Academic Intelligence · Curated Daily

探索全球前沿学术脉络

AcademicHub 汇聚顶级期刊与预印本平台的实时文献。定制您的专属科研雷达,利用大语言模型自动生成交叉领域文献分析简报。

01.
arXiv (CS.AI) 2026-06-12

A Minimal Model of Bounded Trade-Off Screening in Multi-Attribute Choice

arXiv:2606.13201v1 Announce Type: new Abstract: Human decision-making often involves choosing between multi-attribute alternatives, yet classical models assume fully compensatory utility aggregation despite evidence that people reject options with poor performance on critical attributes. We propose a bounded trade-off reasoning framework in which decisions are governed by a screening process that evaluates the balance between gains and losses across attributes. The model introduces a trade-off tolerance parameter that controls acceptable imbalance and can vary across contexts. Through simulation, we show that this mechanism produces preference patterns that differ from standard utility-based models and captures context-dependent variation in trade-off behavior. These results establish bounded trade-off screening as a plausible computational mechanism for multi-attribute choice and generate testable predictions for future behavioral studies.

02.
arXiv (CS.CV) 2026-06-16

BioAutoML-NAS: An End-to-End AutoML Framework for Multimodal Insect Classification via Neural Architecture Search on Large-Scale Biodiversity Data

Insect classification is important for agricultural management and ecological research, as it directly affects crop health and production. However, this task remains challenging due to the complex characteristics of insects, class imbalance, and large-scale datasets. To address these issues, we propose BioAutoML-NAS, the first BioAutoML model using multimodal data, including images, and metadata, which applies neural architecture search (NAS) for images to automatically learn the best operations for each connection within each cell. Multiple cells are stacked to form the full network, each extracting detailed image feature representations. A multimodal fusion module combines image embeddings with metadata, allowing the model to use both visual and categorical biological information to classify insects. An alternating bi-level optimization training strategy jointly updates network weights and architecture parameters, while zero operations remove less important connections, producing sparse, efficient, and high-performing architectures. Extensive evaluation on the BIOSCAN-5M dataset demonstrates that BioAutoML-NAS achieves 96.81% accuracy, 97.46% precision, 96.81% recall, and a 97.05% F1 score, outperforming state-of-the-art transfer learning, transformer, AutoML, and NAS methods by approximately 16%, 10%, and 8% respectively. Further validation on the Insects-1M dataset obtains 93.25% accuracy, 93.71% precision, 92.74% recall, and a 93.22% F1 score. These results demonstrate that BioAutoML-NAS provides accurate, confident insect classification that supports modern sustainable farming.

03.
arXiv (CS.CV) 2026-06-15

SED:Lightweight Saliency prediction for Event-based data via Distillation

Event-based saliency prediction has gained attention recently, as combining event cameras with saliency estimation can act as an upstream stage that naturally improves the efficiency of downstream eventbased perception at the edge. However, current approaches are either neuromorphic, underperforming on event-based saliency benchmarks, or too heavy for resource-constrained edge applications due to their reliance on transformers or 3D convolutions. Drawing inspiration from efficient convolutional modules, SED and aiming to exploit the temporal information in event data, we propose a lightweight network, trained through knowledge distillation, built on a Depthwise Spatio-Temporal Block (DSTconv) – a factorization of the 3D depthwise separable convolution. Relative to its teacher, our model reduces the model size from 180 MB to 0.32 MB (562x) and the parameter count from 45M to 81k (554x), while matching or outperforming it on the N-DHF1K and N-UCF Sports datasets. Moreover, it generalizes strongly beyond its training distribution, transferring from synthetic to real event data where a model trained from scratch fails.

04.
arXiv (quant-ph) 2026-06-17

DRAG-Compatible Leakage Suppression in Landau–Zener Control via Isoprobability Twins

arXiv:2506.19572v4 Announce Type: replace Abstract: Analytically solvable models – particularly the Landau-Majorana-Stückelberg-Zener (LMSZ) and Allen-Eberly-Hioe (AEH) models – underpin many quantum-gate implementations and population-transfer protocols. However, their canonical pulse shapes are incompatible with modern leakage-suppression techniques and some systems. Most notably, the constant Rabi envelope of the LMSZ pulse prevents many leakage-suppression approaches, which require smoothness. We address both limitations by developing the concept of isoprobability twin models: distinct pairs of Rabi frequency $\Omega(t)$ and detuning $\Delta(t)$ that yield identical post-pulse transition probabilities based on the Delos-Thorson transformation. In this work, we formalise the method by experimentally demonstrating the equivalence of multiple LMSZ and AEH twin models on IBM's ibm_kyiv processor. Finally, we show a staggering leakage reduction by more than 3 orders of magnitude using a custom DRAG implementation of a cosine LMSZ isoprobability model.

05.
arXiv (CS.AI) 2026-06-15

Optimizing Agentic Reasoning with Retrieval via Synthetic Semantic Information Gain Reward

arXiv:2602.00845v3 Announce Type: replace Abstract: Agentic reasoning enables large reasoning models (LRMs) to dynamically acquire external knowledge, but yet optimizing the retrieval process remains challenging due to the lack of dense, principled reward signals. In this paper, we introduce InfoReasoner, a unified framework that incentivizes effective information seeking via a synthetic semantic information gain reward. Theoretically, we redefine information gain as uncertainty reduction over the model's belief states, establishing guarantees, including non-negativity, telescoping additivity, and channel monotonicity. Practically, to enable scalable optimization without manual retrieval annotations, we propose an output-aware intrinsic estimator that computes information gain directly from the model's output distributions using semantic clustering via bidirectional textual entailment. This intrinsic reward guides the policy to maximize epistemic progress, enabling efficient training via Group Relative Policy Optimization (GRPO). Experiments across seven question-answering benchmarks demonstrate that InfoReasoner consistently outperforms strong retrieval-augmented baselines, achieving up to 5.4% average accuracy improvement. Our work provides a theoretically grounded and scalable path toward agentic reasoning with retrieval. The code is available at https://github.com/dl-m9/InfoReasoner

07.
arXiv (CS.CV) 2026-06-17

Attention Alignment Between Humans and Vision-Language Models

Visual perception depends on top-down goals and bottom-up sensory mechanisms. Vision-language models implement both, allowing us to treat each component as a separable hypothesis about what drives where we look. We compared spatial attention maps from six vision-language models against human fixation heatmaps recorded on 200 images during two tasks (general description and social captioning). The six models spanned a 2$\times$2 factorial of CNN vs.\ ViT encoders crossed with LSTM vs.\ Transformer decoders, plus Molmo 7B-D and Qwen3.5 9B. We found that both decoder and encoder architecture shaped alignment, but decoder choice dominated. LSTM vs.\ Transformer decoders increased alignment by 40–50 percentage points (80–87\% vs.\ 40–59\% of the human noise ceiling). In contrast, CNN vs.\ ViT encoders contributed a secondary 5–20 point advantage depending on decoder family, with CNN-LSTM the most aligned model overall (85–87\%). Despite their alignment advantage, LSTM-decoder attention maps were spatially diffuse and minimally task-differentiated; ViT-Transformer, the weakest in alignment, showed the sharpest spatial concentration and strongest task differentiation. A hemispatial-neglect simulation confirmed that ablating attention impacted LSTM decoders more than Transformer decoders. In an exploratory extension using TRIBE-simulated synthetic neural responses, fixation alignment and neural relevance dissociate: CNN-Transformer attention maps better predicted synthetic brain activity despite lower fixation alignment, with attention maps best predicting early visual cortex. Together, top-down and bottom-up components trade off what they predict in behavioral and synthetic neural data.

08.
arXiv (CS.AI) 2026-06-18

Bounded Context Management for Tabular Foundation Models on Stream Learning

arXiv:2606.18677v1 Announce Type: cross Abstract: Tabular stream learning requires predictions on sequentially arriving examples under distribution shift. While standard methods adapt by updating model states, tabular foundation models (TFMs) make predictions conditioned on a labeled context in an in-context manner, making them a natural alternative for stream learning. This shifts the challenge from how to update the model to how to manage the context. We propose a future information view that yields three practical requirements for context management: preserve recent examples, retain uncertain examples, and remove redundant examples. We instantiate these requirements as CURE (Context management via Uncertainty-aware admission and Redundancy aware Eviction), a context-managing policy with entropy-gated admission and redundancy-aware eviction. Across seven streams, CURE shows up to 27.0% relative improvement over classical stream learners, remains robust across multiple TFM backbones, and ranks first among other policy variants. Code and datasets are available at https://github.com/morcellinus/CURE-ICML-FMSD.

09.
arXiv (CS.AI) 2026-06-17

SkillJect: Effectively Automating Skill-Based Prompt Injection for Skill-Enabled Agents

arXiv:2602.14211v3 Announce Type: replace-cross Abstract: Agent skills extend LLM agents with task-specific instructions, executable scripts, and auxiliary resources, improving reusability but creating a new supply-chain attack surface. A malicious or compromised skill can be repeatedly loaded as trusted guidance and steer downstream tool use. Existing skill-based prompt-injection attacks are often manual and brittle, because explicit malicious instructions are rejected or ignored when they are not aligned with the original workflow. We propose SkillJect, the first automated framework for generating poisoned skills against skill-enabled agent systems. SkillJect uses two coordinated channels. In the artifact channel, it hides the payload inside an auxiliary helper script. In the instruction channel, it rewrites SKILL.md with a front-loaded inducement strategy, placing injected content at the beginning and framing the helper script as a mandatory prerequisite or initialization step. The rewritten instruction explicitly references the helper-script path and provides an executable example command, making the helper appear to be a legitimate setup step before normal skill operations. SkillJect further adopts a closed-loop multi-agent process to improve attack effectiveness. An Attack Agent generates poisoned skills, a Victim Agent executes downstream tasks with the poisoned skill, and an Evaluate Agent inspects execution traces to determine whether the hidden payload was executed. The Attack Agent then uses this feedback to diagnose failure causes and rewrite SKILL.md, while keeping the payload fixed. Experiments across skill-enabled platforms, backend LLMs, and attack categories show that SkillJect substantially outperforms naive direct injection and prior manual skill-injection attacks, highlighting poisoned skills as a persistent threat in reusable skill ecosystems.

10.
medRxiv (Medicine) 2026-06-24

Self-administered computerized cognitive training for cognitive deficits in individuals with metabolic syndrome: a randomized controlled trial

Background: Metabolic syndrome (MetS) has been associated with cognitive decline. Considering its increasing prevalence worldwide, the goal of this study was to evaluate the feasibility and efficacy of a short-term, self-administered computerized cognitive training programme in individuals with metabolic syndrome and low cognitive performances. Methods: Thirty six participants, aged 40-72 years (mean age: 57.8 years), were randomly assigned to the cognitive training or the passive control group. The cognitive training component of Long Lasting Memories (LLM) Care was used as an interactive software to enhance participants' cognitive functions. Up to 24 sessions, each lasting 45 minutes, were self-administered at home twice per week for 3 months. Thorough cognitive assessments with were performed at baseline (randomization), at the end of intervention, and 12 months after baseline. The primary outcome was performance at nine neuropsychological tests, and the secondary outcome was a self-reported questionnaire assessing everyday functional abilities. Primary analyses were performed employing mixed-effect models using the intention-to-treat principle. Results: Low adherence was observed in the study, as only 9 participants (50%) completed at least 8 sessions of the cognitive training programme (range 9-24 sessions, median 15 sessions). No statistically significant effect of the cognitive training programme on performance in neuropsychological tests or everyday functioning was found. At the end of the 3-month intervention programme, effect for visual memory enhancement in immediate ({beta} = 1.58, 95% CI = -1.84 to 4.99, Cohen's d = 0.39) and delayed recall ({beta} = 2.17, 95% CI = -1.68 to 6.01, Cohen's d = 0.45) was moderate in favour of the intervention group, and at 12-month follow-up, semantic verbal fluency gains for the intervention group were detected ({beta} = 2.78, 95% CI = -0.92 to 6.49, Cohen's d = 0.70), though with wide confidence intervals. Conclusions: Despite some small effects observed in memory and verbal fluency, cognitive training did not yield statistically significant improvements. The observed low adherence and limited benefits on mild cognitive deficits in mostly middle-aged individuals with MetS are likely associated with the self-administered and short-term nature of the computerized intervention. This highlights the need for more intensive and clinician-delivered approaches to enhance engagement. Registry: ClinicalTrials.gov, TRN: NCT05658354, Registration date: 08 December 2022. Keywords: Metabolic syndrome, cognitive deficits, cognitive training, computerized, adults

11.
medRxiv (Medicine) 2026-06-22

Anterior-superior hypothalamic enlargement as specific marker in episodic migraine: converging evidence from an independent discovery-replication design

Background: Growing evidence implicates the hypothalamus as a key structure in migraine pathophysiology; however, our understanding of its precise role and of the specific nuclei involved remains limited. We combined MRI data from our laboratory with publicly available MRI datasets from OpenNeuro to examine hypothalamic subunit volumes in episodic migraine and assess the specificity of these alterations relative to chronic pain conditions. Methods: Structural MRI combined with an automated atlas-based segmentation algorithm and a discovery-replication design was employed to investigate cross-sectional volumetric differences across 5 bilateral hypothalamic subunits in two independent migraine cohorts: DS1-MIG (DS1-MIG-base, n = 111 patients, n = 35 controls) and DS2-MIG (n = 27 patients, n = 31 controls). The adjusted volumes were compared between groups using MANOVA as an omnibus test, followed by Welch t-tests to test univariate follow-up. Longitudinal volumetric changes were additionally assessed in DS1-MIG participants with available follow-up scans using linear mixed models. To assess the specificity of findings to migraine, the same pipeline was applied to two chronic pain datasets, one including patients with fibromyalgia (DS-FM, n = 33 patients, n = 33 controls) and the other including patients with trigeminal neuralgia (n = 119 patients, n = 55 controls). Results: MANOVA revealed significant multivariate group differences in the discovery and replication migraine cohorts (DS1-MIG-base: = .006; DS2-MIG: = .008). Follow-up univariate analyses identified a consistent enlargement of the left anterior-superior subunit across both cohorts (FDR = .023 in DS1-MIG-base and FDR = .046 in DS2-MIG), representing the only cross-cohort replication finding. Beyond this shared signature, DS2-MIG exhibited additional significant enlargements of the right anterior-inferior and right tubular-inferior subunits. Longitudinal analyses in DS1-MIG showed that hypothalamic subunit volumes remained broadly stable over time within both migraine patients and control participants. No significant volumetric alterations were detected in the fibromyalgia or trigeminal neuralgia cohorts, either in multivariate or univariate analyses, underscoring migraine-specific findings. Conclusions: These findings provide evidence for subunit-specific hypothalamic structural alterations in migraine localized in the left anterior hypothalamic subunit. The stability of these differences over time and their absence in other chronic pain conditions suggest a migraine-specific structural organisation of hypothalamic circuitry.

12.
arXiv (CS.LG) 2026-06-15

Time Series Causal Discovery via Context-Conditioned and Causality-Augmented Pretraining

arXiv:2605.26759v2 Announce Type: replace Abstract: Causal discovery from time series is critical for many real-world applications, such as tracing the root causes of anomalies. Existing approaches typically rely on dataset-specific optimization, making it difficult to transfer their causal discovery capabilities to new time series governed by diverse causal mechanisms. In this paper, we propose PTCD, a novel Pretraining framework for Time-series Causal Discovery, which improves cross-task generalization through context-conditioned modeling and transferable causal augmentation. To model complex temporal causal dependencies, PTCD employs a dual-scale iterative attention mechanism to capture window-level causal relationships, and a Gaussian mixture with a context-level routing mechanism to handle heterogeneous exogenous distributions. To further address distribution shifts across causal graphs, PTCD adopts a pretraining paradigm on synthetic datasets that integrates intervention-based learning and a causal mixup strategy, promoting stable causal discovery and stronger generalization. Extensive experiments on multiple real-world out-of-distribution (OOD) datasets demonstrate that PTCD excels in both causal discovery and root cause identification.

13.
arXiv (CS.CL) 2026-06-24

Faithful by Construction: Claim-Anchored Attribution for Multi-Document Summarization

作者:

End-to-end large language models (LLMs) produce fluent multi-document summaries but remain prone to hallucination, and the attributions they offer are typically coarse (whole documents or passages) and generated post hoc, leaving each summary statement hard to verify. We revisit the modular Extract–Select–Rewrite paradigm and recast its intermediate representation as the unit of attribution. We present CAMS, a Claim-Anchored Multi-document Summarization framework that (i) extracts atomic claims with token-level provenance from every source document, (ii) clusters equivalent claims across documents while flagging inter-source conflicts, (iii) selects a support-aware and salient subset, and (iv) rewrites the selection into a summary in which every sentence is anchored to a support-checked claim that links back to one or more source spans. Because content is localized before it is realized, the pipeline is attribution-oriented by construction and faithfulness-oriented by construction: it structurally preserves fine-grained, multi-source traceability while using support-aware selection, constrained rewriting, and verification to encourage, rather than guarantee, factual faithfulness. We evaluate quality, faithfulness, and localization on MultiNews, analyze conflict handling on DiverseSumm, and test zero-shot transfer on WCEP, using a two-regime protocol that separates reference-free citation quality from gold-aligned localization accuracy, and we add an evaluator-decoupled audit that tests citation precision with a support model never used for selection or verification. CAMS matches strong end-to-end and span-attribution baselines on summary quality while substantially improving faithfulness and citation precision, lifting multi-source attribution accuracy by roughly two-thirds, and exposing a controllable faithfulness–coverage trade-off that end-to-end models leave implicit.

14.
arXiv (CS.CL) 2026-06-16

Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents

Online web agents often augment a base actor with memory, workflow, or skill modules. These modules can improve performance, but they also consume test-time tokens, a cost rarely reported alongside the actor's inference cost. We study online augmentation, where this overhead is paid on every task, and re-evaluate its benefits under a fixed total inference budget. We compare AWM, ASI, and ReasoningBank with a token-matched vanilla baseline that uses the same budget for additional actor steps. Across three WebArena domains and three models, Gemini 3 Flash, GPT-5.4-mini, and Qwen 3.6-27B, the vanilla baseline matches or surpasses all three augmentation methods in aggregate success rate while often using fewer total tokens. We observe a similar trend on WorkArena-L1 with Qwen 3.6-27B, indicating that the effect extends to enterprise knowledge-work tasks. Our results suggest that skills and workflow memory can be useful in specific domains, but their apparent gains often vanish against a budget-matched actor. We further show that run-to-run variance materially affects outcomes and should be reported as a core evaluation criterion for online web agents.

15.
arXiv (CS.AI) 2026-06-18

Large-Scale OD Matrix Estimation with A Deep Learning Method

arXiv:2310.05753v2 Announce Type: replace Abstract: The estimation of origin-destination (OD) matrices is a crucial aspect of Intelligent Transport Systems (ITS). It involves adjusting an initial OD matrix by regressing the current observations like traffic counts of road sections (e.g., using least squares). However, the OD estimation problem lacks sufficient constraints and is mathematically underdetermined. To alleviate this problem, some researchers incorporate a prior OD matrix as a target in the regression to provide more structural constraints. However, this approach is highly dependent on the existing prior matrix, which may be outdated. Others add structural constraints through sensor data, such as vehicle trajectory and speed, which can reflect more current structural constraints in real-time. Our proposed method integrates deep learning and numerical optimization algorithms to infer matrix structure and guide numerical optimization. This approach combines the advantages of both deep learning and numerical optimization algorithms. The neural network(NN) learns to infer structural constraints from probe traffic flows, eliminating dependence on prior information and providing real-time performance. Additionally, due to the generalization capability of NN, this method is economical in engineering. We conducted tests to demonstrate the good generalization performance of our method on a large-scale synthetic dataset. Subsequently, we verified the stability of our method on real traffic data. Our experiments provided confirmation of the benefits of combining NN and numerical optimization.

16.
arXiv (CS.LG) 2026-06-11

SpaTeoGL: Spatiotemporal Graph Learning for Interpretable Seizure Onset Zone Analysis from Intracranial EEG

arXiv:2602.11801v2 Announce Type: replace Abstract: Accurate localization of the seizure onset zone (SOZ) from intracranial EEG (iEEG) is essential for epilepsy surgery but is challenged by complex spatiotemporal seizure dynamics. We propose SpaTeoGL, a spatiotemporal graph learning framework for interpretable seizure network analysis. SpaTeoGL jointly learns window-level spatial graphs capturing interactions among iEEG electrodes and a temporal graph linking time windows based on similarity of their spatial structure. The method is formulated within a smooth graph signal processing framework and solved via an alternating block coordinate descent algorithm with convergence guarantees. Experiments on a multicenter iEEG dataset with successful surgical outcomes show that SpaTeoGL is competitive with a baseline based on horizontal visibility graphs and logistic regression, while improving non-SOZ identification and providing interpretable insights into seizure onset and propagation dynamics.

17.
arXiv (CS.AI) 2026-06-16

Open-SWE-Traces: Advancing Dual-Mode Multilingual Distillation for Software Engineering Agents

arXiv:2606.16038v1 Announce Type: cross Abstract: The path toward autonomous software engineering is currently bottlenecked by a severe deficit of diverse, large-scale trajectory data. We address this by introducing \ourdataset, an expansive dataset of 207,489 agentic trajectories spanning nine programming languages (Python, Go, TS, JS, Rust, Java, PHP, C, C++). Sourced from 20,000 real-world PRs via OpenHands and SWE-agent harnesses, the dataset utilizes a hybrid-reasoning synthesis: Minimax-M2.5 generates trajectories with explicit "thinking" processes, while Qwen3.5-122B provides high-quality "non-thinking" traces. Filtered for permissive licenses (MIT, Apache, BSD) from SWE-rebench-V2, this data facilitates the training of models capable of long-horizon reasoning. We validate the dataset by fine-tuning the Qwen3-30B-A3B series (Thinking, Instruct, and Coder). The best performing model achieves resolve rates of 61.7% on SWE-bench Verified, 57.1% on SWE-bench Multilingual, and 36.8% on SWE-bench Pro. These results establish Open-SWE-Traces as a premier resource for distilling human-level software engineering capabilities into efficient, open-source agentic LLMs.

18.
arXiv (CS.AI) 2026-06-18

The More the Merrier: Combining Properties for ABox Abduction under Repair Semantics for ELbot

arXiv:2606.19197v1 Announce Type: cross Abstract: Abduction is a central approach to explain missing entailments from a knowledge base by providing a hypothesis, that would, if added to the knowledge base, make the missing entailment become true. Abduction under repair semantics has recently been investigated in detail, where several desirable properties and optimality criteria were considered, such as signature-restrictions and minimality in size and of introduced conflicts. Naturally, hypotheses that satisfy more than one of these properties or combine a property with an optimality criterion would be even more desirable for applications. So far, such hypotheses have not been investigated in the literature. In the present paper, we consider the ABox abduction problem for hypotheses satisfying more than one property or additional optimality criteria, for EL_bot under brave and AR semantics. Our main observation is that often requiring additional properties for hypotheses does not lead to an increase of complexity.

19.
arXiv (quant-ph) 2026-06-16

Influence of the Electron's Anomalous Magnetic Dipole Moment on High-Atomic-Number Atoms

arXiv:2606.15995v1 Announce Type: new Abstract: Super-heavy atoms ($Z > 100$) are usually studied in the context of the so-called ``Quantum Electrodynamics of Strong Fields''. In this theory the problem of the singularity in the electron energy whenever $Z > 137$ is overcome. This is done by considering the finite size of the nucleus and leads to interesting phenomena, such as the spontaneous production of positrons. Here, we show that taking into account the contribution from the Anomalous Magnetic Dipole Moment of the electron (by means of an effective theory), within a point-nucleus model, is a sufficient condition to obtain regular wave functions and physically acceptable energy values for $Z > 137$.

20.
arXiv (CS.CL) 2026-06-12

Operads for compositional reasoning in LLMs

Question decomposition, i.e. breaking a complex query into simpler sub-queries whose answers are composed to produce a final answer, is a widely used strategy for improving LLM reasoning, yet it currently lacks a rigorous mathematical foundation. In this paper, we propose operads, mathematical structures that model many-in, one-out operations and compositions thereof, as a natural framework for describing question decomposition. We define the questions operad $Q$, in which operations correspond to question templates and composition corresponds to substitution of sub-answers, and show how QA models can be interpreted as algebras over $Q$. Beyond reframing existing practice, this operadic perspective points toward new methods, in particular a notion of operadic consistency, which measures whether a QA model's answers agree across the partial collapses of a question decomposition tree. Empirical evaluation of operadic consistency is reported in our companion paper (Bottman, Liu, and Richardson, 2026), which finds it strongly correlated with accuracy across twelve LLMs and four multi-hop QA datasets and outperforming standard temperature-based self-consistency baselines. We argue that operads are the natural mathematical home for question decomposition, and that invariants such as operadic consistency open new directions for analyzing and improving the reliability of multi-step reasoning.

21.
arXiv (CS.CL) 2026-06-15

AdaSR: Adaptive Streaming Reasoning with Hierarchical Relative Policy Optimization

Large reasoning models typically follow a read-then-think paradigm: they observe the complete input, reason over a static context, and then produce the answer. Yet many real-world scenarios are inherently dynamic, such as audio and video stream, where information arrives as a continuous stream and models must reason, update, and respond under partial observations. Recent streaming reasoning methods allow models to think while reading, but they largely rely on supervised imitation of pre-constructed trajectories, which limits their flexibility. In this paper, we propose AdaSR, an adaptive streaming reasoning framework that enables models to reason during input streaming and perform final deliberation once the stream is complete, learning when to think, and how much computation to allocate across different stages. To optimize this hierarchical reasoning process, we introduce Hierarchical Relative Policy Optimization (HRPO), which decomposes policy optimization into streaming reasoning and deep reasoning phases, providing more fine-grained advantage assignment instead of uniformly distributing a single sequence-level advantage over all tokens. HRPO integrates format, accuracy, and adaptive thinking rewards to enforce valid reasoning protocols, preserve final task performance, and encourage latency-aware computation allocation. Experiments show that AdaSR achieves a better balance among reasoning accuracy, computational efficiency, and streaming latency compared with supervised fine-tuning baseline. We release our code at https://github.com/EIT-NLP/StreamingLLM/tree/main/AdaSR.

22.
arXiv (CS.LG) 2026-06-19

On the Variance of Temporal Difference Learning and its Reduction Using Control Variates

arXiv:2606.20357v1 Announce Type: new Abstract: We analyze the variance of temporal difference (TD) learning using the phased setting with tabular representation, and show that one of the mechanisms behind its ability to reduce variance is by effectively aggregating over a larger number of independent trajectories. Based on this insight, we demonstrate that (1) the variance of TD is asymptotically bounded from above by Monte Carlo (MC) estimators, and (2) shorter horizon updates incurs less variance for a fixed number of samples. Beyond TD, we show that Direct Advantage Estimation (DAE), a method for estimating the advantage function, can be seen as a type of regression-adjusted control variate, which achieves a tighter bound on the variance compared to TD in the large-sample limit. Finally, we numerically illustrate the behaviors of these estimators with carefully designed environments.

23.
medRxiv (Medicine) 2026-06-22

Reform of the intermediate level of the health system in the Democratic Republic of the Congo: Adaptations and limits in the stabilization of the personnel of the Provincial Health Division: A cohort study.

Background: Human resources are one of the pillars of health systems. Since the World Health Organization's report on human resources issues, several countries have integrated this component into the various reforms aimed at strengthening their health systems. This study aims to explore the effects of reforming the intermediate level of a health system operating in a fragile state context. Methodology Our study was conducted in the Democratic Republic of Congo (DRC). It was a cohort study of the staff of the 14 Provincial Health Divisions (PHD) out of the 26 existing in the DRC. We established a database of the staff of these 14 PHD from 2016, just after the implementation of the intermediate level reform and the allocation of this staff by the Ministry of Health. We did a recall in 2021, in each of these PHD to survey this staff through a structured questionnaire and supplemented by the files of the agents available in each PHD. Sociodemographic, economic and academic variables were collected and analyzed. Data were entered into an Excel 2016 database and processed with SPSS software version 25. The chi-square test was used for comparison of proportions with a statistical significance level of p < 0.05. Risk ratios ratios (RR) and their 95% confidence intervals were calculated as measures of association. The error threshold was set at 5%. Results A total of 657 agents with an average age of 45.2 years had been identified in 2016 at the start of the survey and in 2021, 118 or 18% of them were no longer part of the PHD agents. Among the causes of absence noted: 48% of agents placed on leave, 16% promoted to other functions within the health system, 16% desertion and dismissal and 11% cases of death. 19.8% of absentees are executives, 19.5% men against 10.3% women; 22.3% of absentees in unstable provinces against 16.6% in stable ones. The factors associated with the absence of agents in the PHD remain the reaching of retirement age [RR (95% CI) = 5.5 (1.2-24.9) ]and male agents [RR (95% CI) = 3.2 (1.3-7.9)]. Among the agents who remained, 92% kept their initial position, 6% were subject to an internal permutation accompanied by a promotion. The factors associated with the stability of human resources at the level of the Provincial Health Division are: female gender, manager with experience or seniority > 5 years, Age > 35 years, Stable province, Presence of a partner bonus. Conclusion Even in a crisis and fragile context, health system reform is possible. It is possible to organize staff recruitment through a selection process independent of the political authorities of the Ministry of Health and supported by the technical services of the Ministry and partners . Experience and the presence of a financial bonus are motivating factors for staff stability. The involvement of Technical and Financial Support Partners in the recruitment process helped the Ministry of Health to minimize political influence in the recruitment of middle-level executives.

24.
arXiv (CS.AI) 2026-06-24

RIFT-Bench: Dynamic Red-teaming For Agentic AI Systems

arXiv:2606.23927v1 Announce Type: new Abstract: Agentic AI systems powered by large language models (LLMs) are rapidly evolving into autonomous decision-making systems, exposing attack vectors beyond those of traditional LLM vulnerabilities. Existing security evaluations are often tied to specific implementations or domains, limiting unified comparison across heterogeneous systems. To address this gap, we introduce RIFT-Bench, a graph representation-driven methodology for dynamic red-teaming that enables unified evaluations across diverse agentic architectures. Building on a novel hierarchical representation, RIFT-Bench operates in two automated phases: Discovery, which extracts system structure, and Scanning, which deploys adaptive adversarial attacks and produces a comprehensive evaluation report. It evaluates the examined system itself, leveraging a broad set of dynamically adaptable adversarial probes across diverse attack vectors and objectives. We demonstrate the effectiveness of the proposed evaluation pipeline across 45 agentic systems spanning a diverse range of implementations, showing that the approach generalizes effectively to heterogeneous agentic architectures. Beyond systems and attacks, RIFT-Bench also supports direct evaluation of mitigation strategies. These key capabilities make RIFT-Bench a scalable foundation for security evaluation of agentic AI systems.

25.
arXiv (CS.AI) 2026-06-25

The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems

arXiv:2606.26057v1 Announce Type: new Abstract: AI agents are granted access to tools, APIs, and other infrastructure, making them active principals in those systems. The dominant approach places controls inside the agent's own runtime: system prompts, output filters, and guardrail libraries. Any control in the agent's address space is reachable by inputs that influence it; this generalizes to any AI system with sufficient reach into its own runtime, a class we term escapable AI systems. We identify four properties that an authorization mechanism must satisfy for architectural control rather than for cooperative requests: process separation, pre-action enforcement on a structurally only path, fail-closed at both the request and system levels, and externalized signed evidence verifiable outside the controlled system's trust boundary. We position this layer as execution-time AI alignment, complementing training-time alignment (RLHF, Constitutional AI) and inference-time alignment. We present the Unfireable Safety Kernel, a Rust reference implementation realizing all four. Its fail-closed invariant is machine-checked at two levels: an SMT theorem (Z3) and an exhaustive bounded-model-checking proof of the production decision function (Kani, 4/4 harnesses). A Python-to-Rust migration was gated on byte-equivalence (1000/1000 fixtures; 17/17 adversarial classes). We evaluate the kernel governing a live, escapable AI system, a deterministic, self-improving world model, against an escape-seeking adversary driving its real self-modification seam: across 1,000 self-modifications, all 704 attempts on the safety-critical core are refused, with no escape; a further 300, under the operator kill switch, are also refused. A separate campaign of 6,240 authorization round-trips had no successful bypass. Against 3 contemporary systems claiming the agent control plane, the agent invokes control; here, it lacks that choice.