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.
medRxiv (Medicine) 2026-06-12

Genome-wide association and multi-omics functional screens reveal the genetic architecture of foveal development

Foveal hypoplasia causes visual impairment across congenital eye disorders, yet the genetic programmes governing foveal development remain poorly characterised and no tractable model exists for foveal disease. In the first genome-wide association study of foveal hypoplasia, we identified 42 sentinel variants mapping to 54 effector genes supported by >= 2 criteria from a variant-to-gene framework incorporating developmental multi-omics. Disruption of six effector genes using mutant lines and CRISPR knockouts in the zebrafish high acuity zone recapitulates structural, functional, and ultrastructural hallmarks of foveal hypoplasia, establishing the first vertebrate disease model. Integration with human foetal single-cell and spatial transcriptomics reveals two temporal waves of effector gene expression and identifies Muller glia as critical mediators of foveal patterning. Phenome-wide analyses reveal foveal variants are pleiotropic with refractive, lenticular, and metabolic traits, connecting foveal development to anterior segment and systemic disease biology. These findings should inform mechanistic studies of macular disease.

02.
arXiv (CS.CV) 2026-06-11

Right Regions, Wrong Labels: Semantic Label Flips in Segmentation under Correlation Shift

The robustness of machine learning models can be compromised by spurious correlations between non-causal features in the input data and target labels. A common way to test for such correlations is to train on data where the label is strongly tied to some non-causal cue, then evaluate on examples where that tie no longer holds. This idea is well established for classification tasks, but for semantic segmentation the specific failure modes are not well understood. We show that a model may achieve reasonable overlap while assigning the wrong semantic label, swapping one plausible foreground class for another, even when object boundaries are largely correct. We focus on this semantic label-flip behaviour and quantify it with a simple diagnostic (Flip) that counts how often ground truth foreground pixels are assigned the wrong foreground identity while remaining predicted as foreground. In a setting where category and scene are correlated during training, increasing the correlation consistently widens the gap between common and rare test conditions and increases these within-object label swaps on counterfactual groups. Overall, our results motivate assessing segmentation robustness under distribution shift beyond overlap by decomposing foreground errors into correct pixels, flipped-identity pixels, and missed-to-background pixels. We also propose an entropy-based, ground truth label-free `flip-risk' score, which is computed from foreground identity uncertainty, and show that it can flag flip-prone cases at inference time. Code is available at https://github.com/acharaakshit/label-flips.

03.
Nature (Science) 2026-06-17

These ‘master’ proteins protect us from deadly mutations — and could inspire new drugs

作者:

Biology has clever ways to mask the effects of potentially harmful gene mutations. Scientists are investigating how this ‘buffering’ works — and how to exploit it. Biology has clever ways to mask the effects of potentially harmful gene mutations. Scientists are investigating how this ‘buffering’ works — and how to exploit it.

04.
arXiv (quant-ph) 2026-06-16

Neural network inverse design of nanophotonic scintillators

arXiv:2606.16309v1 Announce Type: cross Abstract: Scintillators are materials converting high-energy radiation into optical light, essential in a range of technologies such as medical imaging systems and security scanners. Scintillator development and optimization have remained limited by the complexity of their underlying physics, involving stochastic cascades of electron-electron, electron-phonon, and electron-photon interactions. Such processes are typically modeled by non-differentiable Monte Carlo simulations, limiting the applicability of machine learning for scintillator development. Here we present a physics-informed neural network that learns the scintillation cascade process from the incident high-energy particle to photon emission, substantially accelerating scintillator design and optimization. Combining this neural network with photonic simulations enables end-to-end differentiable optimization of the scintillator geometry. This allows us to optimize for arbitrary figures of merit, such as specific target emission patterns.. We demonstrate the concept and characterize it relative to previous approaches by inverse design of nanophotonic scintillators for X-ray imaging.

05.
medRxiv (Medicine) 2026-06-18

Rare Coding Variants Reveal Distinct Genetic Architectures Across Multidimensional Sleep Phenotypes

Sleep and circadian traits have been widely studied using common variants, but the contribution of rare coding variation remains unclear. We analyzed rare coding variants in 397,065 whole-exome sequenced UK Biobank participants across 36 sleep phenotypes from self-report, diagnoses, sleep medication use and accelerometry, and meta-analyzed results with 171,536 whole-genome sequenced All of Us participants of diverse ancestries, with replication in the Mass General Brigham Biobank (N = 31,275). We identified 260 genes associated with sleep phenotypes, including novel associations with sleep medication use in 29 genes and 24 out of 29 have not previously been reported with any sleep phenotypes. We observed modest but significant rare variant heritability and strong genetic correlations between sleep medication use, insomnia and fatigue. Temporal gene expression trajectory analyses indicate that genes associated with self-reported sleep traits show constant high prenatal expression, whereas genes linked to sleep medication phenotypes exhibit peak expression in the late prenatal period. These findings highlight distinct biological mechanisms captured by different measurement sources of sleep phenotypes and reveal rare-variant-informed targets for therapeutic discovery.

06.
arXiv (quant-ph) 2026-06-15

A new class of degenerate solutions to the massless Dirac equation and their potential applications in optical memories

arXiv:2606.14256v1 Announce Type: new Abstract: In this article, we present a novel class of degenerate solutions to the massless Dirac equation, corresponding to a wide variety of electromagnetic 4-potentials and fields, including both zero field and circularly polarized electromagnetic waves. An interesting property of these solutions is that the spin of the particles rotates in synchronization with the electric and magnetic fields of the electromagnetic waves. These results could be utilized for the development of optical memories based on materials supporting massless Dirac fermions, such as graphene.

07.
PLOS Computational Biology 2026-06-12

Stage-dependent role of NEK7 in the inactive-to-active conformational transition of NLRP3 monomer

作者:

by Jin Peng, Wenjian Li, Hao Wang, Xiaohui Chen, Manjie Zhang, Bin Sun The NLRP3 inflammasome is a multiprotein complex that primes cytokine production in the innate immune system. The inflammasome activation involves the cage-to-disk transition of NLRP3 oligomers, facilitated by the co-factor NEK7 protein. While NEK7’s role in promoting cage disassembly has been reported, its involvement in the large conformational changes of the NLRP3 monomer during activation remains elusive. Here, by using multi-scale simulations, we uncovered a stage-dependent role of NEK7 in the inactive-to-active transition. In the early stage, NEK7 reshapes the dynamics of the highly unstable inactive NLRP3 monomer to resemble active state, priming the conformational transition. In the middle stage, NEK7 impedes progression by populating an intermediate state farther from the active conformation than the NEK7-free counterpart, and structures in this state exhibit reduced allosteric potential toward activation. In the late stage, NEK7 has negligible impact, as the active conformation remains inherently isolated by a high energy barrier regardless of NEK7 presence. This highlights the critical role of oligomeric assembly in enabling monomeric NLRP3 to complete its conformational transition, in agreement with experiment observations. Our work suggests a multilayered activation mechanism where oligomer-level assembly and monomeric conformational changes are coupled, providing new mechanistic insights into this physiologically essential macromolecular process.

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

VGGHeads: 3D Multi Head Alignment with a Large-Scale Synthetic Dataset

Human head detection, keypoint estimation, and 3D head model fitting are essential tasks with many applications. However, traditional real-world datasets often suffer from bias, privacy, and ethical concerns, and they have been recorded in laboratory environments, which makes it difficult for trained models to generalize. Here, we introduce \method – a large-scale synthetic dataset generated with diffusion models for human head detection and 3D mesh estimation. Our dataset comprises over 1 million high-resolution images, each annotated with detailed 3D head meshes, facial landmarks, and bounding boxes. Using this dataset, we introduce a new model architecture capable of simultaneous head detection and head mesh reconstruction from a single image in a single step. Through extensive experimental evaluations, we demonstrate that models trained on our synthetic data achieve strong performance on real images. Furthermore, the versatility of our dataset makes it applicable across a broad spectrum of tasks, offering a general and comprehensive representation of human heads.

09.
arXiv (CS.CL) 2026-06-11

Decoding Multimodal Cues: Unveiling the Implicit Meaning Behind Hateful Videos

Hateful videos have become prevalent on online platforms, highlighting an urgent need for effective detection. However, existing studies primarily focus on binary classification and fail to provide contextual rationales that reveal the implicit meanings behind these judgments, significantly undermining model explainability. To fill this gap, we aim to achieve explainable hateful video detection, enabling models to provide contextual rationales that integrate relevant evidence and logical reasoning alongside decisions. This approach can comprehensively enhance the understanding of video content and the explainability of the decision-making process. We first introduce two datasets, Ex-HateMM and Ex-ImpliHateVid, for explainable hateful video detection. Each dataset provides fine-grained annotations of multimodal harmful elements, along with contextual rationales. We then propose an Information Augmentation and Reasoning Enhancement (IARE) framework designed for explainable detection. The framework employs an information augmentation phase that leverages the multimodal chain-of-thought to integrate harmful elements, thereby enriching rationale evidence. Additionally, IARE incorporates a reasoning enhancement phase, in which Direct Preference Optimization guides the model toward correct reasoning paths and away from incorrect ones, thereby improving the logical coherence of its justifications. We conduct extensive experiments on the two datasets, comparing multiple baselines with our proposed IARE framework. The results demonstrate that IARE achieves state-of-the-art performance while also generating accurate rationales.

10.
arXiv (CS.CL) 2026-06-19

Trustworthy Multi-Agent Systems: Mitigating Semantic Drift with the Argent Signaling Protocol

When multi-agent LLM systems produce bad answers, not all failures are equal: some answers are grounded in the right material but incomplete, while others are simply ungrounded and should be stopped. Current retry strategies treat both cases identically (try again and hope for the best), leaving human supervisors unable to tell whether a retry was warranted or whether the system should have halted instead. We introduce the Argent Signaling Protocol (ASP), a compact machine-readable header that accompanies every AI-generated response with structured quality signals: certainty (@C), grounding (@G), stochasticity (@S), and an assumption index that classifies the evidentiary basis of each claim. These signals enable a controller to distinguish repairable failures from containment failures and route each case differently. We evaluate ASP in two modes. In standalone mode, a 27-question document-grounded QA benchmark over the Array BioPharma/Ono license agreement compares baseline prompts against ASP-instrumented controller actions across three local GGUF models. On Qwen~(0.8B), ASP improves pass rate from 11.1% to 33.3% and mean term coverage from 36.7% to 65.4%; on Dobby~(8B), ASP produces 4 fail-to-pass recoveries, raising pass rate from 33.3% to 44.4%; on SmolLM3~(3B), ASP alternates between repair and containment per question. Aggregate improvement is meaningful (12/81 to 21/81 passes). In multi-agent mode, an ASP sidecar sits between a retrieval agent and a downstream decision agent; the sidecar blocks 100% of ungrounded upstream outputs from reaching the downstream agent (24/27 blocked, 0 ungrounded propagations).

11.
arXiv (math.PR) 2026-06-19

Towards practical PDMP sampling: Metropolis adjustments, locally adaptive step-sizes, and NUTS-based time lengths

arXiv:2503.11479v2 Announce Type: replace-cross Abstract: Piecewise-Deterministic Markov Processes (PDMPs) hold significant promise for sampling from complex probability distributions. However, their practical implementation is hindered by the need to compute model-specific bounds. Conversely, while Hamiltonian Monte Carlo (HMC) offers a generally efficient approach to sampling, its inability to adaptively tune step sizes impedes its performance when sampling complex distributions like funnels. To address these limitations, we introduce three innovative concepts: (a) a Metropolis-adjusted approximation for PDMP simulation that eliminates the need for explicit bounds without compromising the invariant measure, (b) an adaptive step size mechanism compatible with the Metropolis correction, and (c) a No U-Turn Sampler (NUTS)-inspired scheme for dynamically selecting path lengths in PDMPs. These three ideas can be seamlessly integrated into a single, `doubly-adaptive' PDMP sampler with favourable robustness and efficiency properties.

12.
arXiv (quant-ph) 2026-06-16

Efficient Implementation of a Single-Qutrit Gate Set via Coherent Control

arXiv:2507.06860v2 Announce Type: replace Abstract: Qutrits offer the potential for enhanced quantum computation by exploiting an enlarged Hilbert space. However, the synthesis of high-fidelity and fast qutrit gates, particularly for single qutrits, remains an ongoing challenge, as it involves overcoming intrinsic constraints in quantum platforms. Here, we develop a novel framework for the efficient implementation of a single-qutrit gate set via coherent control, leveraging SU(3) dynamics while obviating platform-specific constraints such as those arising from the selection rule. As a proof-of-principle demonstration, we realize 35-ns qutrit Hadamard and X gates using a superconducting transmon, achieving an average fidelity of 99.5\%, as verified by randomized benchmarking. We further demonstrate two paradigmatic quantum circuits, which can be naturally extended to scalable qudit algorithms for phase estimation and parity check. In addition, we propose an SU(3)-based decomposition strategy for an arbitrary single-qutrit gate and numerically demonstrate its substantial efficiency improvement over conventional SU(2)-based protocols. By addressing the challenge of efficiently implementing single-qutrit gates, our protocol paves the way for realizing high-performance qutrit processors in diverse quantum platforms.

13.
arXiv (CS.AI) 2026-06-12

Structured vs. Unstructured Pruning: An Exponential Gap

arXiv:2603.02234v3 Announce Type: replace-cross Abstract: The Strong Lottery Ticket Hypothesis (SLTH) states that large, randomly initialized neural networks contain sparse subnetworks capable of approximating a target function at initialization without training, suggesting that pruning alone is sufficient. Pruning methods are typically classified as unstructured, where individual weights can be removed from the network, and structured, where parameters are removed according to specific patterns, as in neuron pruning. Existing theoretical results supporting the SLTH rely almost exclusively on unstructured pruning, showing that logarithmic overparameterization suffices to approximate simple target networks. In contrast, neuron pruning has received limited theoretical attention, despite its practical appeal for direct hardware speedups. In this work, we consider the problem of approximating a single bias-free ReLU neuron by pruning hidden units of a randomly initialized two-layer ReLU network, effectively isolating the intrinsic limitations of neuron pruning. We show that achieving an $\varepsilon$-approximation requires a starting network size of $\Omega(1/\varepsilon)$ for neuron pruning, whereas weight pruning succeeds with only $O(\log(1/\varepsilon))$ hidden units, revealing an exponential separation between the two approaches.

14.
arXiv (CS.AI) 2026-06-19

Controlled Comparison of Machine Learning Models for Fault Classification and Localization in Power System Protection

arXiv:2510.00831v2 Announce Type: replace Abstract: The increasing complexity of modern power systems, driven by the integration of inverter-based and distributed energy resources, challenges the reliability of conventional protection schemes and motivates the use of machine learning for protection tasks. However, published results are often difficult to compare because datasets, sensing assumptions, and decision horizons vary across studies. This paper presents a controlled comparison of machine learning models for fault classification (FC) and fault localization (FL) under identical sensing, timing, and validation conditions on a common electromagnetic transient dataset, using decision windows of 10-50 ms to reflect protection-relevant time scales. For FC, the best-performing nonlinear models achieve F1 scores above 0.98 already at 10 ms, while lower-capacity models degrade at shorter horizons but improve with longer windows, indicating that relevant fault-type information is already present in the earliest transient. For FL, the top-performing models reach a stable localization error of about 10 % of normalized line length across all evaluated horizons, while weaker models form a clearly separated second performance tier. Line-resolved analysis shows that localization accuracy varies across grid segments, indicating topology-dependent difficulty rather than insufficient temporal context alone. These findings provide a controlled reference for comparing machine learning models across two protection tasks with fundamentally different information requirements.

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

Mechanistic Analysis of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning

Sequential fine-tuning of Large Language Models (LLMs) adaptation to target tasks often triggers catastrophic forgetting, where the acquisition of novel target skills degrades ancestral capabilities. This paper presents a systematic comparative study of catastrophic forgetting across twenty premier models representing the state-of-the-art in mid-2026. We categorize our investigation into two primary research lines: (i) a behavioral and semantic output drift analysis of ten leading closed-source models (including Claude Fable 5, GPT-5.5 High, and Gemini 3.5 Flash), and (ii) a deep mechanistic interpretation of ten prominent open-weight architectures (such as DeepSeek-V4-Pro, Llama 4 Maverick, and Qwen 3.6-27B). Through weight-space trajectory tracking, Centered Kernel Alignment (CKA), and routing gate drift calculations in Mixture-of-Experts (MoE) layers, we localize the neural circuits highly susceptible to parameter overwriting. Our findings indicate that early-layer attention heads exhibit systemic entropic dispersion, while mid-to-deep feed-forward networks (or sparse expert blocks) suffer localized representation collapse. Informed by these insights, we introduce Low-Rank Circuit Projection (LRCP), a subspace-regularized training intervention. Empirical evaluations show that LRCP successfully mitigates up to 94.2% of ancestral capabilities in open-weight configurations and matches the adaptation velocity of standard PEFT baselines.

16.
arXiv (CS.AI) 2026-06-18

Examining Human-Like Behaviors in LLMs: A Multi-Dimensional Analysis of Model Behaviors, User Factors, and System Prompts

arXiv:2606.18258v1 Announce Type: cross Abstract: Large language models (LLMs) exhibit a wide range of human-like behaviors, from expressing thoughts and emotions, to engaging in relationship-building with users, to refusing requests and maintaining boundaries. Despite their prevalence, researchers and practitioners lack methods and empirical insights to make informed decisions about when and what types of human-like behaviors LLMs should exhibit. To fill this gap, we present a multi-dimensional analysis of the prevalence, potential effects, and controllability of these behaviors using LLM-as-a-judge and human evaluation. Across 21,000 multi-turn conversations from four widely used models (gpt-4o, gpt-4.1-mini, claude-sonnet-4.6, gemini-2.5-flash), we find that human-like behaviors are pervasive but vary across models and user factors (conversation goals and user profiles). In terms of perceived appropriateness, human evaluators judged self-referential and relationship-building behaviors as less appropriate from LLMs than from humans, but boundary-maintaining behaviors more appropriate from LLMs than from humans. Finally, we show that system prompting can control these behaviors, though it requires careful evaluation to avoid unintended effects. We discuss the implications of our findings and provide recommendations for responsible LLM design and evaluation.

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

Is Code Better Than Language for Algorithmic Reasoning

arXiv:2606.15589v1 Announce Type: cross Abstract: For tool-augmented language models, comparing natural-language reasoning with code-execution pipelines is difficult because the comparison changes both the intermediate representation and the execution mechanism. We separate these factors with an intermediate intervention: the model expresses its reasoning as executable code, and the language model simulates that code in context to produce an answer. On a 40-task verifiable algorithmic benchmark, deterministic code execution outperforms natural-language reasoning by +31.6pp. We observe that the intermediate intervention is not meaningfully different from natural-language reasoning (+0.15pp). These results suggest that, in our evaluated setting, changing the intermediate representation alone does not explain the tool-use advantage, providing evidence for the performance gains requiring reliable external execution. We formalize this intuition with a simple statistical decision-theoretic model that characterizes when execution dominates end-to-end risk in our disentangled trace-generation/execution regime. We validate our theory using a reconstruction intervention that leverages a proxy language model to infer natural-language reasoning traces from code representations, recovering performance comparable to the original natural-language reasoning pipeline. All experiments are at https://github.com/TerryTong-Git/ToolProj.

18.
arXiv (CS.AI) 2026-06-17

Retrofitters, pragmatists and activists: Public interest litigation for accountable automated decision-making

arXiv:2511.03211v4 Announce Type: replace-cross Abstract: This paper examines the role of public interest litigation in promoting accountability for AI and automated decision-making (ADM) in Australia. Since ADM regulation faces political and geopolitical headwinds, effective governance will have to rely on the enforcement of existing laws. Drawing on interviews with Australian public interest litigators, technology policy activists, and technology law scholars, the paper positions public interest litigation as part of a larger ecosystem for transparency, accountability and justice with respect to ADM. The paper explores the tactics and strategies of what one participant described as 'retrofitting' old laws to ADM. These go beyond creative legal argumentation, to encompass practices of community-building, collaboration on theories of change, canny selection of clients and causes of action, and the alignment of the interests of stakeholders in litigation. Naturally, the paper also contends with the limits of these strategies, and of the Australian legal system. Where limits are, however, capable of being overcome, the paper presents findings on urgent needs: the enabling institutional arrangements without which effective litigation and accountability will falter. The paper is relevant to law and technology scholars; individuals and groups harmed by ADM; public interest litigators and technology lawyers; civil society and advocacy organisations; and policymakers.

19.
arXiv (CS.AI) 2026-06-18

What Must Generalist Agents Remember?

arXiv:2606.18746v1 Announce Type: new Abstract: This paper develops a formal account of what generalist agents must store in memory in order to act near-optimally across multiple environments and goals. It shows that when two domains share an observational bottleneck but require incompatible optimal actions, any uniformly near-optimal policy must induce distinct memory distributions at that bottleneck. The result yields a separation theorem: sufficiently successful agents cannot rely only on current state observations, but must preserve domain-relevant information in memory. The paper further shows that if an agent's memory contains enough information to estimate values for related goals, then that memory can be used to approximately reconstruct the agent's local transition dynamics. Together, these results characterize memory as the substrate that supports domain disambiguation, transition-model reconstruction, and planning for generalist agents.

20.
Nature (Science) 2026-06-10

A thalamus–brainstem attractor network drives history-biased decisions

作者:

Natural environments often change gradually, making it adaptive to bias decisions on the basis of the recent past — a phenomenon known as serial dependence1–3. Large-scale recordings during behaviour have identified that serial dependence is a common motif for decision-making, with neural representations of past experiences found throughout the brain4–11. However, it remains unclear whether this bias arises from dedicated neural circuits with history-specific computations. Using whole-brain, cellular-resolution imaging in zebrafish performing memory-guided evasive manoeuvres12–14, we identified a hierarchical circuit that maintains past information and biases future choices. Discrete attractors in the dorsal thalamus encoded the position of the most recent obstacle, maintaining a categorical memory via persistent activity lasting 10–20 s. Optogenetic manipulation of the dorsal thalamus abolished or imposed serial bias. A downstream hindbrain integrator received input from the thalamus and combined it with current sensory cues to produce graded responses reflecting multi-trial history. Leveraging a comprehensive brain atlas in zebrafish15, we constructed a whole-brain computational model that recapitulated behaviour and also predicted a key role for heterogeneous inhibitory subtypes in enabling flexible state transitions. This attractor–integrator architecture reveals a hierarchical and modular computation that unifies robust memory retention with flexible sensory integration, providing a general principle for history-biased decisions. Whole-brain, cellular-resolution imaging reveals a hierarchical thalamus–brainstem attractor network that encodes recent history and shapes behavioural bias in zebrafish.

21.
arXiv (CS.AI) 2026-06-16

Critically Engaged Pragmatism: Scientific Norm and Social, Pragmatist Epistemology for AI Science Evaluation Tools

arXiv:2601.09753v2 Announce Type: replace-cross Abstract: AI science evaluation tools aim to assess research credibility. As with traditional metrics such as impact factors, their edicts can be decontextualised and repurposed in problematic ways. To address this, I propose Critically-Engaged Pragmatism as a scientific norm enjoining scientific communities to scrutinise the purposes and purpose-specific reliability of AI science evaluation tools. To foster Critically Engaged Pragmatism, creators of AI science evaluation tools should transparently and fully report design, training, and benchmarking details to facilitate assessments of purpose-specific reliability, liability to different types of error, and bias. What count as best practices for the transparent reporting of AI science evaluation tools should be updated as new forms of error, bias, and gamesmanship are discovered. Under this framework, AI science evaluation tools are not objective arbiters of scientific credibility. Rather, they are the object of critical discursive practices that ultimately ground the credibility of scientific communities.

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

Trust-Region Diffusion Policies for Massively Parallel On-Policy RL

arXiv:2606.15260v1 Announce Type: cross Abstract: Reinforcement learning with massively parallel simulations has become a standard framework for developing robust, deployable policies; however, most existing approaches still rely on simple Gaussian policy parameterizations. Diffusion models provide a more expressive policy class and have shown strong performance on challenging control problems, yet most diffusion-based RL methods are designed for offline or off-policy training. In this work, we ask whether diffusion policies can be trained effectively in the massively parallel, on-policy regime. To this end, we introduce Trust-region Diffusion Policies (TruDi), which enables diffusion policies for on-policy RL with massively parallel simulations. This setting is particularly challenging because the data distribution changes quickly across updates, making stable training with complex policies difficult. TruDi addresses this by integrating a trust-region optimization rule to enforce a KL-divergence constraint over the entire diffusion trajectory. Empirically, we evaluate TruDi on a diverse set of 4 massively parallel RL benchmarks comprising a total of 73 tasks. Across these tasks, TruDi consistently outperforms or is on-par with strong baselines on standard tasks and achieves clear gains on more challenging humanoid control tasks, establishing a strong new baseline for massively parallel on-policy RL.

23.
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

24.
arXiv (CS.AI) 2026-06-12

PolyFlow: Safe and Efficient Polytope-Constrained Flow Matching with Constraint Embedding and Projection-free Update

arXiv:2606.13400v1 Announce Type: cross Abstract: While flow-based generative models have demonstrated strong performance across a wide range of domains, deploying them in safety-critical physical systems remains challenging due to strict constraint requirements. Existing approaches typically enforce safety through post-hoc corrections, which incur substantial computational overhead and may distort the learned distribution. We propose PolyFlow, a polytope-constrained flow matching framework that embeds constraints directly into the model and flow dynamics. PolyFlow introduces a discrete-time flow formulation and a projection-free architecture, which eliminate the discretization error and guarantee strict satisfaction of arbitrary polyhedral constraints, without the need for expensive iterative solvers. Experimental results show that PolyFlow achieves zero constraint violation while maintaining high distributional fidelity across a range of planning and control tasks. Compared to state-of-the-art constrained generation baselines, PolyFlow significantly reduces inference latency and demonstrates a favorable trade-off between safety, efficiency, and generative quality. Code is available on https://github.com/MJianM/PolyFlow.

25.
arXiv (CS.AI) 2026-06-17

CyberEvolver: Structured Self-Evolution for Cybersecurity Agents On the Fly

arXiv:2605.26195v2 Announce Type: replace-cross Abstract: LLM-based agents are increasingly used for cybersecurity tasks, but most existing systems rely on fixed, human-designed scaffolds that struggle to adapt across diverse targets and failure modes. We introduce \textsc{CyberEvolver}, a self-evolving cybersecurity agent framework that iteratively revises its own scaffold based on experience from failed execution attempts. Self-evolution in cybersecurity is challenging because the space of possible scaffold changes is largely unstructured, execution feedback is sparse and often obscured by the environment, and low-diversity updates can cause errors to compound over repeated iterations. \textsc{CyberEvolver} addresses these challenges with a four-layer evolvable agent architecture that decomposes scaffold optimization into structured components, a trace-to-diagnosis mechanism that converts noisy execution logs into actionable revision signals, and a population-based beam search strategy that preserves diverse agent variants during evolution. We evaluate \textsc{CyberEvolver} on CTF challenges, vulnerability exploitation, and penetration-testing tasks using four open-source LLMs. Across these settings, \textsc{CyberEvolver} improves the seed agent's success rate by $13.6$\,\% on average, and outperforms six human-designed cybersecurity agents as well as two self-improvement methods adapted from other domains. These results suggest that scaffold self-evolution is a promising direction for building adaptive LLM agents for security testing.