arXiv:2606.03794v1 Announce Type: new
Abstract: Graph Neural Networks (GNNs) have emerged as a powerful tool for wireless resource allocation that leverages the underlying graph structure of communication networks. Their transferability property enables models trained on small-scale graphs to generalize to large-scale deployments with little performance deterioration, a desirable property for currently growing networks. Wireless networks are sparse regimes, where a single node is connected to a small number of other users. This work establishes theoretical results for transferability of GNNs over graphs derived from sparse Random Geometric Graphs (RGGs). In particular, we focus on conflict graphs of RGGs used to model interference among links. Our approach considers the closeness between RGGs and Deterministic Grid Graphs (DGG) to establish bounds in the performance loss when a model is transferred across scales. We validate our theoretical findings through the problem of link scheduling, demonstrating that our learned policies consistently outperform existing benchmarks at scale. Finally, we examine the impact of our theoretical assumptions on empirical performance.
Science Journals
arXiv:2606.02902v1 Announce Type: new
Abstract: Deep reinforcement learning (DRL) is increasingly applied to de novo molecular design, but choices in data, rewards, and evaluation can yield uneven performance across disease areas and chemotypes. Despite this, there is no concise synthesis of how fairness is defined, measured, and tested in DRL-based drug discovery. In this rapid evidence review, we synthesize fairness definitions and metrics for DRL-driven molecule generation in healthcare. We focus on three questions: (i) how dataset composition and split strategies, especially scaffold versus random splits, affect evaluation and distribution shift; (ii) how reward design (e.g., QED, docking, toxicity, synthetic accessibility) can create or mitigate bias, with emphasis on cancer targets; and (iii) which measurable metrics best capture fairness. This includes parity across cancer versus non-cancer indications and across cancer subtypes. It also includes distributional balance in key physicochemical descriptors, scaffold/chemotype diversity, groupwise validity, toxicity, and synthetic accessibility. From 2017 onward, we searched major biomedical, computer science, and engineering literature databases and used arXiv for horizon scanning. Records were screened using PRISMA-style procedures and analyzed via content coding to link reported parity outcomes to dataset and reward choices. Our review provides a concise set of fairness definitions and metrics for DRL molecule generation. It offers practical guidance for reporting distribution parity and outcome parity. It also summarizes how dataset and reward choices relate to observed parity effects and identifies open gaps relevant to trustworthy, cancer-relevant DRL generation.
arXiv:2511.05050v3 Announce Type: replace-cross
Abstract: In this study, a scalable online kernel learning framework is proposed for estimating bidirectional causal effects in systems characterized by mutual dependence and heteroskedasticity. Traditional causal inference often focuses on unidirectional effects, overlooking the common bidirectional relationships in real-world phenomena. Building on heteroskedasticity-based identification, the proposed method integrates a quasi-maximum likelihood estimator for simultaneous equation models with large scale online kernel learning. It employs random Fourier feature approximations to flexibly model nonlinear conditional means and variances, while an adaptive online gradient descent algorithm ensures computational efficiency for streaming and high-dimensional data. Results from extensive simulations demonstrate that the proposed method achieves superior accuracy and stability than single equation and polynomial approximation baselines, exhibiting lower bias and root mean squared error across various data-generating processes. These results confirm that the proposed approach effectively captures complex bidirectional causal effects with near-linear computational scaling. By combining econometric identification with modern machine learning techniques, the proposed framework offers a practical, scalable, and theoretically grounded solution for large scale causal inference in natural/social science, policy making, business, and industrial applications.
arXiv:2606.02879v1 Announce Type: new
Abstract: Informed sampling techniques accelerate sampling-based motion planners by focusing the search on promising regions of the state space, yet most existing methods rely on Euclidean heuristics that become inadmissible under configuration-dependent Riemannian metrics. While scalar eigenvalue bounds restore admissibility by uniformly scaling the Euclidean distance, they discard the directional structure of the metric, producing overly conservative informed sets. We propose a matrix-valued admissible heuristic that exploits the Loewner order on symmetric positive definite matrices to compute the tightest constant lower bound on the metric tensor while preserving its full directional structure. The Cholesky factorization of this bound defines a linear map to an isotropic Euclidean space in which the Riemannian informed set reduces to a standard prolate hyperspheroid, enabling direct, rejection-free sampling using existing algorithms. Experiments on manipulation tasks with a 6-DoF UR5, 7-DoF Franka, and 14-DoF PR2 under three distinct Riemannian metrics show that our heuristic produces consistently tighter informed sets than both the Euclidean and scalar eigenvalue bounds, accelerating convergence across multiple state-of-the-art asymptotically optimal planners.
arXiv:2606.02876v1 Announce Type: new
Abstract: Randomized smoothing (RS) uses a smoothed classifier to provide architecture-agnostic certificates of $\ell_2$ classification robustness, but its dependence on per-input Monte Carlo (MC) sampling undermines its use in real-time systems. We argue that this cost is structural rather than fundamental, such that it can be significantly reduced by sharing information across the deployment stream. We introduce RRISE, an RS framework that compresses certification into a single forward pass through a learned surrogate. RRISE trains the surrogate against precomputed MC class-count targets via a soft-label cross-entropy loss and converts surrogate predictions into provably conservative certified radii through a one-time conformal calibration step. The resulting certificate is deployment-verifiable: whenever the calibrated radius is positive, the surrogate's prediction provably matches the smoothed classifier's and the smoothed classifier is constant on a ball of that radius around the input. Across image classification benchmarks, RRISE matches fixed-budget MC certified accuracy within $0.84$ percentage points while replacing up to $10^4$ noisy base-model evaluations per query with a single surrogate forward pass, recouping MC training cost after $\approx 10^5$ deployment queries. On CIFAR-100 and Tiny ImageNet, where the only prior offline-surrogate method collapses, RRISE achieves $1.23$ to $1.91\times$ higher certified accuracy, establishing efficient randomized smoothing as a practical path to certified robustness in repeated-deployment settings.
arXiv:2606.03629v1 Announce Type: new
Abstract: Assessing the quality of time series (TS) data is fundamental yet inherently challenging due to the multifaceted nature of quality dimensions. Recently, large language models (LLMs) have emerged as a promising paradigm for TS quality assessment via pairwise comparison and per-dimension evaluation. However, existing approaches rely on manually predefined quality dimensions and purely text-based reasoning, leaving it unknown whether LLMs can identify truly relevant quality dimensions or perform grounded and quantitative quality comparisons. To investigate this, we construct TSQBench, a dedicated benchmark for evaluating LLMs on two progressive capabilities: (i) understanding and identifying relevant quality dimensions, and (ii) performing quality comparison under specific dimensions. Our analysis reveals that current LLMs consistently struggle with both dimension identification and evidence-grounded quality comparison. To address these limitations, we propose TSQAgent, a novel agentic reasoning framework for TS quality rating consisting of three collaborative roles: Perceiver for focused dimension selection, Inspector for dimension-wise quantitative analysis, and Adjudicator that aggregates and refines the final judgment. In particular, we introduce an agentic reasoning strategy that instills the ability to identify and prioritize the most relevant quality dimensions, and further propose an agent workflow equipped with external analytical tools to enable precise quantitative comparisons over selected dimensions. Experiments on both the proposed benchmark and eleven real-world datasets demonstrate that our framework not only substantially improves LLMs' capabilities in quality understanding and quantitative comparison but also effectively translates these improvements into better quality-aware data selection, leading to enhanced downstream performance and data efficiency.
arXiv:2606.03644v1 Announce Type: new
Abstract: Comprehensive molecular profiling is essential for modern precision oncology but remains hindered by prohibitive costs, specimen exhaustion, and protracted turnaround times. While pathology foundation models (PFMs) have demonstrated potential for inferring molecular phenotypes from routine hematoxylin and eosin (H&E) whole-slide images (WSIs), current architectures primarily rely on vision-centric self-supervised learning or vision-language alignment, lacking the spatially resolved molecular supervision required to connect subtle morphological features with underlying genomic alterations. Spatial transcriptomics (ST) emerges as a transformative technology that enables transcriptomic quantification within intact tissue sections, thereby preserving the precise spatial link between histology and molecular profiles. In this study, we present a Spatial Transcriptomics-guided Alignment framework for Molecular Profiling (STAMP), which endows PFMs with intrinsic molecular awareness. To support this paradigm, we curated HumanST-1k, a human ST dataset spanning diverse anatomical organs and sequencing platforms. This atlas yields 1.8 million pairs of H&E patches and corresponding transcriptomic profiles, providing a corpus that links histological structures with their molecular states. To mitigate the technical noise inherent to raw transcriptomics, STAMP applies a pathway-informed alignment strategy that aggregates transcriptomic data into biologically functional pathways, which are subsequently integrated into PFMs via parameter-efficient fine-tuning. This alignment enriches the representation space of PFMs and unlocks their capacity to resolve sub-visual molecular signatures. The clinical utility of these augmented representations was validated through a multi-tier evaluation framework.
arXiv:2606.03638v1 Announce Type: new
Abstract: Cyclic codes with dimensions exceeding half of the code length and minimum distance greater than the square root of the code length are of significant interest due to their high transmission efficiency and strong error-correcting capability. Such codes are well suited for demanding applications, including communication and storage systems, post-quantum cryptography, radar and sonar systems, wireless sensor networks, and space communications. Motivated by the work of Ding \cite{P3}, this paper extends the binary framework of Ding and Zhou \cite{P2} to a non-binary setting. By employing power functions with known differential uniformity over finite fields of odd characteristic, we present several infinite families of $q$-ary cyclic codes of length $q^m-1$ with dimensions exceeding $(q^m-1)/2$ and the lower bounds on the minimum distances greater than the square root of the code length, thereby achieving a favorable balance between code rate and error-correcting capability. We also determine the exact minimum distance of some of these codes. Furthermore, we partially resolve Open Problem $5.31$ posed by Ding in \cite{P3}.
arXiv:2606.02884v1 Announce Type: new
Abstract: Reward guidance algorithms steer a learned generative process toward the reward-tilted measure at inference time. While empirically powerful, these methods are prone to reward hacking: the guided model over-optimizes the reward at the cost of fidelity to the learned distribution. Prior work has attributed this to the complexity of neural reward functions or implicit biases in diffusion training, but its fundamental origins remain poorly understood. We show that reward hacking arises from an approximation made in most practical implementations of reward-guided diffusion -- finite-particle plug-in estimation of the Doob h-function -- even in the simplest non-trivial settings of Gaussian and Gaussian mixture targets with quadratic rewards. In closed form, we isolate two distinct failure modes of the plug-in estimator: it leads to reward hacking within each mode and it cannot select high-reward modes. We propose a closed-form reward damping schedule that corrects the within-mode bias with no additional compute, and clarify the role of best-of-n sampling in compensating for the mode selection failure. Experiments on Gaussian mixture targets, a 2D checkerboard, and FLUX.1 text-to-image generation confirm that our theoretical insights carry over to practical settings.
arXiv:2606.02597v1 Announce Type: new
Abstract: The development of brain-computer interfaces (BCIs) based on electroencephalograms (EEGs) has advanced significantly mainly to machine learning. Although the majority of earlier research has been on increasing classification accuracy, relatively little focus has been placed on security and robustness. According to recent research, EEG-based BCIs are susceptible to adversarial attacks, which can cause misdiagnosis due to minute, well-crafted disturbances. Evaluating model robustness against such perturbations is therefore critical for ensuring reliable deployment. In this study, we propose a lightweight custom Convolutional Neural Network (CNN) architecture to investigate adversarial robustness in EEG-based BCIs. The suggested method is assessed using two EEG datasets and contrasted with three novel CNN models tailored to EEG, namely EEGNet, DeepConvNet, and SleepEEGNet, under gradient-based adversarial attack scenarios. According to experimental findings, the suggested model continuously performs better in classification under adversarial perturbations compared to baseline models, indicating improved robustness. These findings highlight the potential of lightweight architectures for enhancing the reliability of EEG-based BCI systems under adversarial conditions.
arXiv:2606.03635v1 Announce Type: new
Abstract: Understanding short online videos involves more than identifying visible objects and actions; video makers often include an underlying message or purpose in the clip. We introduce VidMsg, a benchmark for evaluating implicit message understanding in short, internet-native video clips. VidMsg contains 400 YouTube-derived clips across 9 practical topic areas and 52 fine-grained target messages, covering domains such as career and finance, education, health and well-being, culture, safety, sustainability, and lifestyle. VidMsg is constructed through a message-first pipeline: an LLM first translates target messages into indirect search scenarios, which are used to retrieve candidate clips. Human annotators then retain clips that convey the intended message without being overly explicit. VidMsg is designed primarily for bidirectional message-clip retrieval for scalable applications such as video search and recommendation, where systems must capture holistic video understanding. In addition to retrieval, VidMsg includes a diagnostic multiple-choice QA benchmark, where models select the intended message of a clip from semantically related alternatives. Experiments with contemporary video-language and retrieval models show that strong models often fail on VidMsg, because the task requires pragmatic inference, integration of contextual cues, and discrimination among semantically close messages. We also introduce VidVec-Msg, a baseline method that improves message-oriented retrieval while leaving substantial headroom for future work.
arXiv:2606.03622v1 Announce Type: new
Abstract: The use of digital tools like educational video games and interactive simulations is of great importance to physics education. This study investigates the sequencing effects of an educational game, 'Photon Jump' and PhET Photoelectric Effect simulation on students' perception of such tools for learning about the photoelectric effect. Using a counterbalanced quasi-experimental crossover design, a total of 55 physics and engineering students from a calculus-based physics course were divided into two groups of comparable sizes and administered the game and the simulation in reverse order followed by 9 open ended reflection questions. The results show a clear preference for the game-based activity over the simulation. Additionally, gaming frequency showed no correlation with willingness to use similar tools, suggesting broad accessibility. Thematic analysis revealed that intuitive and explorative learning through the game was reinforced by the analytical aspect of the simulation.
arXiv:2507.17506v4 Announce Type: replace-cross
Abstract: This work presents a cognitive radar (CR) framework designed to track multiple aircraft under unknown disturbances using massive multiple-input multiple-output (MMIMO) systems. Since uniform power allocation is suboptimal across varying signal-to-noise ratios (SNRs), we couple an adaptive waveform design driven by Partially Observable Monte Carlo Planning (POMCP). By assigning an independent POMCP tree to each target, the system efficiently predicts target states. These predictions inform a constrained optimization problem that actively directs transmit energy toward weaker targets while maintaining sufficient power for stronger ones. Results confirm that the proposed POMCP method improves the detection probability for low-SNR targets from 0.6 to nearly 0.9, and yields more accurate tracking of the weakest target than a non-adaptive orthogonal waveform or a cognitive uniform-power POMCP baseline.
arXiv:2403.13449v2 Announce Type: replace-cross
Abstract: String attractors are a combinatorial tool coming from the field of data compression. It is a set of positions within a word which captures an occurrence of every factor. While one-sided infinite words admitting a finite string attractor are eventually periodic, the situation is different for two-sided infinite words. In this article, we characterise the bi-infinite words admitting a finite string attractor as the characteristic Sturmian words and their morphic images. For words that do not admit finite string attractors, we study the structure and properties of their infinite string attractors.
arXiv:2606.00494v2 Announce Type: replace
Abstract: Post-Training Quantization (PTQ) and Low-Rank Adaptation (LoRA) constitute the standard pipeline for efficient Large Language Model (LLM) deployment. However, applying them sequentially poses a problem: PTQ often leaves behind random noise that is spread out (across the model's weights) in a way LoRA can't easily fix, meaning that LoRA ends up wasting its limited capacity trying to fix uncorrectable noise instead of improving task performance. In this paper, we propose \textbf{ProjQ}, a novel framework for constraining quantization noise to the low-rank manifold via orthogonal subspace projection. We derive an efficient alternating algorithm that shapes the quantization noise into a low-rank structure, effectively offloading dominant error components to the subsequent adapter while minimizing the residual error in the orthogonal "uncorrectable" subspace. Our theoretical analysis demonstrates that ProjQ preserves strictly greater model plasticity for downstream tasks compared to standard PTQ. Extensive experiments on LLaMA-2, Qwen2.5 and Qwen3 confirm that ProjQ consistently outperforms existing methods in both quantization error compensation and downstream task fine-tuning, achieving up to $2\times$ lower evaluation loss for compensation and matching the performance of standard 4-bit baselines on language modeling tasks with only 3 bits. The code is available on https://github.com/yy9301/ProjQ .
arXiv:2606.02585v1 Announce Type: new
Abstract: We propose a localized physics-informed kernel function neural network (LPIKFNN), which is an improved physics-informed neural network (PINN) based on physics-informed kernel function (PIKF). In the LPIKFNN framework, the localized collocation scheme discretizes the physical quantities within the local domain, where the physical field is represented as a linear combination of PIKFs. Based on this representation, the multilayer perceptron is trained to iteratively learn the physical quantities. To overcome the computational challenges of conventional PINN in higher-order derivative and high wavenumber problems, the LPIKFNN constructs the loss function using the PIKF and a localized collocation scheme rather than relying on automatic differentiation. As a result, the costly derivative evaluations required to enforce governing equations during iterative training are eliminated, leading to significantly improved computational efficiency and training performance. Moreover, incorporating PIKFs into the loss function enables the proposed LPIKFNN to significantly improve computational accuracy in high-wavenumber problems characterized by highly oscillatory physical fields. To overcome the computational bottleneck of the physics-informed kernel function neural network (PIKFNN) in heterogeneous problems, the LPIKFNN introduces a localized collocation scheme that removes reliance on global PIKFs, enabling accurate predictions where global PIKFs are unavailable. The feasibility and accuracy of the proposed LPIKFNN are demonstrated through a series of benchmark studies, including high wavenumber problems, higher-order derivative problems, nonlinear problems, heterogeneous problems, and potential-based inverse electromyography. The numerical predictions obtained by LPIKFNN show excellent agreement with available analytical solutions and experimental measurements.
arXiv:2606.02581v1 Announce Type: new
Abstract: Retrieval-augmented generation (RAG) faces a fundamental three-way tension: deeper retrieval improves factual grounding but inflates token costs and end-to-end latency. Static retrieval configurations cannot resolve this tension across heterogeneous query workloads -- simple definitional queries waste budget on unnecessary context, while complex analytical prompts are underserved by shallow retrieval. This paper introduces \emph{Cost-Aware RAG} (CA-RAG), a per-query routing framework that selects from a discrete catalog of \emph{strategy bundles} -- each coupling a retrieval depth (from retrieval-free direct inference to top-$k{=}10$ dense retrieval) with a fixed generation profile -- by maximizing a scalar utility that linearly combines an estimated quality prior with normalized penalties for predicted latency and total billed tokens. CA-RAG is implemented with FAISS-backed dense retrieval and OpenAI chat/embedding APIs, and evaluated on a 28-query benchmark spanning four bundles. The router dynamically exercises all bundles, achieving \textbf{26\% fewer billed tokens} than always-heavy retrieval and \textbf{34\% lower mean latency} than always-direct inference while maintaining equivalent answer quality. Per-query delta analysis reveals that savings are non-uniform and concentrated in simpler queries, motivating complexity-aware guardrails. Sensitivity analysis confirms that the same bundle catalog supports multiple cost-latency-quality operating points through weight adjustment alone. All results are generated directly from logged CSV artifacts for full reproducibility. CA-RAG provides a transparent, auditable foundation for cost-conscious LLM deployments.
arXiv:2602.12430v4 Announce Type: replace
Abstract: The transition from monolithic language models to modular, skill-equipped agents marks a defining shift in how large language models (LLMs) are deployed in practice. Rather than encoding all procedural knowledge within model weights, agent skills -- composable packages of instructions, code, and resources that agents load on demand -- enable dynamic capability extension without retraining. It is formalized in a paradigm of progressive disclosure, portable skill definitions, and integration with the Model Context Protocol (MCP). This survey provides a comprehensive treatment of the agent skills landscape, as it has rapidly evolved during the last few months. We organize the field along four axes: (i) architectural foundations, examining the SKILL$.$md specification, progressive context loading, and the complementary roles of skills and MCP; (ii) skill acquisition, covering reinforcement learning with skill libraries, autonomous skill discovery (SEAgent), and compositional skill synthesis; (iii) deployment at scale, including the computer-use agent (CUA) stack, GUI grounding advances, and benchmark progress on OSWorld and SWE-bench; and (iv) security, where recent empirical analyses reveal that 26.1% of community-contributed skills contain vulnerabilities, motivating our proposed Skill Trust and Lifecycle Governance Framework -- a four-tier, gate-based permission model that maps skill provenance to graduated deployment capabilities. We identify seven open challenges -- from cross-platform skill portability to capability-based permission models -- and propose a research agenda for realizing trustworthy, self-improving skill ecosystems. Unlike prior surveys that broadly cover LLM agents or tool use, this work focuses specifically on the emerging skill abstraction layer and its implications for the next generation of agentic systems. Project repo: https://github.com/scienceaix/agentskills
arXiv:2606.02872v1 Announce Type: new
Abstract: This paper considers the problem of simultaneously controlling an interceptor's impact time and impact angle using its lateral acceleration as the sole control input. With a single control input, the nonlinear engagement kinematics is inherently underactuated, which complicates guidance law synthesis. To overcome this challenge, a hierarchical sliding mode-based guidance law is developed to concurrently regulate the two terminal constraints. The proposed architecture consists of a two-layer sliding manifold. The first layer comprises two sub-sliding surfaces corresponding to the impact time and impact angle error dynamics, respectively, while the second layer introduces a composite sliding manifold that combines the two individual sub-surfaces. Then, a variable-gain adaptive guidance law is designed to ensure time and angle-constrained interception against a stationary target, which is further extended to intercept a constant velocity target. Simulations are conducted for various engagement scenarios to attest to the efficacy of the proposed approach.
arXiv:2606.03486v1 Announce Type: new
Abstract: Large language models remain vulnerable to jailbreak attacks that hide harmful intent behind seemingly ordinary requests such as role-play, translation, encoding, adversarial suffixes, and multi-turn buildup. Existing defenses still struggle to handle these attacks without over-blocking benign but sensitive requests, partly because they often apply the same action to every prompt and therefore fail to balance safety and helpfulness. We propose NeuroArmor, a white-box runtime defense that uses prompt-specific safe variants as a local safety reference for deciding when intervention is needed and, once triggered, as safe targets for intervention. For each prompt, NeuroArmor builds K safe variants, compares the prompt state against this local safe reference in hidden-state space, and routes anomalies either to a refusal branch for malicious prompts or to a helpful recovery branch for borderline benign prompts. On Llama-3-8B-Instruct, NeuroArmor reduces malicious attack success rate (ASR) from 41.56% to 1.57% while lowering benign false positive rate (FPR) on the shared benign pool from 30.26% to 22.05%; matched baselines remain substantially weaker on this trade-off. External-judge and manual behavioral evaluations further show that the remaining non-blocked outputs are much less likely to be operationally harmful. Overall, NeuroArmor provides a more effective runtime strategy for jailbreak defense by combining prompt-specific consistency checking, routing, and selective intervention.
arXiv:2509.08726v3 Announce Type: replace-cross
Abstract: This paper focuses on the decentralized stochastic optimization problem $f(\mathbf{x})=\frac{1}{m}\sum_{i=1}^m f_i(\mathbf{x})$ over a connected network of $n$ agents, where each local function has the form of $f_i(\mathbf{x}) = {\mathbb E}\left[F(\mathbf{x};{\boldsymbol \xi}_i)\right]$ which satisfies the $(L_0,L_1)$-smooth condition but possibly nonconvex and each random variable ${\boldsymbol \xi}_i$ follows distribution ${\mathcal D}_i$. We propose a novel algorithm called decentralized normalized stochastic gradient descent (DNSGD), which can achieve an $\epsilon$-stationary point at each local agent. We present a new framework for analyzing decentralized first-order methods in the $(L_0,L_1)$-smooth setting, based on the Lyapunov function related to the product of the gradient norm and the consensus error. We show that the proposed algorithm attains the upper bounds on the sample complexity of ${\mathcal O}(m^{-1}(L_f\sigma^2\Delta_f\epsilon^{-4} + \sigma^2\epsilon^{-2} + L_f^{-2}L_1^3\sigma^2\Delta_f\epsilon^{-1} + L_f^{-2}L_1^2\sigma^2))$ per agent and the communication complexity of $\tilde{\mathcal O}((L_f\epsilon^{-2} + L_1\epsilon^{-1})\gamma^{-1/2}\Delta_f)$, where $L_f=L_0 +L_1\zeta$, $\sigma^2$ is the variance of the stochastic gradient, $\Delta_f$ is the initial optimal function value gap, $\gamma$ is the spectral gap of the network, and $\zeta$ is the degree of the gradient dissimilarity. In the special case of $L_1=0$, the above results (nearly) match the lower bounds of decentralized stochastic nonconvex optimization under the standard smoothness. We also conduct numerical experiments to show the empirical superiority of our method.
arXiv:2606.03768v1 Announce Type: new
Abstract: Extended chain-of-thought (CoT) traces improve LLM reasoning but incur substantial computational and memory costs. While existing CoT compression methods mitigate this by condensing thought steps into compact representations via memory tokens and retaining only these representations at inference time, the loss of fine-grained information makes subsequent steps more error-prone. To alleviate this, we propose \textbf{HybridThinker}, where in addition to preserved these representations, thought steps are also temporarily retained to provide fine-grained details. However, we observe that naively keeping thought steps accessible to subsequent steps \emph{during training} lets the model bypass memory tokens by retrieving information directly from these steps, leaving the model's ability to compress and retrieve information through memory tokens insufficiently trained. We therefore introduce a hybrid training scheme, in which only some thought steps are directly accessible through attention to subsequent steps, while the other thought steps are masked, forcing the model to use memory tokens for compression and retrieval. Across 4 reasoning benchmarks, HybridThinker matches the uncompressed baseline, advancing the state of the art in CoT compression by 5.8 points on average accuracy with similar inference time. Ablation studies confirm that both temporary thought-step retention and the hybrid training scheme contribute to these gains.
arXiv:2606.03485v1 Announce Type: new
Abstract: Visual attention is central to ensemble coordination, yet how musicians allocate gaze during naturalistic rehearsal remains poorly understood. We present a pilot study using mobile eye tracking to examine gaze behaviour in a four-member band across three songs, each practiced twice. Musicians wore Pupil Labs Neon eye trackers, and YOLOv8-assisted scene annotations mapped fixations to ensemble members and objects in view. Analyzing fixation matrices, transition matrices, temporal scarf plots, and dwell-transition correlations, we uncover a hub-and-spoke attention topology: the session leader was the dominant gaze target for all members, while the learning guitarist concentrated up to 97% of interpersonal dwell on this single reference. Between attempts, gaze transitions decreased by up to 65% on average for unfamiliar material (up to 82% for individual participants) as scanning stabilized. Scarf plots reveal how teaching breakdowns fragment attention and uninterrupted runs consolidate it. Post-session participant reflections align with the quantitative patterns, and we discuss implications for gaze-aware tools in ensemble pedagogy.
arXiv:2606.02871v1 Announce Type: new
Abstract: Large reasoning models improve performance by generating extended chain-of-thought (CoT) reasoning, but this behavior becomes inefficient when applied to LLM agents. Current LLM agents often generate verbose textual reasoning at every decision step and allocate reasoning effort nearly uniformly across turns, leading to substantial inefficiency in multi-turn agentic trajectories. We propose Adaptive Latent Agentic Reasoning (ALAR), a dual-mode framework that uses compact latent reasoning for routine turns and selectively escalates to explicit chain-of-thought when deeper deliberation is needed. ALAR learns latent reasoning by using the agent's actions as supervision anchors and is further optimized to use latent reasoning when it is sufficient for task success and reserve explicit CoT for harder decisions. Experiments on agentic search and tool-use benchmarks show that ALAR maintains comparable or better task accuracy while substantially reducing generated tokens by up to 43.6% in search and 84.6% in tool use. These results demonstrate that ALAR improves the accuracy-efficiency trade-off of LLM agents by reducing unnecessary textual reasoning while preserving explicit deliberation for harder decision steps.
arXiv:2606.00486v2 Announce Type: replace
Abstract: GPU workloads with large memory footprints frequently suffer from redundant L2 TLB misses in which a recently evicted translation is immediately re-walked at full page-walk cost. We characterize these dead-entry misses across 24 GPU workloads, finding they account for up to 99% of L2 TLB misses in the most TLB-sensitive applications, yet their performance impact varies widely depending on memory access structure. Workloads where warps share the same virtual page suffer from burst amplification, where a single eviction stalls many warps simultaneously waiting for one translation to return. In contrast, workloads where each warp accesses a distinct set of pages face a capacity-overflow problem that no replacement policy can resolve, a distinction validated by huge page experiments. Building on this two-class taxonomy, we design DEPOT (Dead-Entry PrOTection), a 1 KB Bloom filter mechanism that prevents recently evicted translations from being displaced immediately upon reinstallation, delivering up to 72% IPC improvement on interference-driven workloads with zero overhead on others, and composing with the state-of-the-art TLB prefetching and compaction mechanism, for 2 to 7% additional gain.