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 (quant-ph) 2026-06-16

No Universal Purification in Quantum Mechanics

arXiv:2509.21111v2 Announce Type: replace Abstract: Many central tasks in fundamental physics and quantum information processing are possible only insofar as mixed quantum states can be made purer. In this work, we prove that the linearity and positivity of quantum mechanics impose general restrictions on quantum purification, unveiling a new fundamental principle of quantum information processing. We first establish that no quantum operation can transform a finite number of copies of an unknown quantum state or channel into an exactly pure output that depends non-trivially on the input, thereby ruling out an important form of universal purification in both static and dynamical settings. Building on this, we show that, upon relaxing the requirement of exact purity, one can establish quantitative sample-complexity lower bounds for approximate purification that hold for arbitrary physically allowed strategies, whose scaling matches the performance of purification-related tasks across several different areas of quantum information processing. Moreover, this lower bound leads to a generalized standard quantum limit for learning arbitrary functions of a quantum state, greatly extending earlier results based on quantum Fisher information and revealing a deep connection between purification and quantum learning. Extending this principle to other important settings, we establish, for the first time, an exponential sample-complexity lower bound for approximate pure dilation state preparation and a no-go theorem for approximate bosonic Gaussian state purification with passive Gaussian operations, establishing much more stringent limitations under practical operational constraints.

03.
bioRxiv (Bioinfo) 2026-06-15

SMLMFlow: Improving Structural Resolution in Single Molecule Localization Microscopy with Flow Matching

While Single Molecule Localization Microscopy (SMLM) aims to generate precise coordinates of molecular targets in cells, the resulting point clouds are inherently blurred by additive noise sources across the experimental, imaging, and processing workflow. This blurring often limits SMLM's ability to accurately quantify complex assembled structures required to address biological issues, despite reported localization precision down to a couple of nanometers. Here, we present SMLMFlow, a machine learning framework for improving structural resolution in SMLM datasets that combines a graph neural network and a hierarchical transformer with flow matching. We show that SMLMFlow improves structural resolution and downstream quantification across different structures, including filaments and protein nano-clusters, and generalizes to new unseen photophysics models.

04.
arXiv (CS.CL) 2026-06-17

FeedEval: Pedagogically Aligned Evaluation of LLM-Generated Essay Feedback

Going beyond the prediction of numerical scores, recent research in automated essay scoring has increasingly emphasized the generation of high-quality feedback that provides justification and actionable guidance. To mitigate the high cost of expert annotation, prior work has commonly relied on LLM-generated feedback to train essay assessment models. However, such feedback is often incorporated without explicit quality validation, resulting in the propagation of noise in downstream applications. To address this limitation, we propose FeedEval, an LLM-based framework for evaluating LLM-generated essay feedback along three pedagogically grounded dimensions: specificity, helpfulness, and validity. FeedEval employs dimension-specialized LLM evaluators trained on datasets curated in this study to assess multiple feedback candidates and select high-quality feedback for downstream use. Experiments on the ASAP++ benchmark show that FeedEval closely aligns with human expert judgments and that essay scoring models trained with FeedEval-filtered high-quality feedback achieve superior scoring performance. Furthermore, revision experiments using small LLMs show that the high-quality feedback identified by FeedEval leads to more effective essay revisions. We release our code and curated datasets at: https://github.com/BBeeChu/FeedEval.git.

05.
arXiv (CS.AI) 2026-06-19

Exit-and-Join Dynamics for Decentralized Coalition Formation

作者:

arXiv:2606.19683v1 Announce Type: new Abstract: This paper studies coalition formation as a decentralized dynamical process driven by unilateral exit-and-join decisions. Agents evaluate local moves using the Aumann-Dreze value, so payoffs are computed within the agent's current coalition rather than through a globally negotiated coalition structure. The resulting model links cooperative payoff allocation with noncooperative best-response behavior: a terminal partition is precisely a coalition structure with no admissible, individually profitable exit-and-join deviation. We establish equilibrium characterizations, identify conditions under which the dynamics admit scalar Lyapunov or exact-potential representations, and analyze how switching and acceptance costs shape local stability. Numerical experiments test finite-time stabilization, cost sensitivity, and a special convex-game benchmark.

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

NARRAS: Edge-Triggered Distributed Inference for CSI-Based Localization in Vehicular IoT Networks

arXiv:2606.11914v1 Announce Type: cross Abstract: CSI-based localization with spatially distributed antenna arrays exposes a basic resource trade-off. Each array can provide a rich view of the channel, but forwarding observations from all arrays to a fusion center is wasteful when only a few carry useful information, and the shared uplink supports only a limited number of simultaneous transmissions. We let each array decide locally whether its current observation is worth reporting, subject to a budget on the average number of active transmitters. We refer to this abstraction as Edge-Triggered Distributed Inference (ETDI). It captures a broader class of task-oriented communication problems where resource-constrained devices share an access channel for a common inference task. We instantiate ETDI for CSI-based localization, a common scenario in vehicular IoT networks. Spatially distributed remote antenna arrays (RAAs) encode local channel state information (CSI) from user equipment (UE) transmissions into latent features, and the fusion center estimates the UE position from the subset of reported features. We propose NARRAS, a decentralized reporting policy in which each RAA combines a recurrent summary of its recent observations with a memory of the last latent it transmitted. Training controls an explicit activity budget through differentiable activity penalties and validation-calibrated deterministic thresholds, and uses channel-chart regularization to shape the latent geometry. Experiments show that, at comparable uplink activity, NARRAS improves localization accuracy over learned and heuristic sparse-reporting strategies, while dense full-report models remain useful budget-free references. In low-activity regimes, chart regularization further reduces high-percentile localization errors, suggesting that geometry-aware latent representations are more robust under sparse reporting.

07.
arXiv (CS.CV) 2026-06-12

Towards Effective Waste Segmentation for Automated Waste Recycling in Cluttered Background

Rapid expansion of urban areas and population growth is causing an immense increase in waste production, which demands the need for efficient and automated waste management. In this scenario, automated waste recycling (AWR) using deep learning methods can assist humans in optimal waste management. Recent deep learning approaches for AWR provide promising waste segmentation performance, however, these methods rely on large backbone networks that are inefficient for AWR systems and suffer from performance deterioration in cluttered scenes. To this end, an optimal waste segmentation network is introduced which effectively utilizes the spatial domain to capture localized structural dependencies and the spectral domain to efficiently extract global contextual relationships. This cascaded design allows the network to progressively leverage both local and global representations across complementary domains to highlight the semantic information necessary for effective segmentation of various waste objects. Furthermore, auxiliary feature enhancement module (AFEM) is introduced to enhance the target objects' boundaries and blob amplification for better segmentation in cluttered scenarios. Extensive experimentation on ZeroWaste-aug, ZeroWaste-f and SpectralWaste datasets reveals the merits of the proposed method.

09.
arXiv (CS.AI) 2026-06-18

R2BC: Multi-Agent Imitation Learning from Single-Agent Demonstrations

arXiv:2510.18085v2 Announce Type: replace-cross Abstract: Imitation Learning (IL) is a natural way for humans to teach robots, particularly when high-quality demonstrations are easy to obtain. While IL has been widely applied to single-robot settings, relatively few studies have addressed the extension of these methods to multi-agent systems, especially in settings where a single human must provide demonstrations to a team of collaborating robots. In this paper, we introduce and study Round-Robin Behavior Cloning (R2BC), a method that enables a single human operator to effectively train multi-robot systems through sequential, single-agent demonstrations. Our approach allows the human to teleoperate one agent at a time and incrementally teach multi-agent behavior to the entire system, without requiring demonstrations in the joint multi-agent action space. We show that R2BC methods match, and in some cases surpass, the performance of an oracle behavior cloning approach trained on privileged synchronized demonstrations across four multi-agent simulated tasks. Finally, we deploy R2BC on two physical robot tasks trained using real human demonstrations.

10.
arXiv (CS.CL) 2026-06-16

Building Customer Support AI Agents at 100M-User Scale: An Evaluation-Driven Framework

The rapid rise in LLM capabilities has made AI agents increasingly viable across a broad range of tasks. Among the most promising applications is building production-ready customer-facing agents, a challenge that demands coordinated excellence in evaluation methodology, context engineering, training, and online measurement. Yet these critical pillars are typically developed in isolation, creating blind spots that only surface after deployment. In this paper, we present a unified framework that bridges offline development with online impact for customer support AI agents at Nubank, a company with 100M+ users. Our approach integrates several key components: (1) structured context engineering tailored to customer support agents, (2) systematic human-in-the-loop prompt iteration, (3) rigorous LLM judge evaluation with measured inter-rater agreement and GEPA optimization for consistency, and (4) ideation-to-production validation. A central insight is that evaluation-pipeline quality directly determines iteration velocity. We present results from five production deployments spanning distinct domains: card delivery, debt management, credit-limit support, card management, and product explanation. These deployments deliver consistent customer-satisfaction gains while substantially accelerating iteration. In our card-delivery deployment, large-scale A/B testing yields a 37 percentage-point improvement in AI transactional Net Promoter Score and a 29 percentage-point gain in self-service rate over prior agent variants, alongside a strong correlation between offline simulation metrics and online outcomes, demonstrating that eval-driven development reliably predicts production impact. On most use cases, AI satisfaction reaches within a few percentage points of expert human agents.

11.
arXiv (CS.AI) 2026-06-19

A Deep Generative Model for Resting-State EEG Synthesis and Transferable Representation Learning

arXiv:2503.02636v5 Announce Type: replace-cross Abstract: Resting-state EEG provides a non-invasive view of spontaneous brain activity, but extracting meaningful patterns is often limited by scarce high-quality data and reliance on manually engineered features. Generative adversarial networks (GANs) can synthesize neural signals and learn transferable representations directly from raw data, a dual capability that remains underexplored in EEG research. Here, we introduce REST-GAN, a GAN-based framework for resting-state EEG that combines adversarial training with an auxiliary self-supervised reconstruction objective to support signal synthesis and unsupervised feature extraction. Although trained only on raw time-domain signals, without explicit frequency-domain or sensor-topographic supervision, the generated time series reproduced key temporal, spectral, and connectivity properties of real EEG. In band-power feature space, generated samples showed high precision and recall across eyes-open and eyes-closed conditions (EO: 0.91/0.67; EC: 0.87/0.65), while group-average spectral coherence matrices showed low mean absolute differences from real data across frequency bands (~0.01-0.03). The representations learned by the model's critic transferred to independent resting-state demographic classification tasks, outperforming models trained directly on raw EEG and showing competitive performance relative to a recent EEG foundation model, while requiring substantially less training data and computational resources. These findings highlight a computationally efficient, architecture-driven strategy in which generative models serve not only as EEG signal generators, but also as unsupervised feature extractors. This approach may support more data-efficient EEG analysis while reducing reliance on manual feature engineering. The implementation code for REST-GAN is available at: https://github.com/Yeganehfrh/REST-GAN.

12.
arXiv (math.PR) 2026-06-11

The $K$-th nearest neighbor random walk on a Poisson point process gets trapped

arXiv:2606.11271v1 Announce Type: new Abstract: The $K$-th nearest neighbor random walk $(X_n)_{n \geq 0}$ on a homogeneous Poisson point process $\chi$ on $\R^d$ ($d\geq 1$), starts at the origin and at each step picks its next Poisson point among its closest neighbors according to i.i.d. labels having the same distribution as $K$. Our main result (Theorem 1) states that the number of Poisson points visited by $(X_n)_{n \geq 0}$ admits an exponential decay whenever the random variable $K$ has a bounded support (BS). In particular, the $K$-th nearest neighbor random walk visits finitely many Poisson points if and only if $K$ satisfies Assumption (BS). To prove it, we introduce the key notion of pioneer point which allows us to deal with the region of $\R^d$ already explored by $(X_n)_{n \geq 0}$. Still under Assumption (BS), we also prove an exponential decay for the Euclidean length of the trajectory performed by $(X_n)_{n \geq 0}$ (Theorem 2). Finally, and quite surprisingly, we exhibit an example of label distribution with bounded support for which the $K$-th nearest neighbor random walk discovers new Poisson points after a number of steps whose tail distribution is at least polynomial (Theorem 3).

13.
arXiv (CS.CL) 2026-06-16

From ASR to ASP: Evaluating Prompt Attack Vulnerabilities Against Open-Source LLMs

Recent studies demonstrate that Large Language Models (LLMs) are vulnerable to attacks that generate harmful or sensitive outputs. As open-source LLMs are increasingly adopted in high-impact applications such as finance, law, and healthcare, systematically investigating their security risks is becoming increasingly important towards trustworthy LLM era. This paper comprehensively studies effective prompt injection attacks against 14 widely used open-source and three closed-source LLMs on five attack benchmarks. Moreover, existing evaluation metrics mostly only consider the attack success rate, overlooking uncertainty in model responses. Our proposed Attack Success Probability (ASP) additionally captures uncertain behaviors for evaluation, where the model may initially refuse a harmful request but subsequently provide harmful guidance or vice versa, reflecting inconsistency and ambiguity in attack feasibility. By systematically analyzing the effectiveness of prompt injection attacks, we propose a straightforward and effective hypnotism attack; results show that this attack causes aligned language models, including Stablelm2, Mistral, Openchat, and Vicuna, to generate objectionable behaviors, achieving around 90% ASP. They also indicate that ignore prefix attacks can break all 14 open-source LLMs, achieving over 60% ASP on a multi-categorical dataset. We find that moderately well-known LLMs exhibit higher vulnerability to prompt injection attacks, highlighting the need to raise public awareness and prioritize efficient mitigation strategies.

14.
medRxiv (Medicine) 2026-06-10

Resolving Diagnostic Discordance in Group 2 Pulmonary Hypertension Through Staged Physiologic Testing: Insights From PVDOMICS

Background World Symposium on Pulmonary Hypertension (WSPH) Group 2 pulmonary hypertension (PH) is a clinically integrated phenotype attributed to left heart disease, whereas pre- versus post-capillary classification is operationalized primarily by pulmonary capillary wedge pressure (PCWP). Although current recommendations emphasize contextual interpretation and provocative testing for intermediate PCWP values, the relationship between PCWP-based classification and underlying phenotype has not been systematically evaluated. We aim to quantify phenotype-hemodynamic discordance across the PCWP spectrum and evaluate a staged physiology-guided framework incorporating inhaled nitric oxide (iNO), ventricular geometry, and provocative testing. Methods We studied 1,032 participants from the NHLBI-sponsored PVDOMICS cohort with multidisciplinary adjudicated phenotypes integrating clinical, imaging, physiologic, and hemodynamic data. Stage-specific PCWP thresholds classified pre- versus post-capillary physiology at rest, during iNO, and during provocation (fluid challenge or invasive cardiopulmonary exercise testing [iCPET]). Echocardiographic right ventricular-to-left ventricular (RV/LV) ratio was evaluated as a marker of ventricular interdependence. Restricted cubic spline and staged concordance analyses defined certainty-based PCWP ranges and incremental diagnostic yield. Results Adjudicated Group 2 phenotype was present in 37.0% of participants. Resting PCWP demonstrated good discrimination (AUC 0.86), but substantial bidirectional phenotype-hemodynamic discordance persisted across intermediate PCWP ranges. At a resting PCWP of 12 mmHg, 25% of participants classified as pre-capillary had adjudicated Group 2 PH, whereas at 18 mmHg, 35% classified as post-capillary remained discordant non-Group 2. Concordance did not approach 90% until PCWP values were 24 mmHg. Dynamic testing incrementally improved concordance within these overlap zones. Nearly half of adjudicated Group 2 PH participants (46.5%) were not identified by resting PCWP alone; incorporation of iNO and provocative testing increased cumulative Group 2 identification by 63.4% and improved sensitivity from 79.9% to 83.7%. Model discrimination improved from an AUC of 0.863 to 0.908 (likelihood-ratio P

15.
arXiv (CS.AI) 2026-06-16

MA-SBI: Misspecification-Aware Simulation-Based Inference via Side-Channel Guidance

arXiv:2606.16923v1 Announce Type: new Abstract: Simulation-based inference (SBI) of latent parameters is often hindered by simulator misspecification, the mismatch between simulated and real-world observations caused by inherent modeling simplifications. RoPE, the recent state-of-the-art for robust SBI, addresses this through optimal transport between learned representations of real and simulated observations, but requires ground-truth parameter calibration pairs that are typically unavailable in the very settings where SBI is needed. What practitioners do have is unstructured side-information such as regime labels, instruction text, and policy bulletins. We propose Misspecification-Aware Simulation-Based Inference (MA-SBI), a calibration-free framework that turns this side-channel into a posterior correction. A learned corrector maps side-channel text to an observation-space shift applied before any pre-trained amortized posterior, requiring no retraining and no parameter ground-truth. Our main theorem bounds achievable bias reduction by the mutual information between misspecification and side-channel, with a non-vacuous constant that extends to all sub-Gaussian noise via Donsker-Varadhan. On hide-the-calibration benchmarks, MA-SBI with text alone matches the oracle posterior across 10 seeds and two backbones (TOST equivalence), while RoPE given more data does not. The two approaches are complementary: where misspecification is structural and recoverable from parameter pairs, RoPE dominates, as the theory predicts. A stochastic variant improves posterior-predictive log-likelihood on real COVID and OxCGRT epidemiological data, and correctly leaves the posterior unchanged on a well-specified cognitive-science corpus.

16.
arXiv (CS.AI) 2026-06-12

Beyond Runtime Enforcement: Shield Synthesis as Defensibility Analysis for Adversarial Networks

arXiv:2606.13621v1 Announce Type: new Abstract: Shielded reinforcement learning is typically presented as a runtime safety mechanism that compiles temporal-logic specifications into automata restricting an agent's actions. We argue this is the wrong product. The same automata-theoretic machinery – specification compilation, product game construction, attractor computation, and winning-region extraction – is better read as a design-time analytical instrument whose outputs are structural insights about a system rather than runtime constraints on a deployed agent. We instantiate this through a constrained two-player safety game for network defense. The two specifications are enforced asymmetrically: the defender specification defines the unsafe region of the game, whereas the attacker specification restricts the adversary's legal actions during attractor computation. Solving the game yields a defensibility verdict – a formal certificate that a topology-specification pair is or is not defensible – with the associated winning region and shield. Beyond the binary verdict, we derive topology-level metrics from the attractor structure and combine them with post-convergence behavior from shield-constrained adversarial multi-agent reinforcement learning. Together these form a defensibility fingerprint capturing both a network's formal safety properties and its operational behavior under adaptive play. A what-if analysis shows that formal defensibility and operational effectiveness capture distinct aspects of security: small architectural changes can produce large shifts in operational outcomes while leaving formal safety margins nearly unchanged. Shield synthesis is thus most valuable not as a deployment mechanism for safe agents, but as a framework for answering architectural questions about whether, where, and how a system can be defended. The defensibility verdict is the output, not the safe policy.

17.
bioRxiv (Bioinfo) 2026-06-11

ANCHOR: haplotype-aware allelic and isoform inference from single-cell long-read RNA sequencing with de novo variant calling

Long-read RNA sequencing enables haplotype- and isoform-resolved allelic analysis of transcriptomes, yet extending this capability to single cells and distinct cell types remains computationally challenging due to sparse coverage, sequencing errors, incomplete variant information, and reference-biased transcript assignment. Here we present ANCHOR, a haplotype-aware framework for single-cell long-read RNA sequencing that performs de novo expressed-variant discovery, molecule-level haplotype assignment and isoform-resolved allelic quantification. ANCHOR combines a signed-graph variant caller, pair hidden Markov modelling and beta-binomial UMI aggregation to infer parental allele counts for genes and splice-resolved isoforms, without requiring a pre-existing phased genotype or deep learning. In human single-cell long-read RNA benchmarks, ANCHOR improved variant-calling performance over tested long-read RNA callers at single-cell and low-to-moderate coverage, and its beta-binomial model reduced depth-driven false positives in allele-specific expression testing. Applied to newly generated single-cell long-read RNA-seq data from reciprocal mouse crosses during gastrulation, ANCHOR resolved cell-type- and isoform-specific parent-of-origin imprinting and identified an antagonistic maternally biased Sgce isoform. ANCHOR provides a general framework for allele- and isoform-resolved analysis of diploid single-cell long-read transcriptomes.

18.
arXiv (quant-ph) 2026-06-17

From Period Finding to Lattice Sampling: Experimental Insights into Shor's and Regev's Factoring Algorithms

arXiv:2606.17647v1 Announce Type: new Abstract: Quantum algorithms for integer factorization represent one of the most prominent applications of quantum computation, with far-reaching implications for modern cryptography. While Shor's algorithm provides a polynomial-time solution in the ideal quantum model, its practical implementation is severely constrained by the limitations of current noisy intermediate-scale quantum (NISQ) hardware. These constraints have motivated the exploration of alternative factoring algorithms with different structural and resource trade-offs. In this work, we present an experimental study of Regev's quantum factoring algorithm, implemented on real quantum hardware, and compare its behavior with that of Shor's algorithm under analogous conditions. Focusing on the case N = 15, we execute both algorithms on the QMIO quantum computer at the Centro de Supercomputacion de Galicia (CESGA) and contrast the results with one of IBM's open-access quantum computers and ideal simulations. This parallel execution enables a low-level comparison of the two algorithms, highlighting how their respective quantum implementations interact with hardware noise, limited circuit depth, and finite sampling. Our analysis emphasizes the different ways in which Shor's and Regev's algorithms encode arithmetic structure into quantum states through Fourier sampling in one and higher dimensions, respectively, and how these differences manifest in experimental outcomes. Although neither algorithm demonstrates a practical advantage in the small N regime, the results provide insight into their relative robustness and failure modes on contemporary quantum devices. This study illustrates the value of experimental benchmarking of alternative quantum factoring algorithms as a means of understanding the practical implications of algorithmic design choices in the NISQ era.

19.
arXiv (CS.CL) 2026-06-11

Geometric Metrics and LLMs: What They Measure and When They Work

We present a systematic stress-test of geometric metrics for LLM evaluation. Rank-based geometric properties of internal representations have shown promise as reference-free quality signals, but the conditions under which they are reliable remain unclear. We evaluate eight commonly-used metrics: intrinsic-dimensionality estimators, spectral norms, and related quantities across six tester models (0.5-8B) and eight generators on contrasting tasks, separating genuine geometric signal from text-length effects and from what standard text statistics already capture. Three findings emerge. First, some metrics (notably Schatten Norm and MOM) mainly reflect output length, and their apparent discriminative power collapses once length is controlled. Second, geometric metrics add modest but real information beyond text statistics: combined with them, a classifier reaches 78% accuracy on 6-way generator identification versus 69% for text statistics alone. Third, rather than tracking a general notion of text quality, the metrics demonstrate only moderate association between the intrinsic-dimensionality and lexical diversity (RTTR). We give use-case-specific recommendations and identify failure detection as the most promising near-term application.

20.
arXiv (CS.LG) 2026-06-16

TCHG: Tri-Trust Conditioned Heterogeneous Graph Learning for Reliable Dynamic Trust Prediction

arXiv:2606.16611v1 Announce Type: new Abstract: Trust prediction infers latent user-user trust relations and provides important support for social recommendation, fake-review and manipulation detection, and risk identification. Graph neural networks have become a prominent approach to trust prediction because of their ability to learn network structures and complex trust dependencies. However, existing methods often rely on a unified representation of trust signals and do not disentangle heterogeneous trust evidence into separate evidence channels, failing to exploit the distinct roles that different evidence channels should play during trust modeling. To address this gap, this paper argues that trust evidence should not be treated as an undifferentiated input, but should be decomposed and used as functional control factors over graph propagation. We propose TCHG, a tri-trust conditioned heterogeneous graph learning framework that decomposes trust evidence into three channels and assigns them distinct functional roles in propagation: entity reliability governs message admission, interaction-behavior reliability modulates propagation strength, and contextual trust adjusts the propagation mode through context-conditioned operator selection. Since the three evidence channels evolve at different temporal scales, TCHG maintains independent temporal states with non-uniform decay rates to prevent rapidly changing contextual signals from overwriting slowly accumulated entity reliability. It further predicts trust probability and calibrates the output probability, improving predictive confidence under sparse or conflicting evidence. Extensive experiments on multiple public trust datasets show that TCHG achieves effective and reliable trust prediction compared with representative trust prediction and heterogeneous graph baselines.

21.
arXiv (CS.CL) 2026-06-17

From Trainee to Trainer: LLM-Designed Training Environment for RL with Multi-Agent Reasoning

Reinforcement learning pipelines for Large Language Model (LLM) training often rely on manually redesigned environments between stages, requiring practitioners to heuristically infer which configuration will best improve the current policy. To automate this process, we propose the LLM-as-Environment-Engineer framework in which the current policy model analyzes failure trajectories together with contextual information and proposes modifications to the next-stage training environment configuration. We also introduce MAPF-FrozenLake, a controllable testbed whose generator exposes multi-dimensional environment configurations, making it suitable for studying and benchmarking environment redesign. On this testbed, we condition the environment engineer on structured summaries of policy behavior, failure cases, and environment statistics, from which it produces the configuration for the next training stage. With Qwen3-4B as the backbone, our framework achieves the strongest aggregate performance on our benchmarks, outperforming larger proprietary LLMs (e.g., GPT, Gemini) and fixed-environment training baselines. We further analyze which forms of context are most effective, finding that successful environment updates rely on failure evidence and preserve configurations that already work. Interestingly, the current RL checkpoint serves as a better environment engineer than the original base model, suggesting that policy learning improves the model's ability to diagnose its remaining weaknesses.

22.
arXiv (CS.AI) 2026-06-16

Reward Hacking in Language Model Agents: Revisiting AI Safety Gridworlds

arXiv:2606.15385v1 Announce Type: new Abstract: Reward hacking, where AI systems exploit misspecified objectives to achieve high reward without satisfying intended goals, remains a central challenge in AI safety. Yet most known instances have been discovered post hoc in frontier systems where controlled study is impractical. We adapt the AI Safety Gridworlds framework into a text-based evaluation suite that reformulates classic reinforcement learning safety tasks for language-based agents. Across frontier and mid-scale models, we find that specification gaming emerges zero-shot: models systematically achieve high observed reward while underperforming on hidden safety objectives, and even apparently safe behaviors can reflect misunderstanding rather than principled safety. Reinforcement learning does not correct these failures: direct reward optimization widens the gap between observed and hidden reward, as the model's initial competence causes it to lock into locally rewarding strategies before discovering safer alternatives. This pattern persists across model scales (1.5B–14B) and is not resolved by finer credit assignment, exploration prompts, or entropy regularization. Our results show that reward hacking arises naturally when optimizing proxy objectives with capable language model agents and resists standard mitigations, suggesting that proxy-reward failures in agentic settings may require approaches beyond standard exploration and credit-assignment fixes. To facilitate reproducibility, the code for this work is available at \href{https://github.com/asparius/verl-agent-safety}{our public repository}.

23.
arXiv (quant-ph) 2026-06-12

Block algebra for morphing circuits

作者:

arXiv:2606.12724v1 Announce Type: new Abstract: Morphing circuits are a new paradigm for quantum error correction that relaxes hardware requirements. We present four constructions for CNOT-based CSS morphing circuits with explicit qubit connectivity degrees. All four constructions are specified in block algebra notation, with entries in algebras generated by permutation matrices. The first three are obtained by rewriting existing surface- and color-code morphing circuits; the fourth is a new three-round construction modeled on the 6.6.6 color code. The surface-code construction recovers the morphing circuit of Ref. [ST25] for two-block group algebra codes. Numerical search then instantiates these permutation matrices using regular representations of finite groups. [ST25] M. H. Shaw and B. M. Terhal, Phys. Rev. Lett. 134(9), 090602 (2025).

24.
arXiv (CS.LG) 2026-06-16

Contrastive Regularization for Accent-Robust ASR

arXiv:2605.03297v2 Announce Type: replace-cross Abstract: ASR systems based on self-supervised acoustic pretraining and CTC fine-tuning achieve strong performance on native speech but remain sensitive to accent variability. We investigate supervised contrastive learning (SupCon) as a lightweight, accent-invariant auxiliary objective for CTC fine-tuning. An utterance-level contrastive loss regularizes encoder representations without architectural modification or explicit accent supervision. Experiments on the L2-ARCTIC benchmark show consistent WER reductions across multiple pretrained encoders, with up to 25 – 29\% relative reduction under unseen-accent evaluation. Analysis using within-transcript cosine dispersion indicates that SupCon promotes more compact and stable representation geometry under accent variability. Overall, SupCon provides an effective and model-agnostic regularization strategy for improving accent robustness.

25.
arXiv (CS.AI) 2026-06-16

NeuronFabric: A Software Reference Architecture for On-Chip Transformer Training with Local Adam

arXiv:2606.16440v1 Announce Type: cross Abstract: Publicly documented accelerator architectures generally separate training computation from optimizer-state updates or rely on external memory and host orchestration. This paper presents NeuronFabric, a software reference architecture intended for future FPGA and ASIC implementations of transformer training with local Adam updates. A complete C# prototype implements forward pass, backpropagation, and Adam optimization without external machine-learning frameworks. The goal is to validate numerical correctness and memory requirements before hardware implementation. The evaluated model is a 334K-parameter autoregressive transformer (d=88, H=4, f=264, L=4, vocab=256) trained on the Shakespeare corpus. The BF16W configuration achieves evaluation loss 1.5426 after 80K samples, compared with 1.5224 for an FP32 GPU reference, while producing coherent character-level text. The paper introduces BF16W, which stores weights in BF16 while retaining Adam optimizer moments in FP32. This reduces memory requirements for on-chip training. A 334K-parameter FP32 model with Adam moments requires approximately 4.0 MB, matching the BRAM capacity of a Xilinx ZCU102 device. The BF16W variant requires approximately 3.34 MB, leaving memory available for activation storage. We describe the vocabulary-budget constraint observed during earlier experiments, quantify BF16W memory savings, and outline FPGA training as the next stage of development. No FPGA measurements are included in this paper. This publication serves as a public architectural disclosure and software reference implementation for future FPGA and ASIC exploration of the NeuronFabric architecture.