中 EN
Sign In Register

Academic Intelligence · Curated Daily

Explore the Frontier of Global Academia

AcademicHub aggregates real-time literature from top journals and preprint platforms. Build your personal research radar and let large language models compile cross-disciplinary analysis briefings automatically.

Sign In Register
01.
arXiv (math.PR) 2026-06-19

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

Authors:
Augustin Chevallier↗Sam Power↗Matthew Sutton↗

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.

Read & Discuss → View Source →
02.
arXiv (CS.AI) 2026-06-11

Architecture-Aware Reinforcement Learning Makes Sliding-Window Attention Competitive in Math Reasoning

Authors:
Kai Liu↗Peijie Dong↗Xinchen Xie↗Jianfei Gao↗Qipeng Guo↗Xiaowen Chu↗Shaoting Zhang↗Kai Chen↗

arXiv:2606.11634v1 Announce Type: new Abstract: The rapid progress of reasoning and agentic large language models (LLMs) has increased the demand for long-context inference, but self-attention (SA) scales quadratically with context length. To address this, we study SWARR (Sliding-Window Attention with Reinforced Adaptation for Math Reasoning), a practical recipe for adapting SWA models to mathematical reasoning. SWARR has two stages: (1) efficient conversion from a pretrained SA model to SWA with supervised fine-tuning (SFT), which avoids pretraining a new base model, and (2) policy adaptation with reinforcement learning (RL). We find that SWA still underperforms SA after SFT, and we hypothesize that this gap is caused in part by a data-architecture mismatch: most SFT data are prepared for SA models and may contain long-range dependencies that are difficult for SWA to model. Because on-policy RL optimizes self-generated trajectories under the SWA constraint, it can adapt trajectories to better match SWA. Experiments on mathematical reasoning benchmarks show that this recipe substantially narrows the gap between SWA and SA, recovering much of the accuracy lost during SWA conversion while preserving the efficiency benefits of linear-complexity attention. Our central contribution is the empirical finding that RL changes the conclusion one would draw from conversion and SFT alone about SWA's viability for math reasoning.

Read & Discuss → View Source →
03.
Nature (Science) 2026-06-17

Revealing competitive interfacial reactions in high-energy Li–S batteries

Authors:
Shiyuan Zhou↗

Charge transfer at solid–liquid interfaces plays a critical role in various energy-storage systems1, particularly under dynamically varying reactant concentrations. Deciphering these intricate reaction pathways remains a substantial challenge, notably in lithium–sulfur (Li–S) batteries, in which achieving high energy density requires efficient conversion of highly concentrated lithium polysulfides (LiPSs)2,3. However, the mechanisms governing lithium sulfide (Li2S) deposition and dissolution under lean electrolyte conditions remain poorly understood. Here, using in situ liquid-cell electron microscopy, we directly visualize concentration-driven phase segregation at the electrode–electrolyte interface. Within these high-concentration interfacial layers (HCILs), competitive surface and solution dictate the charge-transfer dynamics and ultimately govern Li2S deposition at different phase boundaries. Density functional theory (DFT) calculations reveal that the aggregation of LiPSs alters molecular geometry, electronic properties and orbital hybridization, collectively facilitating charge transfer through highly concentrated LiPSs clusters. Guided by these insights, we design optimized electrodes that balance interfacial reaction pathways, enabling fast charging (4 C, 26.8 mA cm−2) and achieving high energy densities exceeding 400 Wh kg−1. These findings provide mechanistic understanding of interfacial reactions under practical working conditions and offer a design strategy to advance Li–S batteries. Visualization of concentration-driven phase segregation within high-concentration interfacial layers in the context of high-energy lithium–sulfur batteries using liquid-cell electrochemical transmission electron microscopy reveals competitive interfacial reactions under lean electrolyte conditions at different phase boundaries.

Read & Discuss → View Source →
04.
bioRxiv (Bioinfo) 2026-06-14

Virtual phenotypic screening discovers novel scaffolds inhibiting the PI3K/mTOR pathway

Authors:
Wu↗A. P↗Yao↗Hoeckendorf↗Gaskins↗Kosaisawe↗Lu↗Hanslovsky↗Mayba↗Skelton↗Scalia↗Moffat↗…

Phenotypic drug discovery has yielded many first-in-class small-molecule drugs by discovering modulators of disease phenotypes in physiologically relevant cellular systems. However, high-content phenotypic assays lack the ultra-high-throughput scalability of target-based screens. Recent advances in virtual screening present an opportunity to address this bottleneck, but have been limited to simple phenotypes like viability, restricted to small repurposing libraries, or lack in-depth biological validation. Here, we present PhenoCompass, a multimodal co-embedding model that aligns compound structures and high-content phenotypic imaging to enable virtual phenotypic screening over billion-compound libraries. Following training on the Joint Undertaking in Morphology dataset with more than 100,000 Cell Painting compound profiles, retrospective validation with historical biochemical high-throughput screening data demonstrates that PhenoCompass ranks compounds according to their biochemical target engagement. Leveraging PhenoCompass, we performed a prospective screen of 3.8 billion Enamine REAL compounds for inhibitors of PI3K/mTOR pathway, a critical signaling cascade whose aberrant activation is a common tumor driver. This search identified 11 novel compounds with pathway-consistent Cell Painting readout and diverse scaffolds, a 54-fold enrichment over the training set. Orthogonal validation experiments using a FOXO3A reporter assay and direct kinase inhibition confirmed seven structurally novel inhibitors with distinct mechanisms of action. These results highlight the convergence of diverse molecular target profiles onto a shared morphological pathway signature and establish PhenoCompass as a robust framework for high-content phenotypic virtual screening.

Read & Discuss → View Source →
05.
arXiv (CS.AI) 2026-06-19

Thermodynamic Measure of Intelligence

Authors:
Ishanu Chattopadhyay↗

arXiv:2606.20231v1 Announce Type: new Abstract: Can intelligence be measured? We propose that intelligence can be defined as the lawful amplification of rare but valid futures: a system increases the probability of outcomes that would be unlikely under passive dynamics but remain admissible under the constraints of the domain. We start with the premise that an intelligent system must model the world and its own place within it. Because the system is part of the world it models, this leads naturally to recursive self-simulation: the system represents futures in which its own actions are part of the trajectory. Our central results give a necessity statement and a conditional near-sufficiency statement connecting this architecture to a precise thermodynamic measure of lawful amplification of rare-valid futures: high rare-valid lift is impossible unless the internal simulation identifies rare-valid futures with high fidelity; conversely, when rare-valid fidelity is high and the simulation contains an effective policy, the achievable lift approaches the actuation-limited optimum. Thus recursive self-simulation is not merely a plausible feature of intelligence but, under the stated assumptions, is necessary and nearly sufficient for high thermodynamic intelligence. The resulting framework makes intelligence measurable on a universal scale, from passive matter and feedback controllers, large language models, and humans as text generators to Maxwell-demon-like information engines.

Read & Discuss → View Source →
06.
arXiv (CS.AI) 2026-06-17

Online LLM Selection via Constrained Bandits with Time-Varying Demand

Authors:
Yin Huang↗Qingsong Liu↗Jie Xu↗

arXiv:2606.17489v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly deployed in edge-cloud inference systems to handle diverse user tasks with heterogeneous accuracy, latency, and cost profiles. Selecting the appropriate LLM for each incoming task is critical for ensuring service quality and efficient resource utilization. However, model heterogeneity, stochastic and unknown performance characteristics, and time-varying task demands make static selection strategies inadequate. Real-world deployments often impose hard resource budgets such as monetary expenditure limits, along with soft service-level requirements such as latency guarantees. These constraints introduce additional challenges for online decision-making. We formulate this problem as a constrained stochastic bandit learning task, where the learner sequentially selects models under both packing-type (hard) and covering-type (soft) constraints, while adapting to time-varying task demand. The learner operates without access to the underlying reward, cost, or latency distributions and must rely on partial feedback. We develop a novel online learning algorithm that leverages confidence-bound estimates and demand predictions to balance reward maximization with long-term constraint satisfaction. We provide theoretical guarantees showing sublinear regret and sublinear covering constraint violations compared to an offline benchmark with full information. Experimental results on synthetic workloads demonstrate the effectiveness and robustness of our approach in dynamic, resource-constrained environments.

Read & Discuss → View Source →
07.
arXiv (CS.CV) 2026-06-12

Masked and Predictive Self-Supervised Foundation Models for 3D Brain MRI

Authors:
Esra Erg\"un↗Hersh Chandarana↗Dan Sodickson↗G\"ozde \"Unal↗

Self-supervised foundation models have shown strong promise in medical imaging. However, existing MRI foundation-model studies have primarily emphasized segmentation and dense prediction tasks, while systematic investigation of self-supervised foundation models for MRI-based disease detection remains limited. In this work, we investigate two major self-supervised pretraining paradigms for MRI-based disease detection: reconstruction-based learning via Masked Autoencoders (MAE) and predictive representation learning via Joint Embedding Predictive Architectures (JEPA). We study the role of auxiliary objectives by introducing a novel spectral-domain reconstruction loss for MAE to enhance sensitivity to fine-grained anatomical structure, and by integrating variance–covariance regularization (VCR) within our JEPA framework to encourage decorrelated latent representations. Our models are pretrained on heterogeneous single-contrast MRI volumes in a contrast-agnostic setting, without modality concatenation. Across five downstream disease detection tasks, our results highlight the importance of self-supervised objective design for medical foundation model pretraining, demonstrating that the downstream benefit of each objective is determined by its relevance to the task's structure. Specifically, spectral regularization yields the largest improvements when the downstream discriminative signal is characterized by strong high-frequency anatomical structures, while covariance regularization is most beneficial when discriminative information spans multiple decorrelated feature dimensions. MAE with spectral-domain supervision consistently achieves superior downstream performance for MRI-based disease detection. These findings suggest that self-supervised objectives in medical imaging encode specific biases, and their downstream benefit is fundamentally conditioned on the task's structure.

Read & Discuss → View Source →
08.
arXiv (math.PR) 2026-06-11

Numerical simulations of the spread from the mean of the SLE and Multiple SLE dynamics

Authors:
Phillip Kim↗Vlad Margarint↗

arXiv:2606.11254v1 Announce Type: cross Abstract: The Schramm-Loewner Evolution (SLE) describes a family of fractal curves that arise in the study of the scaling limits of many planar Statistical Physics models. These curves are modeled using the Loewner Differential Equation for the conformal maps $g_t(z)$ with a Brownian motion driver. Using Euler's Method, in the current work we performed numerical experiments to study at a fixed time the quantities $|g_t(z) - \overline{g_t(z)}|$ and $Re(g_t(z)) - Re(\overline{g_t(z)})$, where $Re$ denotes the real part and $\overline{g_t(z)}$ refers to the sample average. These random variables measure the 'spread' of the dynamics from the average behavior at fixed time. One of the scopes of this work is to give numerical predictions for future theoretical investigations on these quantities. When investigating these quantities in the SLE case our experiments predict that the distribution is bimodal when the dynamics started close to the origin, and it can become bell-shaped if the dynamics is started further from the origin. In the second part, we performed experiments for a Multiple SLE model whose driver is Dyson Brownian Motion. Due to singularity in the dynamics of the drivers and the many data points needed, this part is challenging from a computational perspective. In the multiple SLE case, our experiments predict that the distribution is bell-shaped in all cases. In addition, we check the changes in the distributions as we vary the parameter $\kappa$ in the SLE case and $\beta$ in the Multiple SLE case.

Read & Discuss → View Source →
09.
arXiv (CS.AI) 2026-06-16

Your Agent Has a Genome: Sequence-Level Behavioral Analysis and Runtime Governance of LLM-Powered Autonomous Agents

Authors:
Sidi Deng↗

arXiv:2606.15579v1 Announce Type: new Abstract: We propose Base Sequence Analysis, a framework that encodes the runtime behavior of LLM-powered autonomous agents into compact symbolic sequences using a four-letter alphabet: X (Explore), E (Execute), P (Plan), and V (Verify). Drawing an analogy to genomic sequence analysis, we apply n-gram pattern mining, Markov transition matrices, and point-biserial correlation to 347 real-world execution traces collected from a production ReAct agent system over 8 days. Our analysis reveals that (1) the trigram P-X-P is the only statistically significant high-risk pattern, lowering success rate by 10.4%; (2) P-ratio is the strongest negative predictor of success (r=-0.256, pV transition probability is only 2.1%, indicating a systemic verification deficit. Based on these findings, we design Governor, a three-layer runtime intervention system comprising a rule engine, a statistical accumulator, and a chi-square-based threshold adaptor. In a natural before/after deployment evaluation (N=101 vs. N=246), Governor achieves a +6.2% absolute increase in task success rate while simultaneously reducing average token consumption by 44%. To validate cross-system generality, we apply the XEPV encoding to 2,000 public SWE-agent trajectories on SWE-bench, confirming that exploration spirals and the E->V verification deficit replicate in an independent system. We outline six research directions including base sequence language models, cross-agent behavioral fingerprinting, and reward shaping, and release an open-source toolkit for reproducibility.

Read & Discuss → View Source →
10.
arXiv (CS.CV) 2026-06-12

TimeLens: On-Device Artifact Recognition with Retrieval-Augmented Question Answering for the Grand Egyptian Museum

Authors:
Rawan Hesham↗Ali Ashraf↗Amr Ahmed↗Malak Alaa↗Omar Ahmed↗Omar Wagih↗

TimeLens is an AI-powered bilingual mobile guide for the Grand Egyptian Museum (GEM). Pointing a phone at an exhibit, a visitor sees the artifact recognized in real time and can ask follow-up questions answered in English or Arabic. The work addresses three problems specific to in-gallery deployment: fine-grained visual similarity among 51 catalogued artifacts (many near-identical Ramesside statues), the gap between curated training data and handheld camera conditions, and the risk of an AI guide stating unsupported historical facts. Two engineering contributions are reported. First, an on-device artifact detector was developed through a data-quality-driven iteration study – from foundation-model auto-annotation (YOLO-World), through spatial label-cleaning rules, to a fully hand-annotated dataset – isolating label quality as the decisive factor: the final YOLOv8n model resolves every previously failing class while remaining a 5.97 MB TensorFlow Lite asset that runs in real time on a mid-range phone (mAP@0.5 = 0.995, mAP@0.5:0.95 = 0.924). Second, a bilingual Retrieval-Augmented Generation (RAG) guide, grounded in a 108-record ChromaDB knowledge base, was benchmarked across seven candidate language models, with Gemma 4 E2B (Q4 K M) selected; ten targeted optimizations reduce end-to-end latency from over 30 s to approximately 10 s. Both subsystems are integrated in a production Flutter application with bilingual interface, museum location gating, and text-to-speech support.

Read & Discuss → View Source →
11.
arXiv (CS.AI) 2026-06-15

CADET: Physics-Grounded Causal Auditing and Training-Free Deconfounding of End-to-End Driving Planners

Authors:
Zikun Guo↗

arXiv:2606.14438v1 Announce Type: cross Abstract: End-to-end (E2E) autonomous-driving planners trained by imitation are prone to statistical shortcuts: they associate scene elements that merely co-occur with expert actions (a roadside object, a building facade) with driving decisions, rather than the variables that causally determine them. Such causal confusion silently compromises reliability in long-tail scenarios, and it is difficult to detect, because prevailing open-loop metrics (L2 displacement and collision rate) are dominated by ego status and do not indicate whether a planner depends on spurious cues. Existing remedies based on causal-intervention training require retraining large models and cannot audit a planner that is already deployed. We present CADET, a training-free framework that audits, benchmarks, and repairs spurious reliance in pretrained E2E planners without any parameter update.

Read & Discuss → View Source →
12.
arXiv (CS.LG) 2026-06-18

Knockoffs-based False Discovery Rate Control and Simplification for Deep Neural Networks

Authors:
Wenyu Liao↗Yiqing Shi↗Fang Xie↗

arXiv:2606.04404v2 Announce Type: replace-cross Abstract: The deep neural network is a widely used framework in machine learning that has been widely applied in various fields. However, deep neural networks often involve a large number of parameters and inputs, many of which may be irrelevant to the goal or true output. These parameters and input variables not only increase computational complexity, but also contribute to additional computational cost. One solution to this problem is knockoff methods, which have proven successful in controlling false discovery rates in high-dimensional regression. Building on the knockoff methods and using the regularised neural network, this paper proposes three variable screening methods under the condition of controlling false discovery rates: one layer filter, multiple layers filter, and variable weight aggregation filter. In comparison with existing algorithms, we find that our algorithms show satisfactory performance.

Read & Discuss → View Source →
13.
arXiv (CS.LG) 2026-06-11

On Subquadratic Architectures: From Applications to Principles

Authors:
Anamaria-Roberta Hartl↗Levente Z\'olyomi↗David Stap↗Pieter-Jan Hoedt↗Niklas Schmidinger↗Lukas Hauzenberger↗Sebastian B\"ock↗G\"unter Klambauer↗Sepp Hochreiter↗

arXiv:2606.12364v1 Announce Type: new Abstract: Transformers dominate modern sequence modeling, but their quadratic attention incurs substantial computational cost. Subquadratic architectures offer a scalable alternative. However, it remains unclear which designs yield the most effective sequence models. We compare three leading approaches: xLSTM, Mamba-2, and Gated DeltaNet. We evaluate these models on tasks with complex dependencies: (1) code-model pre-training, (2) distillation of code models from large language models, and (3) pre-training of time-series foundation models. Across these settings, xLSTM delivers the strongest overall performance. To explain xLSTM's advantage, we present a unified formulation and analyze the underlying architectural mechanisms, focusing on state tracking and memory dynamics. Our results show that xLSTM enables more flexible and stable memory correction via its gating scheme. We corroborate these findings on controlled synthetic length-generalization tasks. Overall, our findings indicate that xLSTM's gains on complex tasks stem from robust state tracking and accumulation.

Read & Discuss → View Source →
14.
arXiv (CS.CL) 2026-06-16

ACC: Compiling Agent Trajectories for Long-Context Training

Authors:
Qisheng Su↗Zhen Fang↗Shiting Huang↗Yu Zeng↗Yiming Zhao↗Kou Shi↗Ziao Zhang↗Lin Chen↗Zehui Chen↗Lijun Wu↗Feng Zhao↗

Recent development of agents has renewed demand for long-context reasoning capacity of LLMs. However, training LLMs for this capacity requires costly long-document curation or heuristic context synthesis. We observe that agents produce massive trajectories when solving problems, invoking tools and receiving environment observations across many turns. The evidence needed to answer the original question is thus scattered throughout these turns, requiring integration of distant context segments. Nevertheless, standard agent SFT masks tool responses and only trains turn-level tool selection, creating a supervision blind spot where these scattered signals go unused. We propose Agent Context Compilation (ACC), which converts trajectories from search, software engineering, and database querying agents into long-context QA pairs that combine the original question with tool responses and environment observations gathered across multiple turns, training the model to answer directly without tool use. This makes the dependencies between the question and the evidence explicit, enabling direct supervision of long-context reasoning over distant segments without additional annotation. ACC is a simple but effective approach that can be combined with any existing long-context extension or training method, providing scalable supervised fine-tuning data. We validate ACC on long-range dependency modeling tasks through MRCR and GraphWalks, challenging benchmarks requiring cross-turn coreference resolution and graph traversal over extended contexts. Training Qwen3-30B-A3B with ACC achieves 68.3 on MRCR (+18.1) and 77.5 on GraphWalks (+7.6), results comparable to Qwen3-235B-A22B, while preserving general capabilities on GPQA, MMLU-Pro, AIME, and IFEval. Further mechanism analysis reveals that the ACC-trained model exhibits task-adaptive attention restructuring and expert specialization.

Read & Discuss → View Source →
15.
arXiv (CS.AI) 2026-06-18

Sparsity Curse: Understanding RLVR Model Parameter Space from Model Merging

Authors:
Chenrui Wu↗Zexi Li↗Jiajun Bu↗Jiangchuan Liu↗Haishuai Wang↗

arXiv:2606.18521v1 Announce Type: cross Abstract: Reinforcement Learning with Verifiable Reward (RLVR) has emerged as a powerful post-training paradigm that surpasses Supervised Fine-Tuning (SFT) in eliciting reasoning intelligence and resisting catastrophic forgetting. Recent studies further reveal that RLVR induces highly sparse and off-principal parameter updates compared to SFT. This naturally raises the question: does such sparsity make RLVR models more amenable to model merging? If so, model merging would offer a scalable, training-free path to aggregate diverse reasoning capabilities from independently trained RLVR models. Surprisingly, we find the opposite, uncovering a sparsity curse: the sparse RLVR updates are spread farther apart in parameter space, forming near-orthogonal shortcuts that make aggregation inherently fragile. This is likely rooted in the stochasticity of RL optimization and the diversity of emergent reasoning patterns. Unlike SFT models that converge to shared, flat basins and merge naturally, RLVR models suffer severe degradation under standard merging methods. Through systematic empirical analysis of the update geometry, we characterize the mechanisms behind this failure and propose Sensitivity-aware Resolving Merging (SAR-Merging), a merging recipe tailored for the unique structure of RLVR parameter spaces. SAR-Merging resolves conflicts in overlapping update regions via Fisher Information-based sensitivity arbitration, followed by magnitude-aware sparsification and rescaling to preserve fragile reasoning pathways. Experiments on mathematical and coding benchmarks demonstrate that SAR-Merging substantially outperforms existing merging methods on RLVR models, enabling both single-task enhancement and multi-capability fusion.

Read & Discuss → View Source →
16.
arXiv (CS.AI) 2026-06-15

HyperPotter: Spell the Charm of High-Order Interactions in Audio Deepfake Detection

Authors:
Qing Wen↗Haohao Li↗Zhongjie Ba↗Peng Cheng↗Miao He↗Li Lu↗Kui Ren↗

arXiv:2602.05670v2 Announce Type: replace-cross Abstract: Advances in AIGC technologies have enabled the synthesis of highly realistic audio deepfakes capable of deceiving human auditory perception. Although numerous audio deepfake detection (ADD) methods have been developed, most rely on local temporal/spectral features or pairwise relations, overlooking high-order interactions (HOIs). HOIs capture discriminative patterns that emerge from multiple feature components beyond their individual contributions. We propose HyperPotter, a hypergraph-based framework designed to capture high-order relations associated with synergistic patterns through clustering-based hyperedges with class-aware prototype initialization. Extensive experiments on 13 test sets show that HyperPotter improves over the baseline on 11 sets, yielding an average relative EER reduction of 12.68\% across all test sets and 22.15\% on the improved sets. These results demonstrate strong cross-scenario generalization, while also revealing robustness limits under severe codec or channel distortion.

Read & Discuss → View Source →
17.
arXiv (CS.CL) 2026-06-16

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

Authors:
Sina Hajimiri↗Masih Aminbeidokhti↗Jose Dolz↗Ismail Ben Ayed↗Issam H. Laradji↗Spandana Gella↗Nicolas Gontier↗

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.

Read & Discuss → View Source →
18.
arXiv (CS.AI) 2026-06-16

Sample from What You See: Visuomotor Policy Learning via Diffusion Bridge with Observation-Embedded Stochastic Differential Equation

Authors:
Zhaoyang Liu↗Mokai Pan↗Zhongyi Wang↗Kaizhen Zhu↗Haotao Lu↗Haipeng Zhang↗Jingya Wang↗Ye Shi↗

arXiv:2512.07212v3 Announce Type: replace Abstract: Imitation learning with diffusion models has advanced robotic control by capturing the multi-modal action distributions. However, existing methods typically treat observations only as high-level conditions to the denoising network, rather than integrating them into the stochastic dynamics of the diffusion process itself. As a result, the sampling is forced to begin from random noise, weakening the coupling between perception and control and often yielding suboptimal performance. We propose BridgePolicy, a generative visuomotor policy that directly integrates observations into the stochastic dynamics via a diffusion-bridge formulation. By constructing an observation-informed trajectory, BridgePolicy enables sampling to start from a rich and informative prior rather than random noise, substantially improving precision and reliability in control. A key difficulty is that diffusion bridge normally connects distributions of matched dimensionality, while robotic observations are heterogeneous and not naturally aligned with actions. To overcome this, we introduce a semantic aligner to unify the visual and state inputs and align the observations with action representations, making diffusion bridge applicable to heterogeneous robot data. Extensive experiments across 52 simulation tasks on three benchmarks and 5 real-world tasks demonstrate that BridgePolicy consistently outperforms state-of-the-art generative policies. Our code is available at https://jianghcsr.github.io/BridgePolicy_page/.

Read & Discuss → View Source →
19.
arXiv (CS.CL) 2026-06-11

Adaptive Multi-Resolution Procedural Knowledge Compression for Large Language Models

Authors:
Changyue Wang↗Weihang Su↗Qingyao Ai↗Yichen Tang↗Runzhong Qiao↗Xuancheng Li↗Min Zhang↗Yiqun Liu↗

Large language models (LLMs) are widely used to tackle complex tasks with autonomous workflows. Recently, reusable natural language skills have emerged as a popular paradigm to inject procedural knowledge into LLM applications. Since popular skills are often invoked repeatedly, placing their full text in every context significantly increases prefill cost and latency. While text compression techniques have the potential to solve this problem, most existing methods are designed to compress factual knowledge in documents instead of procedural knowledge, making them insufficient for skill compression. In this paper, we argue that an effective skill compression method should: 1) preserve logical dependencies among workflows and tool protocols, 2) enable lightweight, offline compression for frequently updated community skills, and 3) be adaptable to varying complexities across skills. To address this, we present SKIM (SKIll coMpression), an adaptive multi-resolution soft token compression framework for procedural skills. Depending on the complexity of each skill, SKIM creates different numbers of soft tokens that not only improve the efficiency of LLM inference, but also preserve the effectiveness of skill usage. Experiments indicate that SKIM compresses skills to 30 to 60 percent of their original token length while preserving task performance better than existing compression methods.We have released our code at https://github.com/bebr2/SKIM .

Read & Discuss → View Source →
20.
arXiv (CS.CL) 2026-06-16

AuAu: A Benchmark for Auditing Authoritarian Alignment in Large Language Models

Authors:
Andreas Einwiller↗Max Klabunde↗Florian Lemmerich↗

The worldwide surge of authoritarianism, combined with the increasing central role in users' everyday lives, raises the question of to what extent specific models exhibit or promote authoritarian attitudes and characteristics. We introduce AuAu, a comprehensive benchmark that aims to assess the risk of LLMs generating responses with authoritarian tendencies. This benchmark combines three evaluation approaches: (i) psychometric questions from an extensive pool of 15 human validated instruments; (ii) contextual behavior vignettes probing intended actions in concrete situations; and (iii) responses to realistic user prompts. Unlike prior work, AuAu evaluates not only a general closeness towards authoritarianism but also the established sub-concepts Authoritarian Aggression, Authoritarian Submission, and Conventionalism. Evaluating 17 models from China, the EU, Russia, and the USA, we find that all tested models exhibit substantial authoritarian response rates under the psychometric evaluation, though rates drop significantly in increasingly more realistic downstream task. We further find that an authoritarian system prompt easily manipulates 15 out of 17 models to promote increased authoritarianism. Our results underscore the need for continued, systematic auditing of LLM-based AI systems to detect and ultimately mitigate undesired authoritarian tendencies in generated output. Our code and data are available at: https://github.com/andreaseinwiller/AuAu

Read & Discuss → View Source →
21.
arXiv (CS.CL) 2026-06-16

How Much Can We Trust LLM Search Agents? Measuring Endorsement Vulnerability to Web Content Manipulation

Authors:
Yimeng Chen↗Zhe Ren↗Firas Laakom↗Yu Li↗Dandan Guo↗J\"urgen Schmidhuber↗

Large language model (LLM)-based search agents synthesize open-web content into actionable recommendations on behalf of users, creating a risk that attacker-published pages are transformed into endorsed claims. We introduce SearchGEO, a controlled evaluation framework for measuring endorsement corruption in LLM-based web-search agents, combining a web-evidence manipulation pipeline, a five-mode attack taxonomy, and multiple output-level metrics. We evaluate 13 LLM backends on 308 cases each. Results show that vulnerability patterns vary across backends: overall attack success rate (ASR) ranges from 0.0% on Claude-Sonnet-4.6 to 31.4% on Gemini-3-Flash, the strongest attack mode differs by model family, and the same deployment scaffold could amplify or decrease ASR on different backends. An auxiliary agent-skill probe, where endorsement becomes an install command, exposes a sharp split among otherwise robust backends: Claude over-rejects while GPT over-trusts. These findings argue for treating recommendation reliability under adversarial search content as a first-class dimension of backend safety evaluation.

Read & Discuss → View Source →
22.
arXiv (CS.CL) 2026-06-11

Breaking Entropy Bounds: Accelerating RL Training via MTP with Rejection Sampling

Authors:
Yucheng Li↗Huiqiang Jiang↗Yang Xu↗Jianxin Yang↗Yi Zhang↗Yizhong Cao↗Yuhao Shen↗Fan Zhou↗Rui Men↗Jianwei Zhang↗An Yang↗Bowen Yu↗…

Reinforcement learning (RL) has become a key component in modern large language models, yet the rollout stage remains the key bottleneck in RL training pipelines. Although Multi-Token Prediction (MTP) offers a natural solution to accelerate rollouts through speculative decoding, many studies have observed that MTP acceptance rates degrade significantly during RL training, leading to limited speedup performance. To address this bottleneck, we present Bebop, a systematic study of MTP in LLM post-training, and offer practical recipes to integrate MTP into large-scale RL pipelines. First, we reveal that the MTP acceptance rate is fundamentally bounded by the fluctuation of model entropy, which demonstrates a clear negative linear relationship with the rise of entropy in the RL stage. Second, we show that probabilistic rejection sampling largely alleviates the disturbance introduced by entropy in RL compared to greedy draft sampling. We further identify that the conventional MTP training objectives (cross-entropy or KL) are suboptimal in such settings, and therefore we propose a novel end-to-end TV loss that directly optimizes multi-step rejection sampling acceptance rate, yielding ~10% acceptance rate improvements, achieving up to 95% acceptance rates and up to 25% extra inference throughput gains across mathematical reasoning, code generation, and agentic tasks. Third, we test various online MTP training strategies during RL and show that pre-RL MTP training with e2e TV loss and rejection sampling achieves a consistent acceptance rate and speedup throughout the entire RL, eliminating the need for costly online MTP updating. We provide extensive experiments and analysis that validate our findings. Experimental results show our method achieves up to 1.8x end-to-end acceleration in async RL training of Qwen3.5, Qwen3.6, and Qwen3.7 models.

Read & Discuss → View Source →
23.
arXiv (CS.AI) 2026-06-11

Search Discipline for Long-Horizon Research Agents

Authors:
Adithya Srinivasan↗Devesh Paragiri↗

arXiv:2606.11522v1 Announce Type: new Abstract: Autoresearch agents now propose, evaluate, and select scientific candidates against a metric, and that metric is usually an aggregate reduced over a heterogeneous space of regions, slices, or cohorts. We show that when scientific validity lives in that disaggregated structure, the aggregate can rank the wrong candidate first. The headline number improves while the structure underneath inverts, so a decision made on the number accepts a candidate that quietly breaks the model. The failure is not domain-specific. It appears wherever a candidate's validity is multi-dimensional but its verifier is a single reduction. We demonstrate the inversion on a fire-model task in the Ecosystem Demography model. The highest-scoring candidate and a slightly lower one are within noise of each other on global score, yet the top-scoring one collapses the protected boreal regions while the other preserves them. What separates them is the per-region behavior, not the headline number. This decision should not be left to the agent that produced the candidates. The agent optimizing the score is the last party likely to catch the score being wrong, and a prompt has no remaining turn once the agent has stopped. We move the decision to an external control loop that audits each candidate on its disaggregated behavior and acts after the agent has decided. It can demote a candidate the agent would have accepted, and it can reopen a run the agent had declared finished. Our contribution is the inversion finding itself, and a search-discipline protocol that decides on reviewable candidate-effect evidence instead of the score.

Read & Discuss → View Source →
24.
arXiv (CS.CV) 2026-06-18

Experimental Analysis of Neural Network-Based Image Classification on the CIFAR-10 Dataset

Authors:
Necati Kagan Erkek↗Emre Balci↗Berkin Halay↗

An experimental investigation of neural image classification on the CIFAR-10 benchmark is presented through fully connected and convolutional network formulations. The analysis emphasizes the complete learning pipeline: image vectorization, normalization, one-hot class encoding, supervised loss minimization, learning-rate selection, mini-batch training, convolutional feature extraction, max-pooling, and validation-based generalization assessment. A convolutional architecture with six convolutional layers and three max-pooling stages is evaluated for ten training epochs using a batch size of 128 and an Adam optimizer with a learning rate of 0.001. The validation accuracy reaches approximately 74.77%, while the validation loss begins to increase after the middle of training despite continued reduction in training loss. The resulting behavior illustrates the practical difference between representation learning and memorization, and it provides a compact experimental baseline for future studies on regularization, data augmentation, deeper architectures, and reproducible image-classification education.

Read & Discuss → View Source →
25.
arXiv (CS.CV) 2026-06-16

Akasha 2: Hamiltonian State Space Duality and Visual-Language Joint Embedding Predictive Architectur

Authors:
Yani Meziani↗

We present Akasha 2, a state-of-the-art multimodal architecture that integrates Hamiltonian State Space Duality (H-SSD) with Visual-Language Joint Embedding Predictive Architecture (VL-JEPA). The system leverages the Mamba-3 Selective State Space Model (SSM) augmented by a Sparse Mixture of Hamiltonian Experts (SMoE-HE) that enforces latent physical conservation laws through symplectic integration. For visual synthesis, we introduce Hamiltonian Flow Matching (HFM) and persistent 3D Gaussian Splatting (3DGS), enabling ultra-low latency (

Read & Discuss → View Source →