MeGan.

---

### ProtoFair: Fair Self-Supervised Contrastive Learning via Pseudo-Counterfactual Pairs

Authors: Marah Halawa, Olaf Hellwich
Published: 2026-05-03
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.01971v2

Self-supervised learning methods learn high-quality visual representations, yet recent studies show that these representations often capture demographic biases present in the training data. Existing fairness-aware methods address this by redesigning the self-supervised objective itself, limiting portability across the rapidly evolving landscape of self-supervised learning (SSL) frameworks. We propose ProtoFair, a fairness-aware contrastive loss designed to work alongside existing SSL objectives without modifying them. ProtoFair leverages unsupervised prototype clustering to identify pseudo-counterfactual pairs: samples sharing the same cluster assignment but belonging to different sensitive groups. By pulling these content-matched, cross-group samples together in the embedding space, ProtoFair encourages the encoder to learn representations that are invariant to the sensitive attribute. The method requires only sensitive attribute annotations, no target labels, and integrates seamlessly with both SimCLR and SupCon. Experiments on CelebA and UTKFace demonstrate consistent fairness improvements while maintaining competitive accuracy.

---

### Trojan Hippo Bench: A Dynamic Benchmark for Persistent Memory Attacks and Defenses in LLM Agents

Authors: Debeshee Das, Julien Piet, Darya Kaviani, Luca Beurer-Kellner, Florian Tramèr, David Wagner
Published: 2026-05-03
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2605.01970v4

Memory systems enable otherwise stateless LLM agents to persist user information across sessions, but also introduce a new attack surface. The Trojan Hippo attack is a class of persistent memory attacks that operates under a more realistic threat model than prior memory poisoning work. The attacker plants a dormant payload into an agent's long-term memory via a single untrusted tool call (e.g., a crafted email), which activates only when the user later discusses sensitive topics such as finance, health, or identity, and exfiltrates high-value personal data to the attacker. While anecdotal demonstrations of such attacks have appeared against deployed systems, no prior work systematically evaluates them across heterogeneous memory architectures and defenses. We introduce Trojan Hippo Bench, a dynamic evaluation framework for persistent memory attacks and memory-layer defenses, comprising two components, (1) an OpenEvolve-based adaptive red-teaming benchmark that stress-tests defenses and memory backends against continuously refined attacks, and (2) a capability-aware security-utility analysis for persistent memory systems, enabling principled reasoning about defense deployment across different usage profiles. Instantiated on an email assistant across four memory backends (explicit tool memory, agentic memory, RAG, and sliding-window context), the undefended Trojan Hippo attack achieves up to 85-100% attack success rate (ASR) against current frontier models from OpenAI and Google, with planted memories successfully activating even after 100 benign sessions. We evaluate four memory-system defenses inspired by basic security principles; they reduce attack success to as low as 0-5%, but at utility costs that vary widely with the task mix. Effective real-world deployment thus remains an open challenge that Trojan Hippo Bench is built to study; we release it to support reuse and extension.

---

### Trojan Hippo: Weaponizing Agent Memory for Data Exfiltration

Authors: Debeshee Das, Julien Piet, Darya Kaviani, Luca Beurer-Kellner, Florian Tramèr, David Wagner
Published: 2026-05-03
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2605.01970v3

Memory systems enable otherwise-stateless LLM agents to persist user information across sessions, but also introduce a new attack surface. We characterize the Trojan Hippo attack, a class of persistent memory attacks that operates in a more realistic threat model than prior memory poisoning work: the attacker plants a dormant payload into an agent's long-term memory via a single untrusted tool call (e.g., a crafted email), which activates only when the user later discusses sensitive topics such as finance, health, or identity, and exfiltrates high-value personal data to the attacker. While anecdotal demonstrations of such attacks have appeared against deployed systems, no prior work systematically evaluates them across heterogeneous memory architectures and defenses. We introduce a dynamic evaluation framework comprising two components: (1) an OpenEvolve-based adaptive red-teaming benchmark that stress-tests defenses and memory backends against continuously refined attacks, and (2) the first capability-aware security/utility analysis for persistent memory systems, enabling principled reasoning about defense deployment across different usage profiles. Instantiated on an email assistant across four memory backends (explicit tool memory, agentic memory, RAG, and sliding-window context), Trojan Hippo achieves up to 85-100% ASR against current frontier models from OpenAI and Google, with planted memories successfully activating even after 100 benign sessions. We evaluate four memory-system defenses inspired by basic security principles, finding they substantially reduce attack success rates (to as low as 0-5%), though at utility costs that vary widely with task requirements. Because of this substantial security-utility tradeoff, the effective real-world deployment of defenses remains an open challenge, which our evaluation framework is specifically designed to address.

---

### AdamO: A Collapse-Suppressed Optimizer for Offline RL

Authors: Nan Qiao, Sheng Yue, Shuning Wang, Ju Ren
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.01968v1

Offline reinforcement learning (RL) can fail spectacularly when bootstrapped temporal-difference (TD) updates amplify their own errors, driving the critic toward extreme and unusable Q-values. A key counterintuitive insight of this work is that collapse is not only a property of the backup rule or network architecture: optimizer dynamics themselves can directly trigger or suppress instability. From a control-theoretic viewpoint, we model offline TD learning as a feedback system and analyze Adam-based critic updates. This yields a necessary and sufficient condition for stability of the induced local update dynamics: within the regime we analyze, these dynamics are stable if and only if the spectral radius of the corresponding update operator is strictly below one. Further analysis suggests that standard Adam updates can inadvertently distort the parameter geometry, motivating explicit orthogonality constraints to prevent TD error amplification. To this end, we propose AdamO, an Adam-based optimizer with a decoupled orthogonality correction regulated by a strict task-alignment budget. We prove that this design theoretically guarantees worst-case task safety and preserves Adam's continuous-time dissipative dynamics. Empirically, AdamO is broadly compatible with diverse offline RL baselines, improving stability and returns across a broad suite of benchmarks.

---

### MER-DG: Modality-Entropy Regularization for Multimodal Domain Generalization

Authors: Yavuz Yarici, Ghassan AlRegib
Published: 2026-05-03
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2605.01967v1

Deploying multimodal models in real-world scenarios requires generalization to new environments where recording conditions differ from training, a challenge known as multimodal domain generalization (MMDG). Standard architectures employ separate encoders for each modality and a fusion module, training the system end-to-end by optimizing on the fused features. In this paper, we identify that such joint optimization causes encoders to exploit cross-modal co-occurrences, statistical relationships between modalities that arise from source-specific recording conditions, rather than learning domain-invariant features. We term this failure mode Fusion Overfitting. To address this, we propose Modality-Entropy Regularization for Domain Generalization (MER-DG), which maximizes the entropy of each encoder's feature distribution to preserve feature diversity. MER-DG is architecture-agnostic and integrates into existing multimodal frameworks as an additive loss term. Extensive experiments on EPIC-Kitchens and HAC benchmarks demonstrate average improvements of approximately 5% over standard fusion and approximately 2% over state-of-the-art methods.

---

### Retrieval with Multiple Query Vectors through Anomalous Pattern Detection

Authors: Allassan Tchangmena A Nken, Baimam Boukar Jean Jacques, Miriam Rateike, Celia Cintas, Skyler Speakman
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.01965v2

A classical vector retrieval problem typically considers a \emph{single} query embedding vector as input and retrieves the most similar embedding vectors from a vector database. However, complex reasoning and retrieval tasks frequently require \emph{multiple query vectors}, rather than a single one. In this work, we propose a retrieval method that considers multiple query vectors simultaneously and retrieves the most relevant vectors from the database using concepts from anomalous pattern detection. Specifically, our approach leverages a set of query vectors $Q$ (with $|Q|\geq 1$), and identifies the subset of vector dimensions within $Q$ that standout (anomalous) from the rest of dimensions. Next, we scan the vector database to retrieve the set of vectors that are also anomalous across the previously identified vector dimensions and return them as our retrieved set of vectors. We validate our approach on two image datasets, a text dataset, and a tabular dataset. Overall, we observe that, across most datasets, larger query sets lead to improved retrieval performance. The improvement is most pronounced when increasing the query sets from 1 to 8, while the gains become smaller beyond that.

---

### Multi-User Dueling Bandits: A Fair Approach using Nash Social Welfare

Authors: Maheed H. Ahmed, Mahsa Ghasemi
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.01961v3

Learning from human preference data is becoming a useful tool, from fine-tuning large language models to training reinforcement learning agents. However, in most scenarios, the model is trained on the average preference of all human evaluators, which, under large variations of preferences, can be unfair to minority groups. In this work, we consider fairness in dueling bandits, a standard framework for online learning from preference data. We assume that each user has a (potentially distinct) Condorcet winner, which is an arm preferred to every other arm. Using these user-specific Condorcet winners as reference points, we evaluate and score arms according to their performance relative to the corresponding winner. To promote fairness across heterogeneous users, we adopt the well-established Nash Social Welfare objective, which maximizes the product of user utilities, thereby inherently penalizing inequality and preventing the marginalization of any single user. Within this framework, we construct a hard instance to establish a regret lower bound of $Ω(T^{2/3}\min(K,D)^\frac{1}{3})$ for a time horizon $T$, $K$ arms, and $D$ users, which, to the best of our knowledge, is the first result quantifying the cost of fairness in dueling bandits with heterogeneous preferences. We then present the Fair-Explore-Then-Commit and Fair-$ε$-Greedy algorithms with a Condorcet winner identification phase. We further derive their regret upper bounds that match the lower-bound dependence on $T$ up to logarithmic factors.

---

### Multi-User Dueling Bandits: A Fair Approach using Nash Social Welfare

Authors: Maheed H. Ahmed, Mahsa Ghasemi
Published: 2026-05-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.01961v2

Learning from human preference data is becoming a useful tool, from fine-tuning large language models to training reinforcement learning agents. However, in most scenarios, the model is trained on the average preference of all human evaluators, which, under large variations of preferences, can be unfair to minority groups. In this work, we consider fairness in dueling bandits, a standard framework for online learning from preference data. We assume that each user has a (potentially distinct) Condorcet winner, which is an arm preferred to every other arm. Using these user-specific Condorcet winners as reference points, we evaluate and score arms according to their performance relative to the corresponding winner. To promote fairness across heterogeneous users, we adopt the well-established Nash Social Welfare objective, which maximizes the product of user utilities, thereby inherently penalizing inequality and preventing the marginalization of any single user. Within this framework, we construct a hard instance to establish a regret lower bound of $Ω(T^{2/3}\min(K,D)^\frac{1}{3})$ for a time horizon $T$, $K$ arms, and $D$ users, which, to the best of our knowledge, is the first result quantifying the cost of fairness in dueling bandits with heterogeneous preferences. We then present the Fair-Explore-Then-Commit and Fair-$ε$-Greedy algorithms with a Condorcet winner identification phase. We further derive their regret upper bounds that match the lower-bound dependence on $T$ up to logarithmic factors.

---

### Flexi-LoRA with Input-Adaptive Ranks: Efficient Finetuning for Speech and Reasoning Tasks

Authors: Zongqian Li, Yixuan Su, Han Zhou, Zihao Fu, Nigel Collier
Published: 2026-05-03
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2605.01959v1

Parameter-efficient fine-tuning methods like Low-Rank Adaptation (LoRA) have become essential for deploying large language models, yet their static parameter allocation remains suboptimal for inputs of varying complexity. We present Flexi-LoRA, a novel framework that dynamically adjusts LoRA ranks based on input complexity during both training and inference. Through empirical analysis across question answering, mathematical reasoning, and speech tasks, we demonstrate that maintaining consistency between training and inference dynamics is important for effective adaptation, particularly for sequential reasoning tasks. Our findings reveal that input-dependent parameter allocation achieves higher performance with fewer parameters by optimally matching rank configurations to question complexity. Furthermore, task-specific dependency on rank dynamics varies, with mathematical reasoning tasks exhibiting higher dependency than QA tasks. Successful adaptation manifests not only in correctness but also in reasoning quality and instruction adherence. Flexi-LoRA consistently outperforms static LoRA while using fewer parameters, with performance gains more pronounced on tasks requiring strict reasoning chains. Our approach realizes key benefits of mixture-of-experts frameworks through a more streamlined implementation, reducing parameter redundancy while improving model capabilities. We provide comprehensive empirical studies across diverse tasks, establishing a basis for future work in input-adaptive and efficient fine-tuning approaches.

---

### LLM-Augmented Semantic Steering of Text Embedding Projection Spaces

Authors: Wei Liu, Eric Krokos, Kirsten Whitley, Rebecca Faust, Chris North
Published: 2026-05-03
Categories: cs.HC, cs.CL
arXiv: https://arxiv.org/abs/2605.01957v1

Low-dimensional projections of text embeddings support visual analysis of document collections, but their spatial organization may not reflect the relationships an analyst intends to examine. Existing semantic interaction approaches encode semantic intent indirectly through geometric constraints or model updates, limiting interpretability and flexibility. We introduce LLM-augmented semantic steering, which enables analysts to express semantic intent by grouping a small set of example documents within the projection. A large language model externalizes this intent as natural-language representations and selectively extends it to related documents; the resulting semantic information is then incorporated into document representations via text augmentation or embedding-level blending, without retraining the underlying models. A case study illustrates how the same corpus can be reorganized from different semantic perspectives, while simulation-based evaluation shows that semantic steering improves global and local alignment with target semantic structures using only minimal interaction. Embedding-level blending further enables continuous and controllable steering of projection layouts. These results position projection spaces as intent-dependent semantic workspaces that can be reshaped through explicit, interpretable, language-mediated interaction.

---

### Moira: Language-driven Hierarchical Reinforcement Learning for Pair Trading

Authors: Polydoros Giannouris, Yuechen Jiang, Lingfei Qian, Yuyan Wang, Xueqing Peng, Jimin Huang, Guojun Xiong, Sophia Ananiadou
Published: 2026-05-03
Categories: cs.AI, cs.CL, cs.MA
arXiv: https://arxiv.org/abs/2605.01954v1

Many sequential decision-making problems exhibit hierarchical structure, where high-level semantic choices constrain downstream actions and feedback is delayed and ambiguous. Learning in such settings is challenging due to credit assignment: performance degradation may arise from flawed abstractions, suboptimal execution, or their interaction. We study this challenge through pair trading, a domain that naturally combines long-horizon semantic reasoning for asset pair selection with short-horizon execution under partial observability. We formulate pair trading as a hierarchical reinforcement learning problem and propose a language-driven optimization framework in which both high-level and low-level policies are parameterized by large language models (LLMs) and optimized exclusively through prompt updates. Our approach leverages pretrained LLMs as hierarchical policies and uses trajectory- and episode-level textual feedback to adapt abstractions and execution without gradient-based fine-tuning. By explicitly separating abstraction selection from execution, the framework reduces non-stationarity across hierarchical levels and enables targeted adaptation under delayed feedback. Experiments on real-world market data show consistent improvements over traditional and LLM-based baselines, demonstrating the effectiveness of language-driven hierarchical reinforcement learning.

---

### TRAP: Tail-aware Ranking Attack for World-Model Planning

Authors: Siyuan Duan, Ke Zhang, Xizhao Luo
Published: 2026-05-03
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2605.01950v1

World models enable long-horizon planning by internally generating and evaluating imagined trajectories, making them a promising foundation for generalist agents. However, this imagination-driven decision process also introduces new security risks. Existing backdoor attacks typically aim to manipulate local features, one-step predictions, or instantaneous policy outputs. While such objectives may suffice for weaker reactive models, they are often ineffective against world models, where the learned dynamics prior and planning process can absorb or wash out the effects of shallow perturbations. More importantly, we find that world models exhibit a distinct backdoor vulnerability rooted in the long-tailed ranking structure of imagined trajectories, where disrupting the ordering of a few decision-critical trajectories can systematically hijack planning. To exploit this vulnerability, we propose TRAP, a backdoor attack framework for world models that targets imagined trajectory ranking. TRAP combines a tail-aware ranking loss to focus optimization on decision-critical trajectories with dual gating mechanisms that stabilize optimization and regulate when and where the attack penalty is applied. Under trigger conditions, TRAP alters the relative ranking of imagined trajectories to redirect planning outcomes, while largely maintaining the normal ranking structure on clean inputs. Experiments on DreamerV3 and TD-MPC2 across diverse tasks show that TRAP consistently induces sustained behavioral deviations and significant performance degradation, highlighting the need for dedicated security evaluation of world-model-based agents.

---

### Sonar-GPS Fusion for Seabed Mapping in Turbid Shallow Waters with an Autonomous Surface Vehicle

Authors: Yisheng Zhang, Michael Xu, Alan Williams, Matthew Gray, Nare Karapetyan, Miao Yu
Published: 2026-05-03
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2605.01949v1

Accurate seabed mapping is essential for habitat monitoring and infrastructure inspection. In turbid, shallow coastal waters, such as shellfish aquaculture farms, the effectiveness of traditional optical methods is limited. Autonomous surface vehicles (ASVs) equipped with forward-looking sonar (FLS) offer a promising alternative. However, existing sonar-based systems face challenges in achieving fine resolution mapping over long trajectories due to low-resolution positioning measurements and accumulated drift over long trajectories. In this paper, we present a drift-resilient seabed mapping framework that integrates local FLS frame alignment using the Fourier-Mellin transform (FMT) with global trajectory optimization based on an extended Kalman filter (EKF) that fuses global positioning system (GPS), inertial measurement unit (IMU), and compass data. A variance-based image blending strategy is used to further reduce visual artifacts in overlapping regions. Field trials on a structured oyster farm site show that our framework helps reduce drift in RMSE by 9.5% relative to the FMT-only baseline. This framework also enables sub-meter reconstruction accuracy and preservation of high-resolution textures needed for oyster inventory estimation within the mapped areas.

---

### Phone2Act: A Low-Cost, Hardware-Agnostic Teleoperation System for Scalable VLA Data Collection

Authors: Om Mandhane, Bipin Yadav, Sangeetha Prasanna Ram, Gopalakrishnan Narayanan
Published: 2026-05-03
Categories: cs.RO, cs.AI, cs.HC
arXiv: https://arxiv.org/abs/2605.01948v1

Collecting diverse, high-quality manipulation data for Vision-Language-Action (VLA) model training remains prohibitively expensive for many research groups, as existing teleoperation frameworks rely on specialized hardware or are tightly coupled to specific robot platforms. We present Phone2Act, a low-cost, hardware-agnostic teleoperation framework that transforms a commodity smartphone into a 6-DoF robot controller via Google ARCore. Built on a modular ROS 2 architecture, Phone2Act decouples control logic from hardware specifics through interchangeable bridge nodes, supporting platforms from industrial cobots to low-cost bimanual arms without code modification. A Universal Recorder synchronizes multi-camera RGB streams with robot state feedback and exports demonstrations natively in the LeRobot dataset format, eliminating post-processing and enabling immediate VLA fine-tuning. We validate the framework by fine-tuning GR00T-N1.5 on 130 collected episodes, achieving a 90% success rate on a real-world multi-stage pick-and-place task deployed on a physical Dobot CR5.

---