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.

01.
arXiv (quant-ph) 2026-06-24

A Universal All-Fiber Quantum Buffer for the Telecom Band

arXiv:2606.24681v1 Announce Type: new Abstract: The realization of a scalable quantum internet relies on the ability to temporally align asynchronous photonic signals through on-demand buffering. While matter-based quantum memories achieve long storage times, their extremely narrow bandwidths and cryogenic requirements pose significant barriers to integration with existing telecommunications infrastructure. Conversely, current all-optical memories operate at room temperature but are hampered by high input/output losses and a lack of universality across different photonic degrees of freedom. Here, we demonstrate a universal, fully fiber-integrated quantum buffer operating over the full telecom C-band that overcomes these fundamental trade-offs. By implementing an actively switched dual-Sagnac cavity driven by cross-phase modulation, we achieve an ultra-low input/output loss of 0.46 dB and a storage time exceeding 18 $\mu$s. The device exhibits an operational bandwidth exceeding 12.5 THz ($\sim$100 nm), covering the full telecom C-band. We show the simultaneous buffering of over 200 temporal modes with the ability to address them either collectively or one by one. We demonstrate high-fidelity storage for all three degrees of freedom compatible with optical fiber propagation, namely time-bin, frequency-bin, and polarization qubits, along with faithful preservation of entanglement, confirming the platform's true universality. These results provide a robust, room-temperature solution for the high-rate synchronization of multidimensional quantum states, clearing a major hurdle for the deployment of global photonic quantum networks.

02.
arXiv (CS.AI) 2026-06-15

Low-Burden LLM-Based Preference Learning: Personalizing Assistive Robots from Natural Language Feedback for Users with Paralysis

arXiv:2604.01463v2 Announce Type: replace-cross Abstract: Physically Assistive Robots require personalized behaviors to ensure user safety and comfort. However, traditional preference learning methods, like exhaustive pairwise comparisons, cause substantial physical and cognitive fatigue for users with severe motor impairments. To solve this, we propose a low-burden, offline framework that translates unstructured natural language feedback directly into deterministic robotic control policies. To safely bridge the gap between ambiguous human speech and robotic code, our pipeline uses Large Language Models (LLMs) grounded in the Occupational Therapy Practice Framework. This clinical reasoning decodes subjective user reactions into explicit physical and psychological needs, which are then mapped into transparent decision trees. Before deployment, an automated "LLM-as-a-Judge" verifies the code's structural safety. We validated this system in a simulated meal preparation study with 10 adults with paralysis. Results show our natural language approach significantly reduces user workload compared to traditional baselines. Additionally, occupational therapists confirmed the generated policies are safe and accurately reflect user preferences.

03.
arXiv (CS.CV) 2026-06-12

BrainDINO: A Brain MRI Foundation Model for Generalizable Clinical Representation Learning

Brain MRI underpins a wide range of neuroscientific and clinical applications, yet most learning-based methods remain task-specific and require substantial labeled data. Here we show that a single self-supervised representation can generalize across heterogeneous brain MRI endpoints. We trained BrainDINO, a self-distilled foundation model, on approximately 6.6 million unlabeled axial slices from 20 datasets encompassing broad variation in population, disease, and acquisition setting. Using a frozen encoder with lightweight task heads, BrainDINO supported transfer across tumor segmentation, neurodegenerative and neurodevelopmental conditions classification, brain age estimation, post-stroke temporal prediction, molecular status prediction, MRI sequence classification, and survival modeling. Across tasks and supervision regimes, BrainDINO consistently equaled or exceeded natural-image and MRI-specific self-supervised baselines, with particularly strong advantages under label scarcity. Representation analyses further showed anatomically organized and pathology-sensitive feature structure in the absence of task-specific supervision. Our findings indicate that large-scale slice-wise self-supervised learning can yield a unified brain MRI representation that supports diverse neuroimaging tasks without volumetric pretraining or full-network fine-tuning, establishing a scalable foundation for robust and data-efficient brain imaging analysis. Code is available at https://github.com/mclwu22/BrainDINO

04.
arXiv (CS.CL) 2026-06-19

Multi-Agent Transactive Memory

The decentralized deployment of LLM agents with diverse capabilities across diverse tasks motivates infrastructure for knowledge sharing across heterogeneous agent populations. Just as search engines index human-generated artifacts to support human problem solving, retrieval systems can organize agent-generated artifacts for reuse across agent populations. We extend retrieval-augmented generation - which demonstrates the value of human-authored artifacts to individual agents - to retrieval of agent-generated artifacts supporting a population of agents. In particular, agent trajectories encode reusable procedural knowledge, yet these artifacts are typically discarded after a single use or retained only by the producing agent, forcing newly instantiated agents to repeatedly rediscover existing solutions. We propose Multi-Agent Transactive Memory (MATM), a framework for population-level storage and retrieval of agent-generated trajectories, where producer agents contribute trajectories to a shared repository and consumer agents retrieve them to improve task execution. We focus on interactive environments (ALFWorld and WebArena), where trajectories are long and encode especially rich procedural structure. Our experiments demonstrate that retrieving trajectories from MATM improves downstream task performance and reduces interaction steps without coordination or joint training. These results position MATM as a design pattern for population-level experience sharing in open agent ecosystems.

05.
arXiv (quant-ph) 2026-06-19

Effects of interaction range on the mean-field dynamics of Bose polarons

arXiv:2606.20020v1 Announce Type: cross Abstract: We consider the three-dimensional Bose polaron problem in the regime of finite range interactions and competing length scales. Working in the reference frame of the impurity, we study both static and out of equilibrium properties of the system, in particular the transfer of momentum between the impurity and the host gas. We find that relaxation dynamics can occur via damped oscillations of the impurity velocity with simple dependence on the interaction strength. Furthermore, the equilibration process is sensitive to the type of the impurity-bath interaction. Specifically, interatomic forces describing ion-atom systems lead to much longer timescales and more pronounced oscillations in the strong coupling regime with respect to local interaction potentials. We also find that the effective masses can differ by a large amount between the two scenarios, even if the number of atoms in the polaron cloud remains similar for both cases.

06.
medRxiv (Medicine) 2026-06-15

Primary care practitioners preconception health literacy and information-seeking: A cross-sectional survey.

Background Parental health before pregnancy influences maternal and child outcomes. Primary care professionals, including general practitioners [GPs], midwives, and naturopaths, can provide preconception care, yet many report limited knowledge and difficulty accessing relevant information. This study described Australian GPs, midwives, and naturopaths preconception health literacy, including knowledge and ability to access information. Methods Between July and September 2022, Australian GPs, midwives, and naturopaths completed a 32-item online cross-sectional survey. Participants were recruited through professional associations, and data were analysed using descriptive and inferential statistics Results Participants (N=373) included naturopaths (40.7%), GPs (32.4%), and midwives (26.8%). Reported barriers to clinician health literacy including lack of preconception care resources (25.5%), and limited clinician knowledge (23.6%). The proportion identifying limited clinician knowledge differed significantly between professions (GP: 31.4%; midwives: 23.0%; naturopaths: 17.8%; p=0.030). The highest level of accurate knowledge regarding preconception exposures was for pre-pregnancy obesity (82.7%), while low birth weight was the most accurately identified preconception outcomes (83.7%). Incorrect responses were most common for maternal multivitamin use as an exposure (28.3%) and childhood leukaemia as an outcome (26.3%). Differences between professions were strongest for infant outcomes, with moderate associations observed for shoulder dystocia (V=.2355), precipitous labour (V=.2173), macrosomia (V=.2060), labour dystocia (V=.2018) and cryptorchidism (V=.2018). Discussion Preconception health literacy varies across primary care professions. Clinicians require greater access to targeted resources and education tailored to their differing scopes of practice and experience. Improving clinician preconception health literacy may strengthen consistent evidence-based care and support better maternal, child, and long-term family health outcomes.

07.
arXiv (CS.AI) 2026-06-16

On-Policy Distillation with Curriculum Turn-level Guidance for Multi-turn Agents

arXiv:2606.15912v1 Announce Type: cross Abstract: Multi-turn agents that plan, invoke tools, and interact with environments offer a promising paradigm for solving complex tasks, yet their capabilities typically rely on very large models whose inference cost is prohibitive in practice.On-Policy Distillation (OPD) is a natural recipe for transferring such capabilities to smaller students, but we find that it suffers a characteristic failure mode in this setting: small student errors compound across turns and push the trajectory out of the teacher's familiar state distribution, so the teacher's supervision becomes least reliable precisely where the student needs it most.We propose Guided On-Policy Distillation (Guided-OPD), a simple yet effective algorithm that mixes teacher- and student-generated turns within each rollout and schedules the teacher's intervention probability along a curriculum that decays to zero.Strong guidance keeps early trajectories close to the teacher distribution and is then gradually withdrawn to recover the purely on-policy regime used at inference.On ALFWorld, ScienceWorld, and WebShop, distilling Qwen3 students from a Qwen3-30B-A3B teacher, Guided-OPD improves Score by 21.1\% and Success Rate by 25.5\% over vanilla OPD on average, with larger gains on smaller students.

08.
PLOS Computational Biology 2026-06-02

Data-driven model reveals increased stability of CAG-expanded <i>huntingtin</i> RNA due to MID1 binding

Authors:

by Yuhong Liu, Annika Reisbitzer, Domagoj Dorešić, Jan Hasenauer, Sybille Krauß, Tatjana Tchumatchenko RNA-binding proteins (RBP) are important regulators of RNA metabolism. In neurodegenerative disorders such as Huntington’s Disease (HD), disrupted RBP-RNA interactions contribute to neuronal dysfunction. One such RBP, Midline 1 (MID1), has been shown to aberrantly associate with mutant huntingtin (Htt) RNA, enhancing its translation, yet the mechanism driving this effect remains unknown. Here, we develop a computational model to understand the role of MID1. Based on previously published data, our model predicts that MID1 increases the stability of the Htt RNA. We experimentally validate this prediction, showing that overexpression of MID1 significantly prolongs the half-life of mutant Htt RNA. Furthermore, we evaluate model refinements, including clustering of MID1-bound RNA, which allow capturing all key observations in the data. Together, we provide a data-driven framework that underlines the importance of RBP-RNA interaction in post-transcriptional regulation. This framework also shows how individual molecular reactions jointly determine RNA stability and protein levels in HD.

09.
arXiv (CS.AI) 2026-06-15

tap: A File-Based Protocol for Heterogeneous LLM Agent Collaboration

Authors:

arXiv:2606.14445v1 Announce Type: cross Abstract: Existing multi-agent software development systems have proposed many forms of agent collaboration, including role-based collaboration and automated code review. However, many systems assume a common runtime, a central conversation server, or the same API family. Under these assumptions, LLM agents from different vendors cannot easily exchange messages directly from their own execution environments while dividing development and review work on a shared codebase. This paper presents tap, a file-based collaboration protocol that allows Claude (Anthropic) and Codex (OpenAI) to collaborate on one codebase without shared memory or an identical runtime. The core of tap is a file-first design that preserves markdown files with metadata as original messages, combines a file inspection path (file communication, Tier 1) with real-time notification paths for Claude and Codex (real-time communication, Tier 2), and isolates work through separate git worktrees. Even if real-time notification fails or a receiver restarts, the message file remains available and the same content can be inspected again. In a 27-day, 37-generation self-applied operation where tap was used to develop and review itself, we collected 209 tap-related pull requests and 717 operational artifacts. An analysis of 375 review artifacts showed that the share of reviews recording at least one defect or requested change was 69.8% for heterogeneous model pairs and 53.1% for homogeneous model pairs. These results show that tap, which combines file-based message preservation with real-time notification, operates in a real production repository, and that combining heterogeneous models and execution environments can broaden review perspectives. tap is distributed as the open-source npm package @hua-labs/tap (v0.5.2).

10.
medRxiv (Medicine) 2026-06-23

What Is the Optimal Timing and Frequency of Workload-Matched Postprandial Physical Activity Breaks? A Randomized Controlled Crossover Study of Cardiometabolic and Cognitive Responses During Sedentary Behavior

Purpose Postprandial sedentary behavior is associated with negative health effects and constitutes a large part of daily life in modern society. This study investigated how the timing of physical activity after eating influences glucose levels, cerebral and muscle oxygenation, cognitive performance, and well-being during subsequent sitting. Methods In a four-armed randomized crossover trial, healthy adults consumed four standardized meals separated by 48-hour washout periods. Each meal was followed by 2 hours of sitting combined, in random order, with one of four interventions: (1) sitting only, (2) 15 minutes of moderate intensity cycling immediately after eating, (3) 15 minutes of cycling 20 minutes after eating, or (4) three workload-matched five-minute cycling bouts during sitting. Interstitial glucose (continuous glucose monitoring), cerebral and muscle oxygenation (Functional near infrared spectroscopy), cognitive performance (Stroop test), heart rate, blood pressure, and subjective ratings were assessed every 30 minutes. Data were analyzed using repeated-measures ANOVA. Results Twenty participants (mean age 27.1{+/-}10.3 years, 12 females) completed the study. Cycling immediately after eating reduced mean glucose levels during postprandial sitting, while both 15-minute cycling bouts increased cerebral oxygenation. All active conditions enhanced muscle oxygenation. Heart rate and arousal increased with delayed cycling and active breaks. No effects were observed for blood pressure, cognitive performance, focus, or well-being. Conclusion A short bout of physical activity immediately after eating reduces postprandial hyperglycemia and improves brain oxygenation during sitting, whereas delayed activity and brief breaks increase physiological activation without cognitive or perceptual benefits.

11.
arXiv (CS.AI) 2026-06-18

scGTN: Deep Siamese Graph Transformer Network for Single-cell RNA Sequencing Clustering

arXiv:2606.18672v1 Announce Type: cross Abstract: Single-cell RNA sequencing (scRNA-seq) serves a pivotal role in characterizing gene expression at the cellular level, enabling the identification of cell types and advancing the understanding of cellular heterogeneity. Despite the significant progress in scRNA-seq data clustering, we argue that current methods always ignore the sparsity and noise, as well as the complex intercellular structural information inherent in scRNA-seq data. Toward this end, in this paper, we propose a novel single-cell RNA-seq clustering framework via deep Siamese Graph Transformer Network (termed scGTN), which explicitly integrates gene expression profile and intercellular structural dependencies for cell clustering. In particular, we formulate scRNA-seq data as a graph and construct two augmented graph views that serve as dual views to capture complementary intercellular information. Then, a Siamese graph transformer network is employed to explicitly incorporate shortest-path information and node-wise distances for capturing richer structural relationships between cells. Finally, we employ an optimal transport strategy to guide the cell clustering in a self-supervised manner. Extensive experiments on multiple benchmark scRNA-seq datasets demonstrate that our scGTN consistently outperforms existing methods. Our code is available at https://github.com/W-RMSL/scGTN.

12.
arXiv (CS.LG) 2026-06-17

Overcoming the Incentive Collapse Paradox

arXiv:2603.27049v2 Announce Type: replace-cross Abstract: AI-assisted task delegation is increasingly common, yet human effort in such systems is costly and typically unobserved. Recent work by Bastani and Cachon (2025); Sambasivan et al. (2021) shows that accuracy-based payment schemes suffer from incentive collapse: as AI accuracy improves, sustaining positive human effort requires unbounded payments. We study this phenomenon in a budget-constrained principal-agent framework with strategic human agents whose output accuracy depends on unobserved effort. Our first contribution is a general impossibility result showing that incentive collapse is not merely a limitation of simple linear payments, but arises for any payment rule based only on observed task accuracy.To overcome this barrier, we propose a sentinel-auditing payment mechanism that enforces a strictly positive and controllable level of human effort at finite cost, independent of AI accuracy. Building on this incentive-robust foundation, we develop an incentive-aware active statistical inference framework that jointly optimizes (i) the auditing rate and (ii) active sampling and budget allocation across tasks of varying difficulty to minimize the final statistical loss under a single budget. Experiments demonstrate improved cost-error tradeoffs relative to standard active learning and auditing-only baselines.

13.
arXiv (CS.LG) 2026-06-11

Visual-Redundancy-Controlled Parallel Decoding for Diffusion-Based Multimodal Large Language Models

arXiv:2605.25820v2 Announce Type: replace Abstract: Diffusion-based multimodal large language models (dMLLMs) decode by iteratively predicting tokens at multiple masked positions in parallel. This turns each decoding step into a position-selection problem: the model must choose not only which predictions are reliable in isolation, but also which positions should be committed together as context for later decoding steps. Existing confidence-based decoding ranks masked positions independently and commits the top-K positions, largely ignoring whether the committed tokens provide complementary visual grounding. We identify a step-level limitation of this strategy in multimodal settings: high-confidence tokens selected in the same step can rely on overlapping visual grounding, introducing visual redundancy among the committed tokens and leaving less complementary visual grounding available for later decoding. To quantify this effect, we introduce the Visual Redundancy Index (VRI), which measures visual grounding overlap among tokens committed in parallel. To control this redundancy during decoding, we propose Visual-Redundancy-Controlled Decoding (VRCD), a training-free inference-time decoding method that uses token-to-image attention to prioritize visually complementary positions. Across diverse multimodal benchmarks, VRCD reduces visual redundancy and remaining-position entropy with modest runtime overhead. In longer decoding experiments, it also achieves relative accuracy gains of up to 18.8% on M^3CoT and 6.9% on MMBench over confidence-based decoding. Code is available at https://github.com/infiniteYuanyl/VRCD.

14.
arXiv (quant-ph) 2026-06-12

Proper and improper mixed states serve as different prior beliefs for quantum state retrodiction

arXiv:2502.10030v2 Announce Type: replace Abstract: A mixed quantum state can be taken as capturing an unspecified form of ignorance; or as describing the lack of knowledge about the true pure state of the system ("proper mixture"); or as arising from entanglement with another system that has been disregarded ("improper mixture"). These different views yield identical density matrices and therefore identical predictions for future measurements. But when used as prior beliefs for inferring the past state from later observations ("retrodiction"), they lead to different updated beliefs. This is a purely quantum feature of Bayesian agency. Based on this observation, we establish a framework for retrodicting on any quantum belief and we prove a necessary and sufficient condition for the equivalence of beliefs. We also illustrate how these differences have operational consequences in quantum state recovery.

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

Link-Free Multi-Node Timing Synchronization for Scalable Quantum Networking

arXiv:2606.14077v1 Announce Type: new Abstract: Precise timing synchronization is essential for distributed quantum networking, enabling entanglement distribution, quantum teleportation, and entanglement swapping across remote nodes. Existing synchronization architectures rely on dedicated timing-distribution infrastructure, most notably White Rabbit networks, which constrain topology, scalability, and deployment in free-space and satellite environments. Here we demonstrate link-free synchronization of quantum network nodes using independently operating miniature rubidium atomic clocks and computational post-processing. We validate the approach on a deployed metropolitan-scale telecom fiber network spanning three geographically separated nodes. Following drift correction, atomic-clock-based synchronization achieves timing performance approaching that of a White Rabbit benchmark and remains stable over continuous 8-hour operation. As a stringent test of quantum-network functionality, we observe Hong-Ou-Mandel interference across spatially separated nodes with visibility exceeding 70%, statistically equivalent to that obtained using dedicated White Rabbit timing links. To the best of our knowledge, this represents the first observation of quantum interference across a deployed metropolitan-scale telecom fiber network synchronized entirely without dedicated timing-transfer infrastructure. These results establish atomic-clock-based synchronization as a scalable, topology-independent alternative to conventional timing-distribution architectures and a practical pathway toward terrestrial, airborne, and space-based quantum networks where dedicated timing links are unavailable.

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

A Computational Audit of Demographic Association Encoding in ClinicalBERT Language Predictions

Transformer-based clinical language models are increasingly integrated into high-stakes clinical decision support pipelines, yet the computational mechanisms through which demographic associations encoded in medical documentation propagate into model probability distributions remain empirically underspecified. We present a systematic computational audit of representational bias in ClinicalBERT (Alsentzer et al., 2019), a BERT-based model pretrained on MIMIC-III discharge summaries, employing two complementary probing methodologies: Log Probability Bias Analysis (LPBA), which quantifies demographic descriptor-induced shifts in masked token probability distributions across behavioral and evaluative semantic categories, and Masked Language Model-based analysis (MLM), which probes internal representational structure for demographic agency attribution encoding across 98 real clinical sentence templates and eight intersectional race-gender combinations. Corpus frequency analysis operationalizes the distinction between statistical disparity and bias amplification by benchmarking model outputs against empirical term frequencies in the MIMIC-III training corpus. Of 32 statistically significant findings, 65.6% contradict observed corpus distributions, rising to 80% for Black patients and 87.5% for agency attribution under MLM probing, providing direct empirical evidence that representational bias in ClinicalBERT operates predominantly through model-internal amplification rather than training data inheritance. Keywords: natural language processing, clinical documentation, algorithmic auditing, representational bias, health equity 1

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

daVinci-kernel: Co-Evolving Skill Selection, Summarization, and Utilization via RL for GPU Kernel Optimization

GPU kernel optimization represents a paradigm where functional correctness is assumed and execution efficiency is the objective. We present daVinci-kernel, a reinforcement learning framework that couples skill discovery with skill exploitation through a dynamically evolving skill library. daVinci-kernel jointly trains three agents sharing one LLM backbone: a Skill Selection Agent that retrieves relevant techniques via BM25 and LLM reranking, a Policy Agent that generates multi-turn CUDA/Triton kernels conditioned on selected skills, and a Skill Summary Agent that distills successful rollouts into reusable skills. Candidate skills are added only after execution-based verification confirms reproducible speedups. All three agents share a single LLM backbone, are initialized via a structured SFT cold start on diversity-filtered data, and are then jointly optimized end-to-end with multi-turn REINFORCE and per-agent advantage estimation. On KernelBench, daVinci-kernel-14B achieves 37.2%, 70.6%, and 32.2% on Level 1, Level 2, and Level 3 under the Fast$_1$ threshold, outperforming the strongest prior RL-trained model, Dr.Kernel-14B.

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

Prompt Perturbation for Reliable LLM Evaluation over Comparison Graphs

Evaluating large language models (LLMs) is important for understanding their capabilities, comparing competing systems, and supporting the deployment of reliable models in practice. For open-ended tasks, pairwise evaluation has become a popular paradigm, in which two responses to the same prompt are compared and the resulting judgments are aggregated into an overall ranking. A central challenge of this paradigm is intransitivity: the induced comparison outcomes may fail to support any coherent global ranking. For example, one may observe cyclic preferences such as $A \succ B \succ C \succ A$, or inconsistencies involving ties such as $A \equiv B\equiv C\neq A$. Such contradictions make the resulting leaderboard unstable and challenging to interpret. In this paper, we propose a prompt perturbation framework for improving the consistency of pairwise LLM evaluation. Our approach generates perturbed variants of each prompt, uses the resulting comparison graphs to identify and filter out structurally inconsistent comparison patterns, and then applies standard ranking methods to the filtered comparisons. A key feature of the proposed framework is that graph-level structural consistency is incorporated explicitly into the evaluation pipeline before ranking aggregation. This provides a simple and principled way to reduce cyclic inconsistencies and improve the reliability of LLM rankings.

19.
arXiv (CS.CL) 2026-06-16

DRA-GRPO: Your GRPO Needs to Know Diverse Reasoning Paths for Mathematical Reasoning

Post-training LLMs with Reinforcement Learning, specifically Group Relative Policy Optimization (GRPO), has emerged as a paradigm for enhancing mathematical reasoning. However, standard GRPO relies on scalar correctness rewards that are often non-injective with respect to semantic content: distinct reasoning paths receive identical rewards. This leads to a Diversity-Quality Inconsistency, where the policy collapses into a narrow set of dominant modes while ignoring equally valid but structurally novel strategies. To bridge this gap, we propose Diversity-aware Reward Adjustment (DRA), a theoretically grounded framework that calibrates the reward signal using the semantic density of sampled groups. By leveraging Submodular Mutual Information (SMI), DRA implements an Inverse Propensity Scoring (IPS) mechanism that effectively de-biases the gradient estimation. This creates a repulsive force against redundancy, driving the policy to achieve better coverage of the high-reward landscape. Our method is plug-and-play and integrates seamlessly with GRPO variants. Empirical evaluations on five math benchmarks demonstrate that DRA-GRPO consistently outperforms strong baselines, achieving an average accuracy of 58.2% on DeepSeek-R1-Distill-Qwen-1.5B with only 7,000 training samples and $55 cost, highlighting the critical role of diversity calibration in data-efficient alignment. The code is available at https://github.com/xiwenc1/DRA-GRPO.

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

Auditing Reward Hackability in Code RL Training Environments

arXiv:2606.16062v1 Announce Type: new Abstract: We measure the rate at which code RL environments accept incorrect solutions as correct. On a 49-task sample of SWE-bench Verified, 28.5% of tasks have test suites weak enough that a Docker-verified incorrect patch passes them. On 20 R2E-Gym tasks across 6 repositories, the same pipeline at single-shot exploit generation yields 25.0%. A random-effects meta-analysis over 134 frontier model submissions to SWE-bench Verified finds, within the same human-rated difficulty stratum, model Pass@1 is +14.14 percentage points higher on flagged-hackable tasks than on robust ones (95% CI [+11.80, +16.48]; one-sided p < 10^-6; I^2 = 0%; 123 of 134 models positive). We then describe a procedure for hardening the broken tasks. An inline LLM judge with a Docker gold-sanity gate runs each generated test against the gold solution before the judge is consulted. On the 11 broken tasks in the audit, the gate flags 65 of 105 decisive LLM-generated tests as failing on the gold patch itself, a 61.9% per-augmentation defect rate the LLM judge alone misses. With diversity-biased retry, the loop converges 9 of 11 tasks to a gated upgrade.

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

Training-free sparse attention based on cumulative energy filtering

Sparse attention accelerates Diffusion Transformers (DiTs) for video generation by computing only the important tokens while skipping the rest. The token selection strategy is key to balancing sparsity and accuracy. We formulate the token filtering process as a dual-goal optimization problem: maximizing sparsity and minimizing accuracy degradation. Existing algorithms cannot fulfill both objectives simultaneously. For example, Top-p only considers the accuracy constraint, while Top-k maintains a fixed computational budget but loosens the accuracy constraint. This paper demonstrates that maintaining a fixed recall rate is sufficient for ensuring accuracy, whereas a fixed threshold is suboptimal for reducing computational cost. Therefore, we propose a dynamic thresholding scheme to improve sparsity while maintaining the same level of accuracy. Furthermore, our algorithm is deeply integrated with Flash Attention (FA), eliminating the need for any additional masking computation overhead. Experimental results on Wan 2.2 validate that, compared to the BLASST algorithm which is also integrated with FA, our dynamic thresholding strategy enhances sparsity from 61.42\% to 82\% with a VBench metric drop of less than 5\%. This results in an approximate 15\% in attention computation and a $1.61\times$ increase in computational efficiency, which is 1.18x higher than that of BLASST.

22.
arXiv (CS.CL) 2026-06-18

DreamReasoner-8B: Block-Size Curriculum Learning for Diffusion Reasoning Models

Block diffusion language models accelerate decoding through parallel block-wise denoising, yet whether they can be reliably scaled for long chain-of-thought (CoT) reasoning remains unresolved. To this end, we develop DreamReasoner-8B, an open-source block diffusion reasoning model, and conduct a systematic study of how training and inference block sizes affect long-CoT reasoning. Our analysis reveals a stark performance disparity: training with large block sizes yields remarkably poor reasoning, whereas small block sizes preserve effective reasoning. To bridge this granularity gap, we propose block-size curriculum learning, which gradually transitions training from fine-grained to coarse-grained block sizes, thereby overcoming this limitation and enabling strong reasoning performance that generalizes across diverse inference block sizes. On mathematical and code reasoning benchmarks, DreamReasoner-8B achieves results competitive with leading open autoregressive models such as Qwen3-8B. This work establishes a practical foundation for efficient, reasoning-capable diffusion language models. We release our model at https://github.com/DreamLM/DreamReasoner.

23.
arXiv (CS.CL) 2026-06-15

UP-NRPA: User Portrait based Nested Rollout Policy Adaptation for Planning with Large Language Models in Goal-oriented Dialogue Systems

To address the challenge that current dialogue policy planning methods struggle to dynamically adapt to diverse user characteristics, this paper proposes a User Portrait based Nested Rollout Policy Adaptation (UP-NRPA) online framework with Large Language Models. In contrast to conventional approaches dependent on model training and require offline reinforcement learning policy models for user groups, UP-NRPA enables dynamic customization of dialogue strategies through an adaptive mechanism. This is achieved by leveraging real-time user feedback alongside personality, preferences, and objectives mapped from the current user portrait, thereby adapting to user characteristics without offline reinforcement learning. In collaborative and non-collaborative dialogue benchmarks, UP-NRPA demonstrated considerable benefits, achieving an impressive 100% success rate in multiple dialogue tasks. Particularly in negotiation tasks, the sale-to-list ratio (SL) increased by 56.41%. This demonstrates that UP-NRPA can adapt to diverse user needs without requiring a training mechanism, enabling the dialogue system to adapt to user characteristics.

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

A First-Principles Derivation of LLM Policy Optimization: From Expected Reward to GRPO and Its Structural Extensions

arXiv:2606.16733v1 Announce Type: new Abstract: Policy gradient algorithms for language models optimize the same objective $J(\theta) = \mathbb{E}*{\tau \sim p*\theta(\tau)}[R(\tau)]$, which has exactly two factors: the trajectory probability $p_\theta(\tau)$ and the reward $R(\tau)$. Every method from REINFORCE to PPO to GRPO and their descendants modifies one or both factors to address a specific failure in the preceding formulation. Existing surveys organize these methods by domain or chronology, which obscures the rationale behind each design choice and the precise location of its intervention within the gradient estimator. This survey revisits the landscape of LLM policy optimization from $J(\theta)$ on first principles and uses the trajectory side, induced by $p_\theta(\tau)$, and the reward side, induced by $R(\tau)$, as the two axes along which methods are located. It covers the path from REINFORCE and PPO to GRPO, as well as post-GRPO variants, Agentic RL, and GRPO-OPD. The resulting framework is unified, diagnostic, and extensible: it analyzes methods from a shared objective, identifies which side each method modifies and why, and applies the same trajectory and reward axes across these settings. Across these settings, the framework also exposes compound failures that no single-side fix resolves and that therefore require joint design of the trajectory side and the reward side. The boundary cases and coupled failures identified by this map mark where existing solutions run out and provide a principled starting point for designing the next generation of LLM policy optimization algorithms.

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

Generalization Guarantees for Multi-Input Neural Operator Learning in Sobolev Spaces

arXiv:2606.17419v1 Announce Type: new Abstract: We develop approximation and generalization error estimates for multi-input neural operators, with the output error measured in Sobolev norms. In contrast to standard operator-learning settings with a single input function, our framework allows multiple input functions defined on possibly different domains, with different dimensions and Sobolev regularities. The derived rates explicitly quantify the contribution of each input space to the final error bound. In particular, in the balanced regime, the approximation and generalization rates are governed by the interaction between the input dimensions, regularities, and Sobolev orders, while the dependence on the model complexity retains a \(\log\log/\log\)-type structure. Our analysis provides a general theoretical framework for multi-input operator learning, including Sobolev training, and is applicable to operator learning problems arising from partial differential equations and scientific computing.