Showing 1–50 of 515 results for author: Ren, L

A Systematic Gaia--ZTF Search for Short-Period Blue Compact-Binary Candidates

Authors: Jiamao Lin, Liangliang Ren, Yilong Li, Bo Ma, Di-Chang Chen, Zi-Heng Yu, Sen Yang, Shun-Jia Huang, Yi-Ming Hu, Chengyuan Li

Abstract: We present a catalog of 147 short-period (10.34--106.46~min) blue compact-binary candidates, identified by combining Gaia DR3 astrometry and photometry with ZTF DR23 light curves via a Gaia selection, period searches, and machine-learning morphology ranking. Of these, 111 lack prior compact-binary classifications. Multiwavelength data (DESI DR1, GALEX, AllWISE) reveal a heterogeneous sample: on th… ▽ More We present a catalog of 147 short-period (10.34--106.46~min) blue compact-binary candidates, identified by combining Gaia DR3 astrometry and photometry with ZTF DR23 light curves via a Gaia selection, period searches, and machine-learning morphology ranking. Of these, 111 lack prior compact-binary classifications. Multiwavelength data (DESI DR1, GALEX, AllWISE) reveal a heterogeneous sample: on the Gaia colour--magnitude diagram, 52 sources lie on the white-dwarf locus, 69 in the hot-subdwarf region, and 26 are intermediate. Among 26 sources with DESI spectra, only about one third follow the white-dwarf cooling sequence; the rest are more luminous blue stars with white-dwarf-like low-resolution spectra. We highlight a prioritized subset of new white-dwarf-locus candidates for follow-up, including ten with periods below 40~min and none with existing radial-velocity data. Under fiducial binary assumptions, 17 of these newly identified white-dwarf-locus candidates would exceed the adopted LISA signal-to-noise threshold (led by a 37~pc white dwarf), with the count depending on chirp mass (9 for $0.15\,M_\odot$, 17 for $0.3\,M_\odot$, 21 for $0.6\,M_\odot$), assuming orbital modulation. However, for most of the white-dwarf-locus sample, observed modulation amplitudes exceed any plausible ellipsoidal signal by three to five orders of magnitude, implying that rotating magnetic or chemically inhomogeneous single white dwarfs offer a viable alternative that ZTF photometry alone cannot rule out---the catalog includes at least one confirmed case. We release the full 147-source catalog, including periods, Gaia/spectroscopic classifications, harmonic/ellipsoidal diagnostics, and supplementary tables of fiducial GW estimates and UV--IR photometry. △ Less

Submitted 19 August, 2026; originally announced August 2026.

Comments: 39 pages, 17 figures, accepted in ApJ

Code as Representation: A Compilable Parsing Paradigm for Academic Documents

Authors: Rihui Jin, Jun Wang, chengyuan zhu, Liang Mingyu, Yue Gao, Li Yunxuan, Kuicai Dong, Guilin Qi, Lin Ren, Yongrui Chen, Xinbang Dai, Jiaqi Li, Tongtong Wu, Gholamreza Haffari

Abstract: Academic papers are a primary carrier of scientific knowledge, yet most of this knowledge remains locked in PDFs that are optimized for human reading rather than machine use. For Multimodal Large Language Models (MLLMs), the core challenge is not only perception, but representation: scientific pages interleave text with Structured Academic Elements (SAEs) such as tables, formulas, charts, and pseu… ▽ More Academic papers are a primary carrier of scientific knowledge, yet most of this knowledge remains locked in PDFs that are optimized for human reading rather than machine use. For Multimodal Large Language Models (MLLMs), the core challenge is not only perception, but representation: scientific pages interleave text with Structured Academic Elements (SAEs) such as tables, formulas, charts, and pseudocode, whose structure, data, and logic are poorly preserved by common surrogates like Markdown. We therefore propose Compilable Academic Document Parsing (CADP), a paradigm that reconstructs a full page as contextual \LaTeX{} plus executable Python, so that structure-preserving elements and executable chart representations can be reconstructed, recompiled, and directly verified against the source page. To support this setting, we introduce CADP-Bench, an expert-verified benchmark of full academic pages containing tightly coupled text and multiple SAE types, evaluated through a re-injection compilation protocol. We further study current capabilities using SOTA MLLMs and an exploratory multi-agent baseline that incorporates common agentic techniques. Results show that even frontier models still struggle to produce high-fidelity executable reconstructions, highlighting substantial room for improvement in structure-aware scientific document parsing. CADP-Bench is released for future research. △ Less

Submitted 18 August, 2026; originally announced August 2026.

Comments: Accepted by ACM MM 2026

Physics-informed Diffusion Generative Model for Time-Series Data Synthesis in Dynamic Systems

Authors: Haiteng Wang, Yunfei Zhu, Tao Wang, Yikang Li, Jiabao Dong, Xiaoge Zhang, Lei Ren

Abstract: Industrial time-series signals, such as turbine temperature and rotational speed in aero-engines, are essential for monitoring the health and operational status of complex dynamical systems. However, collecting such data is often limited by harsh environments (e.g., high temperature and high pressure) and the high cost of experimental testing. To address this challenge, we introduce PhysDGM, a ste… ▽ More Industrial time-series signals, such as turbine temperature and rotational speed in aero-engines, are essential for monitoring the health and operational status of complex dynamical systems. However, collecting such data is often limited by harsh environments (e.g., high temperature and high pressure) and the high cost of experimental testing. To address this challenge, we introduce PhysDGM, a stepwise physics-embedded diffusion generative model for synthesizing time-series data that are consistent with the underlying physical laws of dynamical systems. PhysDGM embeds physical laws directly into each reverse diffusion step of the generative process, ensuring trajectory-level physical consistency, rather than enforcing constraints only at the final output. A large-scale AI-synthetic dataset (4.4 million samples, 20x scale-up) constructed by PhysDGM demonstrates strong fidelity across 34 datasets spanning turbofan engines, aero-engines, batteries, and chemical processes. After incorporating the synthetic data, the downstream task performance substantially surpassed that using real data alone by 48% for remaining useful life prediction, 15% for health indicator estimation, 22% for state-of-health assessment, and 20% for fault diagnosis. Moreover, it requires 10-20x less training data than existing approaches, substantially reducing the high cost of data collection in dynamical systems. We further demonstrate PhysDGM's potential in identifying early-stage faults in aero-engines by incorporating AI-synthesized data. In summary, PhysDGM provides a solid foundation for generating physically consistent industrial time-series, paving the way for expanding physics-guided AI into diverse data-scarce environments, including both industrial machinery and complex chemical reaction dynamics. △ Less

Submitted 11 August, 2026; originally announced August 2026.

Comments: 27 pages; 5 main figures and 4 extended data figures

VANE: Reliable Test-Time Training for Vision-Language-Action Models via Future Visual Representation Prediction

Authors: Hongjin Ji, Guoyang Xia, Luoyang Sun, Fangxiang Feng, Lei Ren

Abstract: Test-time training (TTT) offers a lightweight way to adapt vision--language--action (VLA) policies from unlabeled deployment streams, but it remains difficult to use reliably in closed-loop manipulation. A shared adaptation space can mix incompatible task corrections, while an online update can alter subsequent actions before its consequences are known. We introduce a reliable TTT framework for VL… ▽ More Test-time training (TTT) offers a lightweight way to adapt vision--language--action (VLA) policies from unlabeled deployment streams, but it remains difficult to use reliably in closed-loop manipulation. A shared adaptation space can mix incompatible task corrections, while an online update can alter subsequent actions before its consequences are known. We introduce a reliable TTT framework for VLA policies (VANE). VANE conditions prompt adaptation on the current vision--language context and learns from the future visual consequences of executed actions. Candidate updates are isolated from the live policy, evaluated on subsequent observations, and committed only when supported by future evidence, making adaptation selective and reversible. On SimplerEnv WidowX, VANE improves average success by $3.2$ percentage points over the corresponding TTT baseline. Results on Google Robot further show that deployment-time gains remain task- and embodiment-dependent. Together, these results demonstrate a constrained, evidence-based approach to adapting VLA policies during interaction. △ Less

Submitted 12 August, 2026; v1 submitted 10 August, 2026; originally announced August 2026.

Test-time Adaptation of Pelvic Bone Segmentation Models via Dynamic Reliability-Guided

Authors: Ling Ren, Chao Deng, Ziming Wang, Yuecong Xu, Kai Zheng

Abstract: Reliable pelvic bone segmentation (PBS) from CT is essential for robot-assisted pelvic trauma surgery, yet deploying a source-trained model to a new hospital suffers from severe performance degradation due to cross-center domain shifts. While test-time adaptation (TTA) enables online model adaptation without accessing source data, existing methods show limited effectiveness for PBS, facing challen… ▽ More Reliable pelvic bone segmentation (PBS) from CT is essential for robot-assisted pelvic trauma surgery, yet deploying a source-trained model to a new hospital suffers from severe performance degradation due to cross-center domain shifts. While test-time adaptation (TTA) enables online model adaptation without accessing source data, existing methods show limited effectiveness for PBS, facing challenges including boundary degradation, anatomical inconsistency under domain shifts, and voxel-level class imbalance. To address these challenges, we propose a novel closed-loop dynamic Reliability-Guided TTA framework (ReGA) for PBS. Specifically, we introduce a pseudo-label reliability criterion termed Segmentation Inference Consistency Evaluation (SICE), which jointly measures region overlap and boundary deviation via dropout-based ensemble predictions. Based on SICE, a trust-weighted refinement module adaptively updates features to mitigate boundary errors in pseudo-labels. Furthermore, a confidence-weighted region-level contrastive learning strategy is proposed to enforce anatomical consistency. Finally, ReGA follows the teacher-student (TS) scheme to alleviate voxel-level class imbalance. Experiments on three heterogeneous 3D pelvic CT datasets demonstrate that ReGA consistently outperforms state-of-the-art TTA methods, enabling effective adaptation of the source-trained PBS model to unseen clinical domains. The code is available at https://github.com/Ren-ling/ReGA. △ Less

Submitted 1 August, 2026; originally announced August 2026.

Comments: Provisionally accepted for presentation at MICCAI 2026

MSC Class: Computer Vision and Pattern Recognition (cs.CV)

Enforcing Access Control in Distributed Version Control Systems

Authors: Xin Xu, Quanwei Cai, Jingqiang Lin, Shiran Pan, Liangqin Ren

Abstract: Version control systems (VCS), including central VCS (CVCS) and distributed VCS (DVCS), are widely adopted to manage the changes to various types of data. Unlike the CVCS where all the entities obtain the data from the server and the access control is enforced with the cooperation of the server, each entity in the DVCS stores the entire repository, obtains the repository shared by any entity and i… ▽ More Version control systems (VCS), including central VCS (CVCS) and distributed VCS (DVCS), are widely adopted to manage the changes to various types of data. Unlike the CVCS where all the entities obtain the data from the server and the access control is enforced with the cooperation of the server, each entity in the DVCS stores the entire repository, obtains the repository shared by any entity and is free to share its own repository. Therefore, existing access control schemes for CVCS are not suitable for DVCS. In this paper, we present a distributed access control scheme (Disac) for DVCS. Disac makes each entity have full control over its data, while the access control is enforced at each entity independently. We adopt Attribute-based Encryption (ABE) and Attribute-based Signature (ABS) to achieve the read and write permission control. The analysis of the Git client demonstrates that Disac can be integrated easily. △ Less

Submitted 31 July, 2026; originally announced August 2026.

Comments: 6 pages, 2 figures. Author-accepted manuscript

Journal ref: 2019 IEEE International Conference on Multimedia and Expo (ICME), pp. 772-777, 2019

Enforcing Cryptographic Distributed-VCS Access Control with No Trust on Servers

Authors: Xin Xu, Zhen Yang, Quanwei Cai, Jingqiang Lin, Liangqin Ren, Bo Chen, Yongfeng Huang

Abstract: Version control systems (VCS), including central VCS (CVCS) and distributed VCS (DVCS), are widely adopted to manage changes to software code and various types of documents. Unlike CVCS, where entities obtain data from a central server, each entity in DVCS stores the entire repository and shares it independently. In VCS, existing access control schemes require the participation of a central server… ▽ More Version control systems (VCS), including central VCS (CVCS) and distributed VCS (DVCS), are widely adopted to manage changes to software code and various types of documents. Unlike CVCS, where entities obtain data from a central server, each entity in DVCS stores the entire repository and shares it independently. In VCS, existing access control schemes require the participation of a central server and cannot be deployed in a completely distributed scenario. Additionally, these schemes often fail to enforce fine-grained access control for write permissions, which is crucial for collaborative work in a distributed environment. In this paper, we propose a distributed version control system access control scheme (named DVAC), which enforces cryptographic access control on distributed user nodes based on attribute-based encryption (ABE) and attribute-based signature (ABS). DVAC is designed to enforce a cryptographic access control protocol for DVCS, which enables file granularity read and write separation access control without the support of a central server. To ensure the integrity of the core version control functions in DVCS while protecting data security, DVAC incorporates a version control adaptation protocol. Additionally, DVAC leverages Ethereum smart contracts to maintain access control policies, ensuring distributed storage and trusted management of access policies. The architecture of DVAC is designed to seamlessly integrate with existing mature DVCS, such as Git, with minimal modifications. We have implemented a prototype of DVAC and integrated it with Git. A comprehensive performance evaluation was conducted to assess the overhead introduced by DVAC, and it was demonstrated that the overhead is modest. △ Less

Submitted 31 July, 2026; originally announced July 2026.

Comments: Accepted manuscript. 15 pages, 7 figures. Published in Journal of Information Security and Applications, Vol. 93 (2025), Article 104103

Journal ref: Journal of Information Security and Applications, 93 (2025), 104103

Zero-Mem: Zero-Token Memory Operations for LLM Agents

Authors: Yilin Xiao, Zhehan Zhu, Yujing Zhang, Jin Chen, Zijin Hong, Luyao Zhuang, Qinggang Zhang, Shengyuan Chen, Xiaocao Ouyang, Lingfei Ren, Xiao Huang

Abstract: LLM agents need memory to act consistently over long interactions, yet many systems use additional LLM calls to operate that memory. Generating intermediate records and mediating their retrieval adds recurring token and time costs, while omitted or merged details can obscure the original evidence. We ask whether structured memory access requires generation at all. Zero-Mem introduces \emph{zero-to… ▽ More LLM agents need memory to act consistently over long interactions, yet many systems use additional LLM calls to operate that memory. Generating intermediate records and mediating their retrieval adds recurring token and time costs, while omitted or merged details can obscure the original evidence. We ask whether structured memory access requires generation at all. Zero-Mem introduces \emph{zero-token memory operations}: no step outside final question answering invokes an LLM or consumes LLM input or output tokens; encoder computation is accounted for separately. Zero-Mem preserves original interaction traces as its source of record. It organizes the traces in two complementary ways. An entity--context graph exposes connections across interactions, while a temporal hierarchy preserves conversational locality and session state. For each query, Zero-Mem weighs the two views, retrieves from both, and follows their structure to recover supporting relations or surrounding context. Deterministic calibration first discards conflicting evidence and then keeps the reader's answer grounded in the retrieved traces. Only the final-QA reader invokes an LLM. Across long-memory and long-context question-answering benchmarks, Zero-Mem achieves competitive performance while eliminating LLM calls and LLM-token consumption from memory operations. With the same final-QA reader and context budget, it reduces memory-operation time cost by 57.6\% relative to the fastest compared baseline. Ablations support the contribution of the two views and their query-dependent coordination. Overall, the results show that structured agent memory need not generate an intermediate representation of the past. After peer review, the code and implementation details will be available at \textcolor{blue}{https://github.com/TheMoon0815/Zero-mem}. △ Less

Submitted 31 July, 2026; originally announced July 2026.

Comparative Periodogram Analysis of 22 Years of Super-Kamiokande Solar $^{8}\mathrm{B}$ Neutrino Data: Classical, Phase-Based, and Information Theoretic Methods

Authors: Liangliang Ren, Ze-Lin Zhang, Bing Xu, Tian-Cheng Huang, Ran Wang, Jia-Xin Dong, Yan-Ping Wang

Abstract: Solar $^8\mathrm{B}$ neutrinos offer a unique probe of solar interior dynamics and neutrino electromagnetic properties. We present a systematic, multi-method periodogram analysis of the 22-year Super-Kamiokande solar neutrino dataset (1996--2018), comparing nine algorithms. Through hierarchical temporal segmentation, we disentangle astrophysical signals from detector systematics. The Generalized L… ▽ More Solar $^8\mathrm{B}$ neutrinos offer a unique probe of solar interior dynamics and neutrino electromagnetic properties. We present a systematic, multi-method periodogram analysis of the 22-year Super-Kamiokande solar neutrino dataset (1996--2018), comparing nine algorithms. Through hierarchical temporal segmentation, we disentangle astrophysical signals from detector systematics. The Generalized Lomb-Scargle (GLS) method provides the most statistically robust detections by correctly handling heteroscedastic uncertainties, whereas classical Lomb-Scargle systematically underestimates significance. The Lafler--Kinman method generally fails, whereas independent algorithms like MHAOV and PDM1 recover consistent periodicities, providing vital cross-validation. In pre-2001 and SK-I data, seven algorithms provide \textit{weak evidence} ($\ln B > 0$) for a $\sim 38.8$ d periodicity. However, this signal is entirely absent in the highest-statistics SK-IV modified flux data, where the Bayes factor decisively favors the null model ($\ln B \ll -5$), indicating it is a transient feature of the early low-statistics era. Conversely, a $\sim 24.3$ d signal in post-2001 raw flux is decisively rejected by the Bayesian framework and vanishes in modified flux, confirming its seasonal systematic origin. Furthermore, no evidence is found for an $\sim 11$-year solar cycle modulation, yielding a stringent amplitude upper limit of $<0.2\%$ of the mean flux. By highlighting the stark contrast between frequentist significance and Bayesian model selection ($\ln B$) in low signal-to-noise regimes, we establish a rigorous, multi-metric best-practice framework for periodicity searches. This work provides a direct methodological blueprint for next-generation observatories like Hyper-Kamiokande and JUNO. △ Less

Submitted 31 July, 2026; v1 submitted 30 July, 2026; originally announced July 2026.

Comments: 42 pages,24 figures,3 tables,submitted to JCAP,comments welcome

ReLATE: Reliability-Guided Evidence Fusion for Robust UAV--Satellite cross-view Geo-Localization

Authors: Haochen Jiang, Jialei Pan, Yuzhe Sun, Zhe Dong, Lecheng Ren, Yanfeng Gu, Tianzhu Liu

Abstract: Unmanned aerial vehicle (UAV)-satellite cross-view geo-localization matches UAV images against satellite imagery and has achieved impressive accuracy on clean (non-degraded) image benchmarks. In real-world flights, however, UAV observations are frequently affected by adverse weather, illumination changes, platform motion, sensor noise, and compression, while the robustness of existing methods unde… ▽ More Unmanned aerial vehicle (UAV)-satellite cross-view geo-localization matches UAV images against satellite imagery and has achieved impressive accuracy on clean (non-degraded) image benchmarks. In real-world flights, however, UAV observations are frequently affected by adverse weather, illumination changes, platform motion, sensor noise, and compression, while the robustness of existing methods under such degradations remains largely unexamined. In this paper, we present UAVSat-Deg, a large-scale robustness benchmark for degraded UAV-satellite geo-localization, comprising University-1652-Deg and SUES-200-Deg. UAVSat-Deg covers 27 corruption types, including 19 core and 8 compound corruptions, at three severity levels, supports bidirectional drone-to-satellite and satellite-to-drone retrieval as well as multi-height UAV acquisition, and contains more than 11.7 million pre-generated corrupted test images. Benchmarking representative methods under this protocol reveals substantial robustness gaps, particularly under severe and compound corruptions. To address this problem, we propose ReLATE, a Reliable Evidence Learning framework with Adaptive Token Evidence Regulation, which realizes reliability-adaptive feature fusion during descriptor construction. ReLATE estimates a structure-smoothed reliability field over visual tokens, aggregates trustworthy local evidence, and adaptively integrates it into query-derived representations; the regulated query representations are then combined with the CLS-token and GeM-pooled branches to form the final cross-view descriptor. Across both test sets and retrieval directions, ReLATE achieves the best average corrupted-test performance among the compared methods while maintaining competitive accuracy on clean images. The code and dataset will be available at https://github.com/JHC626/ReLATE. △ Less

Submitted 28 July, 2026; originally announced July 2026.

Mass/electric versus NUT/magnetic charges: duality from scattering amplitudes in $D\geq4$ and for all bosonic spins

Authors: Ricardo Monteiro, Lecheng Ren, Daniel Siretanu

Abstract: We revisit Kerr-NUT metrics and related spin-$s$ fields in $D\geq4$, and study their associated scattering amplitudes. Starting in position space, we highlight the interpretation of mass and (multiple) NUT charges as being associated to distinct solutions of the rotation-deformed radial equation, which is made explicit in Cartesian multi-Kerr-Schild coordinates. This interpretation extends to elec… ▽ More We revisit Kerr-NUT metrics and related spin-$s$ fields in $D\geq4$, and study their associated scattering amplitudes. Starting in position space, we highlight the interpretation of mass and (multiple) NUT charges as being associated to distinct solutions of the rotation-deformed radial equation, which is made explicit in Cartesian multi-Kerr-Schild coordinates. This interpretation extends to electromagnetism with magnetic-type charges, and also extends to higher-spin counterparts, in accordance with the classical double or multi copy. We then establish a notion of ``electric-magnetic" duality in higher dimensions, relating mass/electric to NUT/magnetic charges, which generalises the $D=4$ case in a novel manner. For $D\geq6$, this involves the choice where the multiple magnetic charges are equal. The duality is revealed in momentum space, by the 3-point scattering amplitudes that generate the solutions. For all spins, these amplitudes are constructed from a spin-raising operator $\mathcal S_μ$ and take the form $\varepsilon^{μ_1\cdotsμ_s}{\mathcal S}_{μ_1}\cdots {\mathcal S}_{μ_s}$ acting on a scalar seed. The scalar seed of the electric sector is dual to that of the magnetic sector: where the former's rotation dependence resums to a Bessel function $J_{\frac{D-5}{2}}$, the latter resums to a Bessel function $J_{-\frac{D-5}{2}}$. Finally, we explore the notion of self-duality that arises from this picture. Studying the classical $2\!\mapsto\!2$ amplitudes that determine leading-order scattering, we find no evidence of a higher-dimensional analogue of the integrability of $D=4$ self-dual gravity. △ Less

Submitted 24 July, 2026; originally announced July 2026.

Comments: 47 pages

Report number: QMUL-PH-26-27

PrivDNN: A Secure Multi-Party Computation Framework for Deep Learning using Partial DNN Encryption

Authors: Liangqin Ren, Zeyan Liu, Fengjun Li, Kaitai Liang, Zhu Li, Bo Luo

Abstract: In the past decade, we have witnessed an exponential growth of deep learning models, platforms, and applications. While existing DL applications and Machine Learning as a service (MLaaS) frameworks assume fully trusted models, the need for privacy-preserving DNN evaluation arises. In a secure multi-party computation scenario, both the model and the data are considered proprietary, i.e., the model… ▽ More In the past decade, we have witnessed an exponential growth of deep learning models, platforms, and applications. While existing DL applications and Machine Learning as a service (MLaaS) frameworks assume fully trusted models, the need for privacy-preserving DNN evaluation arises. In a secure multi-party computation scenario, both the model and the data are considered proprietary, i.e., the model owner does not want to reveal the highly valuable DL model to the user, while the user does not wish to disclose their private data samples either. Conventional privacy-preserving deep learning solutions ask the users to send encrypted samples to the model owners, who must handle the heavy lifting of ciphertext-domain computation with homomorphic encryption. In this paper, we present a novel solution, namely, PrivDNN, which (1) offloads the computation to the user side by sharing an encrypted deep learning model with them, (2) significantly improves the efficiency of DNN evaluation using partial DNN encryption, (3) ensures model accuracy and model privacy using a core neuron selection and encryption scheme. Experimental results show that PrivDNN reduces privacy-preserving DNN inference time and memory requirement by up to 97% while maintaining model performance and privacy. Codes can be found at https://github.com/LiangqinRen/PrivDNN △ Less

Submitted 23 July, 2026; originally announced July 2026.

Comments: Published in Proceedings on Privacy Enhancing Technologies (PoPETs 2024)

Journal ref: Proceedings on Privacy Enhancing Technologies, 2024(3), 477-494

Energy-Energy Correlators at Strong Coupling

Authors: Max Jackson, Lecheng Ren, Bo Wang, Congkao Wen

Abstract: We study energy-energy correlators (EEC) in planar $\mathcal{N}=4$ super Yang-Mills theory at strong 't Hooft coupling $λ$. We consider the EEC in states created by half-BPS operators of arbitrary dimension $p$, and determine the corresponding event-shape function up to order $λ^{-3/2}$ from the worldsheet representation of the AdS Virasoro-Shapiro amplitude with Kaluza-Klein external states. For… ▽ More We study energy-energy correlators (EEC) in planar $\mathcal{N}=4$ super Yang-Mills theory at strong 't Hooft coupling $λ$. We consider the EEC in states created by half-BPS operators of arbitrary dimension $p$, and determine the corresponding event-shape function up to order $λ^{-3/2}$ from the worldsheet representation of the AdS Virasoro-Shapiro amplitude with Kaluza-Klein external states. For $p=2$ we compute the second curvature correction, which completes the EEC through order $λ^{-2}$; the new contribution improves the agreement with recently derived non-perturbative bounds at intermediate coupling. We further develop a complementary method in which the strong-coupling expansion coefficients of the EEC are extracted directly from the Wilson coefficients of low-energy expansion of the AdS Virasoro-Shapiro amplitude, and find the two approaches in perfect agreement. △ Less

Submitted 23 July, 2026; originally announced July 2026.

Comments: 28 pages + 4 appendices, 3 figures

Report number: QMUL-PH-26-26

TOUR: A Trajectory-Level Unlearning Benchmark for Offline Reinforcement Learning

Authors: Chaofan Pan, Lingfei Ren, Xiangyu Jiang, Yanhua Li, Xuemei Cao, Xiangkun Wang, Hao Yu, Wei Wei, Xin Yang

Abstract: Offline Reinforcement Learning (RL) agents are trained on fixed behavioral trajectories, which makes trajectory-level deletion important when selected data must be removed after training. Evaluating such deletion is difficult because a lower membership score can reflect trajectory removal, residual memorization visible to another attack, or policy collapse that destroys useful behavior. We introdu… ▽ More Offline Reinforcement Learning (RL) agents are trained on fixed behavioral trajectories, which makes trajectory-level deletion important when selected data must be removed after training. Evaluating such deletion is difficult because a lower membership score can reflect trajectory removal, residual memorization visible to another attack, or policy collapse that destroys useful behavior. We introduce Trajectory-level memOrization and Unlearning in offline RL (TOUR), a benchmark that combines trajectory-level partitioning, matched non-member controls, retraining references, retained-performance anchors, and multi-attack privacy auditing. Across D4RL locomotion experiments and an exploratory AntMaze extension, TOUR shows that common deletion baselines have environment-dependent privacy-utility behavior. Retraining and fine-tuning often provide stronger retained-utility references than uniform GA+Refit, while TrajDeleter remains a useful comparator but is not uniformly stronger under the same audit. Reference-model, threshold, deviation, equivalence, action-error, representation-based, and query-limited attacks further show that a single likelihood-based membership score can overstate deletion quality. In the evaluated settings, conclusions about offline RL unlearning are therefore not stable under single-score auditing. They depend on matched non-member construction, retraining-relative calibration, attack family, retained utility, and explicit scope for diagnostic architecture or component-level evidence. △ Less

Submitted 23 July, 2026; originally announced July 2026.

PhantomSeal: Proactive Deepfakes Defense with Identity/Context Protection and Forensic Tracing

Authors: Liangqin Ren, Zeyan Liu, Ye Wang, Yuxin Chen, Fengjun Li, Bo Luo

Abstract: Deepfakes, especially face-swapping attacks, pose significant challenges to authenticity, security, and ethics across science, engineering, and society. While most existing detection/tracing approaches operate post hoc, proactive defenses that aim to intervene before deepfake generation remain limited in terms of real-world effectiveness. In this paper, we present PhantomSeal, the first proactive… ▽ More Deepfakes, especially face-swapping attacks, pose significant challenges to authenticity, security, and ethics across science, engineering, and society. While most existing detection/tracing approaches operate post hoc, proactive defenses that aim to intervene before deepfake generation remain limited in terms of real-world effectiveness. In this paper, we present PhantomSeal, the first proactive defense to simultaneously protect both the identity and the context of users' images from being used in face-swapping attacks, while supporting forensic tracing. We present a novel cloaking technique that embeds a selected identity as a stealthy identifier. This mechanism steers the deepfake generation process toward producing content that resembles the chosen cloak identity, thereby preventing successful face-swapping while enabling effective feature-based forensic analysis. The effectiveness and robustness of PhantomSeal is demonstrated in extensive experiments across different face-swapping architectures and models. For example, it reduces the attack success rate of SimSwap, an advanced deepfake model, to 0.30%, and correctly identifies 97.97% of manipulated content. The source codes is available at https://github.com/LiangqinRen/PhantomSeal. △ Less

Submitted 3 August, 2026; v1 submitted 20 July, 2026; originally announced July 2026.

Comments: 22 pages, 5 figures, and 24 tables. Extended version with additional technical details and appendices. Accepted by ACM CCS 2026

The Eos detector: a demonstrator of hybrid optical detection technology

Authors: Eos Collaboration, S. Arora, M. Askins, A. J. Bacon, Z. Bagdasarian, A. Baldoni, L. Bartoszek, M. Bergevin, Y. Bezawada, E. Blucher, J. Boissevain, R. Bonventre, E. J. Callaghan, D. F. Cowen, K. DeHolton, M. Diwan, M. Dubnowski, P. Englezos, S. Gadamsetty, C. Grant, B. Harris, M. R. Hebert, S. Jeon, T. Kaptanoglu, A. Katt , et al. (42 additional authors not shown)

Abstract: Eos is an R&D testbed for hybrid detector technologies, featuring state-of-the-art sub-ns photosensors, the first implementation of dichroicons in a large-scale demonstrator, and the deployment of novel detection media such as water-based liquid scintillator (WbLS). By separating Cherenkov and scintillation light, Eos leverages the benefits of both to explore the potential of next-generation neutr… ▽ More Eos is an R&D testbed for hybrid detector technologies, featuring state-of-the-art sub-ns photosensors, the first implementation of dichroicons in a large-scale demonstrator, and the deployment of novel detection media such as water-based liquid scintillator (WbLS). By separating Cherenkov and scintillation light, Eos leverages the benefits of both to explore the potential of next-generation neutrino technologies. An extensive radioactive source calibration program enables the characterization of position, direction, and energy reconstruction performance of a variety of target materials. Furthermore, Eos will provide data to refine optical models and inform the development and simulation of future neutrino experiments. This paper describes the as-built design and data-taking plan of Eos, outlining its scientific motivations and role in the development of future detector technologies. △ Less

Submitted 22 July, 2026; v1 submitted 22 July, 2026; originally announced July 2026.

Comments: 48 pages, 39 figures

EvoThink: Evolving Thinking in Large Reasoning Models via Self-Pruning and Aha-Moment Preference Optimization

Authors: Xinbang Dai, Zheyu Xin, Huikang Hu, Lin Ren, Rihui Jin, Guohui Xiao, Guilin Qi, Kuicai Dong, Zhaocheng Du, Yuyang Zhang

Abstract: Large Reasoning Models (LRMs) often suffer from overthinking due to redundant verification steps. Existing approaches for mitigating overthinking, such as fast-slow thinking switching and reasoning trajectory compression, fail to make a fine-grained distinction between beneficial and redundant steps within the LRM's reasoning process, and may thus impair reasoning capability in their pursuit of ef… ▽ More Large Reasoning Models (LRMs) often suffer from overthinking due to redundant verification steps. Existing approaches for mitigating overthinking, such as fast-slow thinking switching and reasoning trajectory compression, fail to make a fine-grained distinction between beneficial and redundant steps within the LRM's reasoning process, and may thus impair reasoning capability in their pursuit of efficiency. To simultaneously improve reasoning efficiency and capability, we propose EvoThink, a framework that reduces redundant verification and encourages the exploration of new reasoning paths. EvoThink comprises two key components: Self-Pruning Training (SPT), an unsupervised method that iteratively prunes redundant reasoning steps and self-trains on the concise trajectories; and Aha-Moment Preference Optimization (AMPO), which, inspired by genetic algorithms, identifies valuable failed reasoning attempts, synthesizes from-wrong-to-right aha-moment data, and optimizes the model to internalize this reasoning pattern. Extensive evaluations across mathematical reasoning and code generation benchmarks demonstrate that EvoThink not only substantially reduces inference-time token usage but also improves the reasoning capability of LRMs. △ Less

Submitted 22 July, 2026; originally announced July 2026.

Comments: 9 pages, 7 figures, accepted by IJCAI 2026

Plover: Steering GUI Agents through Plan-Centric Interaction

Authors: Madhumitha Venkatesan, Shicheng Wen, Jiajing Guo, Jorge Piazentin Ono, Liu Ren, Dongyu Liu

Abstract: Graphical user interface (GUI) automation remains challenging in real-world environments, where dynamic layouts, unexpected dialogs, and evolving interface states can cause autonomous agents to drift from user intent. Recent vision-based multimodal agents improve flexibility by operating directly over screenshots and natural language instructions, but planning and adaptation often remain internal,… ▽ More Graphical user interface (GUI) automation remains challenging in real-world environments, where dynamic layouts, unexpected dialogs, and evolving interface states can cause autonomous agents to drift from user intent. Recent vision-based multimodal agents improve flexibility by operating directly over screenshots and natural language instructions, but planning and adaptation often remain internal, limiting users' ability to inspect, supervise, or correct system behavior. We present Plover, a plan-centric vision-based GUI automation system that externalizes task plans and replanning as persistent, inspectable, and revisable artifacts. Through a planner--executor architecture, Plover supports explicit supervision of evolving execution, localized correction through editable plans, natural-language guidance, and screenshot-grounded interventions, while preserving prior progress during repair. A formative study with six participants informed the interaction design. We then evaluate Plover through benchmark failure-case repair and scenario-based workflow analyses. Our results show that many autonomous GUI-agent failures are structurally repairable when plans remain visible and interventions are localized, and that explicit replanning helps make GUI automation more transparent, controllable, and adaptable. △ Less

Submitted 16 July, 2026; originally announced July 2026.

VLT: A Vision-Language-Time Series Multimodal Foundation Model for Industrial Intelligence

Authors: Haiteng Wang, Jingheng Yan, Xiaokang Wang, Lei Ren

Abstract: Industrial time series serve as the foundation for Prognostics and Health Management (PHM) to ensure the reliability and safety of industrial equipment such as aero-engines. However, existing approaches are typically limited to single-modality modeling, which restricts their generalization in complex scenarios. Although recent advances in large language models (LLMs) provide new opportunities for… ▽ More Industrial time series serve as the foundation for Prognostics and Health Management (PHM) to ensure the reliability and safety of industrial equipment such as aero-engines. However, existing approaches are typically limited to single-modality modeling, which restricts their generalization in complex scenarios. Although recent advances in large language models (LLMs) provide new opportunities for multimodal learning, bridging continuous time-series signals and discrete textual semantics remains an open challenge. To this end, we propose VLT, a multimodal foundation model that jointly models time-series, frequency-spectrum visual representations, and textual knowledge. A key insight is to utilize the frequency spectrum as a visual bridge to connect continuous temporal signals with discrete semantics. Specifically, a Time-aware Mixture-of-Experts (Time-MoE) is designed to capture heterogeneous temporal dynamics, while a Frequency-Text Augmented Learner enables joint modeling of spectral and semantic features within a shared representation space. Furthermore, a time-centric gradient alignment mechanism is introduced to mitigate cross-modal optimization conflicts via gradient normalization and reliability-aware dynamic reweighting. Extensive experiments on multiple industrial datasets demonstrate that VLT outperforms state-of-the-art methods, achieving superior robustness and generalization under few-shot, noisy, and incomplete-modality settings. △ Less

Submitted 15 July, 2026; originally announced July 2026.

Comments: 18 pages, 13 figures, and 13 tables, including supplementary material. Haiteng Wang and Jingheng Yan contributed equally to this work

$K^{*}(892)^0$ production and the time between freeze-outs in $^{40}$Ar+$^{45}$Sc collisions by NA61/SHINE at the CERN SPS

Authors: NA61/SHINE Collaboration, :, P. Adrich, K. K. Allison, M. Bajda, Y. Balkova, D. Battaglia, M. Bielewicz, A. Blondel, M. Bogomilov, Y. Bondar, J. Brzychczyk, M. Buryakov, A. F. Camino, Y. D. Chandak, M. Csanád, M. Ćwiok, T. Czopowicz, C. Dalmazzone, N. Davis, A. Dmitriev, P. von Doetinchem, W. Dominik, J. Dumarchez, R. Engel , et al. (88 additional authors not shown)

Abstract: The analysis of the production of strange $K^{*}(892)^0$ resonances allows us to better understand the temporal evolution of high-energy nucleus--nucleus collisions. In particular, the ratio of $K^{*}(892)^0$ to charged kaon yields is used to determine the time interval between chemical and kinetic freeze-outs. In this paper, the first measurements of $K^{*}(892)^0$ production in central $^{40}$Ar… ▽ More The analysis of the production of strange $K^{*}(892)^0$ resonances allows us to better understand the temporal evolution of high-energy nucleus--nucleus collisions. In particular, the ratio of $K^{*}(892)^0$ to charged kaon yields is used to determine the time interval between chemical and kinetic freeze-outs. In this paper, the first measurements of $K^{*}(892)^0$ production in central $^{40}$Ar+$^{45}$Sc collisions at the CERN Super Proton Synchrotron are reported. They were performed by NA61/SHINE at collision center-of-mass energies per nucleon pair $\sqrt{s_\mathrm{NN}}$ = 8.8, 11.9, 16.8 GeV. The obtained $\langle K^{*}(892)^0 \rangle/\langle K^{+} \rangle $ and $\langle K^{*}(892)^0 \rangle/\langle K^{-} \rangle$ mean multiplicity ratios are compared with corresponding results in $p$+$p$ collisions, allowing for an estimate of the time interval between chemical and thermal freeze-outs in the $^{40}$Ar+$^{45}$Sc system. These are the first such results reported for $^{40}$Ar+$^{45}$Sc collisions. △ Less

Submitted 6 July, 2026; originally announced July 2026.

Comments: 12 pages, 3 figures

EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments

Authors: Deyao Zhu, Xin Zhou, Shengling Qin, Xuekai Zhu, Hangliang Ding, Shu Zhong, Zixin Wen, Zhonglin Xie, Chenhui Gou, Linxuan Ren, Yueyang Wang, Junfeng Zhong, Rui Liu, Tian Gao, Yangguang Lin, Jingyuan Zhang, Maojia Song, Xuan Qi, Jinhong Wu, Chenyang Zhang, Yinzhu Piao, Ziru Niu, Hongbin Lin, Lingxiang Meng, Peng Tang , et al. (22 additional authors not shown)

Abstract: Pretraining scaling laws reveal that model capability improves predictably with data and compute. But learning from real world environments after deployment remains far less understood. Analyzing roughly 38,000 hours of agent interaction with the environment across 134 real world tasks, we find, to the best of our knowledge, the first evidence that overall performance during environment learning f… ▽ More Pretraining scaling laws reveal that model capability improves predictably with data and compute. But learning from real world environments after deployment remains far less understood. Analyzing roughly 38,000 hours of agent interaction with the environment across 134 real world tasks, we find, to the best of our knowledge, the first evidence that overall performance during environment learning follows a log-sigmoid scaling law with remarkably high precision, reaching R^2 = 0.998. Across model generations, we also find that agent learning speed roughly doubles every three months. This discovery stems from EdgeBench, a suite of 134 real world tasks with ultra-long horizons, spanning scientific discovery, software engineering, combinatorial optimization, professional knowledge work, formal mathematics, and interactive games. Each task sustains at least 12 hours of continuous agent operation under rich, multilevel feedback, and is built through substantial expert effort. We publicly release 51 tasks and our full evaluation framework to accelerate the study of how agents learn from real world experience. △ Less

Submitted 6 July, 2026; originally announced July 2026.

Rethinking Speech-LLM Integration for ASR: Effective Joint Speech-Text Training by Interleaving

Authors: Ruchao Fan, Yiming Wang, Rui Zhao, Liliang Ren, Keqi Deng, Xiaoyang Chen, Ali Zare, Bo Ren, Yuxuan Hu, Junkun Chen, Yan Huang, Yelong Shen, Jinyu Li

Abstract: Speech-LLM integration has shown promising results by leveraging extensive textual pretraining, yet its specific benefits for automatic speech recognition (ASR) remain unclear. We observe that as supervised ASR training data increases, the contribution of LLM priors becomes less evident, and simple speech-text joint training under-utilizes textual knowledge. We therefore propose Joint Speech-Text… ▽ More Speech-LLM integration has shown promising results by leveraging extensive textual pretraining, yet its specific benefits for automatic speech recognition (ASR) remain unclear. We observe that as supervised ASR training data increases, the contribution of LLM priors becomes less evident, and simple speech-text joint training under-utilizes textual knowledge. We therefore propose Joint Speech-Text Interleaved Pretraining (JSTIP), an ASR-oriented pretraining strategy that constructs word-level and segment-level interleaved speech-text sequences within aligned pairs for speech-LLM architectures that accept continuous inputs. Experiments on 38k hours of ASR data show consistent entity accuracy improvement compared to ASR-only and joint speech-text training baselines. JSTIP achieves on-par entity recognition performance using domain transcription text compared to synthetic speech-text pairs, simplifying domain adaptation. Benefiting from textual pretraining and domain text data, JSTIP is competitive with open-source ASR and Speech-LLM systems in medical entity recognition. The zero-shot speech question answering behaviors further suggest that interleaving reduces the speech-text modality gap and preserves the LLM generative prior, which is likely the reason for the entity improvements on the ASR task. △ Less

Submitted 2 July, 2026; originally announced July 2026.

A Single Patch Is Not Enough: Deterministic Fusion of Repair Candidates

Authors: Boyang Yang, Xiangliang Hu, Luyao Ren, Yanjun Chen, Bach Le, Tegawendé F. Bissyandé, Haoye Tian

Abstract: Modern LLM coding agents are commonly evaluated using pass@k, but developers typically apply a single final patch in real-world settings. This pass@k-to-pass@1 gap is a post-generation problem: a candidate patch pool may contain a correct patch, but the system must decide which one to suggest to developers. Existing post-generation approaches mainly rank whole candidates, filter them with tests, o… ▽ More Modern LLM coding agents are commonly evaluated using pass@k, but developers typically apply a single final patch in real-world settings. This pass@k-to-pass@1 gap is a post-generation problem: a candidate patch pool may contain a correct patch, but the system must decide which one to suggest to developers. Existing post-generation approaches mainly rank whole candidates, filter them with tests, or query an LLM judge, but none deterministically reuse shared edit-atom evidence to both select and construct the final patch. Thus, we propose PatchFusion, a deterministic atomic evidence fusion approach for candidate patches that consults no test outcome at decision time. PatchFusion first fuses whole-diff agreement into a repair neighborhood, selects an auditable representative, and then applies evidence-constrained fusion (ECF) to retain repeated edit atoms and prune unsupported parts. To evaluate this setting, we build PatchFuseBench, a fixed-pool benchmark covering SWE-bench Verified, SWE-bench Multilingual, and Defects4J candidate patches. On PatchFuseBench, PatchFusion solves 426/500 bugs on SWE-bench Verified and 236/300 on SWE-bench Multilingual, and reaches 87/371 plausible patches on Defects4J, outperforming every matched candidate-pool selector on all three. PatchFusion recovers 41 and 27 bugs that no single source solves (30 and 18 more over the best single source). Ablation studies show that ECF adds +5/+6/+9 solved bugs by recovering in-pool repairs that selection misses, with no observed regression, and that PatchFusion's gains remain stable as candidate pools are resampled. On these complementary multi-source pools, cross-candidate evidence recovers more correct patches than the test-based and LLM-based selectors we evaluate, at orders-of-magnitude lower cost, reaching within 96.2% and 89.7% of the candidate-reachable ceiling on the two SWE-bench benchmarks. △ Less

Submitted 1 July, 2026; originally announced July 2026.

VLAFlow: A Unified Training Framework for Vision-Language-Action Models via Co-training and Future Latent Alignment

Authors: Guoyang Xia, Fengfa Li, Hongjin Ji, Lei Ren, Fangxiang Feng, Kun Zhan, Yan Xie

Abstract: Vision-language-action models (VLAs) have recently advanced robotic manipulation, yet the effects of different robot-data pre-training paradigms remain difficult to compare because existing models often differ in architecture, data, action space, and evaluation protocol. We present VLAFlow (Vision-Language-Action Flow), a unified flow-matching framework for controlled comparison of VLA training ob… ▽ More Vision-language-action models (VLAs) have recently advanced robotic manipulation, yet the effects of different robot-data pre-training paradigms remain difficult to compare because existing models often differ in architecture, data, action space, and evaluation protocol. We present VLAFlow (Vision-Language-Action Flow), a unified flow-matching framework for controlled comparison of VLA training objectives. Using a heterogeneous robot corpus, OXEMix, containing approximately 5,000 hours of data from DROID, OpenX-Embodiment, OpenX-Augmented, and RoboCOIN, we evaluate four paradigms under the same pi0-style architecture, shared VLM backbone, action expert, and 14-dimensional action space: action-only modeling (MindPI), language-supervised co-training (MindLPI), future latent alignment (MindWPI), and their combination (MindLWPI). Experiments on LIBERO, LIBERO-Plus, and SimplerEnv show that action-only pre-training is sensitive to heterogeneous data. In contrast, language supervision helps preserve vision-language generalization, while future latent alignment improves state-transition and action-outcome modeling. By combining both signals, MindLWPI achieves the most stable overall transfer performance across benchmarks. These results suggest a meta-action space view: language and future latent representations provide complementary intermediate constraints that make heterogeneous action supervision smoother and more transferable. △ Less

Submitted 4 August, 2026; v1 submitted 1 July, 2026; originally announced July 2026.

Mitigating Positional Leakage in 3D Masked Autoencoders for Robust Representation Learning

Authors: Xu Yan, Huiqun Wang, Chen Wang, Lei Ren, Di Huang

Abstract: Masked autoencoding has emerged as a prominent paradigm for self-supervised learning on 3D point clouds, achieving competitive performance across downstream tasks. Unlike its 2D counterpart, 3D masked autoencoding directly reconstructs spatial coordinates, making it inherently susceptible to positional leakage. In this work, we identify that the decoder in existing 3D MAE frameworks tends to over-… ▽ More Masked autoencoding has emerged as a prominent paradigm for self-supervised learning on 3D point clouds, achieving competitive performance across downstream tasks. Unlike its 2D counterpart, 3D masked autoencoding directly reconstructs spatial coordinates, making it inherently susceptible to positional leakage. In this work, we identify that the decoder in existing 3D MAE frameworks tends to over-rely on positional information, which weakens semantic representation learning and leads to suboptimal feature quality. To address this issue, we propose MPL-MAE, a masked point learning framework that mitigates positional over-reliance while enhancing the utilization of encoder features. Specifically, we introduce a recalibrated positional embedding module that suppresses metric-dominant coordinate signals while preserving geometric topology, together with a gated positional interface module that dynamically regulates positional injection during reconstruction. These designs promote a more balanced interaction between spatial priors and semantic features, yielding robust and informative representations. Extensive experiments across downstream tasks demonstrate that MPL-MAE consistently achieves competitive performance, validating its effectiveness. Code is available at https://github.com/yanx57/MPL-MAE. △ Less

Submitted 30 June, 2026; originally announced June 2026.

Thermo-responsive self-oscillating gel: mathematical model and theoretical analysis

Authors: Yunjie Wang, Ling Yuan, Lin Ren, Zihao Liu, Qingyu Gao

Abstract: Internally heated LCST thermo-responsive gels can show self-sustained swelling and collapse oscillations through feedback between temperature-induced collapse and collapse-suppressed heating. In this work, a minimal two-variable model is developed by coupling gel swelling dynamics with a lumped thermal balance. The analysis shows that stable large-amplitude oscillations are mainly controlled by gl… ▽ More Internally heated LCST thermo-responsive gels can show self-sustained swelling and collapse oscillations through feedback between temperature-induced collapse and collapse-suppressed heating. In this work, a minimal two-variable model is developed by coupling gel swelling dynamics with a lumped thermal balance. The analysis shows that stable large-amplitude oscillations are mainly controlled by global bifurcations of limit cycles, rather than by the local Hopf bifurcation. The Hopf bifurcation is subcritical in the studied parameter range, leading to a broad coexistence region where a stable fixed point and a stable limit cycle are both possible. The oscillatory behavior remains robust for different heating-gate functions, indicating that local linear instability is neither necessary nor sufficient for self-oscillation. Fast-slow analysis further shows that the oscillation period is mainly governed by the cooling rate, while the amplitude is determined by the geometry of the swelling equilibrium manifold. These results clarify the bifurcation mechanism of thermo-responsive gel oscillations and provide guidance for controlling their period, amplitude, and waveform. △ Less

Submitted 19 June, 2026; originally announced June 2026.

Comments: 16 pages, 6 figures

A Graphical Coaction for FRW Integrals from Partial/Relative Twisted (Co)homology

Authors: Andrew J. McLeod, Andrzej Pokraka, Lecheng Ren

Abstract: We construct a graphical coaction for Friedmann-Robertson-Walker (FRW) integrals at all loop orders in conformally-coupled scalar theories with non-conformal polynomial interactions. Our construction makes use of intersection theory in the context of (partial/relative) twisted (co)homology, which we use to decompose FRW integrals (and their discontinuities and derivatives) into building blocks tha… ▽ More We construct a graphical coaction for Friedmann-Robertson-Walker (FRW) integrals at all loop orders in conformally-coupled scalar theories with non-conformal polynomial interactions. Our construction makes use of intersection theory in the context of (partial/relative) twisted (co)homology, which we use to decompose FRW integrals (and their discontinuities and derivatives) into building blocks that can be represented as decorations of the original Feynman diagram. This facilitates a purely graphical description of the coaction, up to rational prefactors that can be read off from the graph. Our construction provides a comprehensive combinatorial framework for dissecting the analytic properties of cosmological observables; in particular, we demonstrate that the combinatorics of the differential equations that govern FRW integrals -- their so-called kinematic flow -- is a natural consequence of our coaction. We have also developed a user-friendly web application that computes the graphical coaction of any graph: https://frwcoaction.ca. Whenever possible, the web application also computes the differentials and discontinuities. A Mathematica notebook with the same functionality is also hosted at on a public GitHub repository. △ Less

Submitted 11 June, 2026; originally announced June 2026.

Comments: 60+8 pages; many figures

Report number: QMUL-PH-26-22

Performance of the Eos detector with water

Authors: Eos Collaboration, S. Arora, M. Askins, A. J. Bacon, Z. Bagdasarian, A. Baldoni, L. Bartoszek, M. Bergevin, Y. Bezawada, E. Blucher, J. Boissevain, R. Bonventre, E. J. Callaghan, D. F. Cowen, K. DeHolton, M. Diwan, M. Dubnowski, P. Englezos, S. Gadamsetty, C. Grant, B. Harris, M. R. Hebert, S. Jeon, T. Kaptanoglu, A. Katt , et al. (42 additional authors not shown)

Abstract: In this manuscript we present the first results from Eos, a four tonne optical detector located at the University of California, Berkeley. The primary goal of Eos is to demonstrate the performance capabilities of scintillation-based, 'hybrid' detector technology for future neutrino detectors. The data presented were collected while both the inner target vessel and the outer buffer vessel were fill… ▽ More In this manuscript we present the first results from Eos, a four tonne optical detector located at the University of California, Berkeley. The primary goal of Eos is to demonstrate the performance capabilities of scintillation-based, 'hybrid' detector technology for future neutrino detectors. The data presented were collected while both the inner target vessel and the outer buffer vessel were filled with water. The water target acts as a well-understood medium that produces only Cherenkov light, which can be used to calibrate and develop the detector model and reconstruction algorithms prior to the deployment of scintillating material. Using deployed optical and radioactive calibration sources, a series of detailed detector calibrations are performed. These enable a suite of tests for various reconstruction algorithms. Simulations that use calibrated models are compared with the data across a variety of different types of calibration sources, source positions, and rotations. △ Less

Submitted 8 June, 2026; originally announced June 2026.

OneReason Technical Report

Authors: OneRec Team, Biao Yang, Boyang Ding, Chenglong Chu, Dunju Zang, Fei Pan, Han Li, Hao Jiang, Honghui Bao, Huanjie Wang, Jian Liang, Jiangxia Cao, Jiao Ou, Jiaxin Deng, Jinghao Zhang, Kun Gai, Lu Ren, Peiru Du, Pengfei Zheng, Rongzhou Zhang, Ruiming Tang, Shiyao Wang, Siyang Mao, Siyuan Lou, Teng Shi , et al. (59 additional authors not shown)

Abstract: Generative recommendation models in the OneRec family have been widely deployed in many real-world services, such as short-video, live-streaming, advertising, and e-commerce. However, these generative models can only benefit from the scaling advantage, while their reasoning ability is hard to activate, since we cannot construct meaningful Chain-of-Thought (CoT) sequences consisting of itemic token… ▽ More Generative recommendation models in the OneRec family have been widely deployed in many real-world services, such as short-video, live-streaming, advertising, and e-commerce. However, these generative models can only benefit from the scaling advantage, while their reasoning ability is hard to activate, since we cannot construct meaningful Chain-of-Thought (CoT) sequences consisting of itemic tokens only. Inspired by the success of the reasoning-style ``think before answer'' paradigm in the LLM field, we conduct preliminary studies (i.e., OneRec-Think, OpenOneRec) to explore reasoning capability in generative recommendation. Nevertheless, we notice an unexpected phenomenon: the thinking mode does not show advantages over the non-thinking mode. Drawing insights from recent findings on CoT robustness in multi-modal language models, we argue that effective reasoning in recommendation rests on two factors: perception, the ability to ground itemic tokens in their underlying language semantics, and cognition, the ability to reorganize a user's behavior sequence into coherent latent interest points. We therefore propose OneReason, which includes: (1) strong itemic token perception in pre-training, (2) a three-level cognition-enhanced CoT format for recommendation tasks in SFT, and (3) a specialize-then-unify training recipe in RL to enhance the thinking ability. △ Less

Submitted 4 June, 2026; originally announced June 2026.

Comments: Work in progress

Scaling Expert Feedback with Reflective Edit Propagation in Compositional Knowledge Bases

Authors: Jiajing Guo, Xueming Li, Jorge Piazentin Ono, Wenbin He, Liu Ren

Abstract: Domain-specific knowledge bases (KBs) encode vertical expertise and proprietary information that organizations depend on, but curating them at scale is a persistent challenge. Although Large Language Models (LLMs) can draft initial entries efficiently, technical accuracy still requires human expert validation, and reviewing entries one by one at scale is impractical. We present Reflective Agent fo… ▽ More Domain-specific knowledge bases (KBs) encode vertical expertise and proprietary information that organizations depend on, but curating them at scale is a persistent challenge. Although Large Language Models (LLMs) can draft initial entries efficiently, technical accuracy still requires human expert validation, and reviewing entries one by one at scale is impractical. We present Reflective Agent for Identifier Dictionary (RAID), a novel system that transforms individual expert edits into systematic knowledge updates. Unlike traditional "correct-and-save" paradigms, RAID utilizes a reflective agent to infer the underlying semantic intent behind a single expert edit and propagates that correction across the entire KB through a three-step architecture: Intent Inference, Reflection-based Planning, and User Controlled Execution. We evaluated the reflection and propagation performance on a public dataset and conducted a user study with subject matter experts with proprietary data. The evaluation shows RAID's technical feasibility in capturing expert intent and its potential to scale specialized expertise across industrial knowledge bases. △ Less

Submitted 3 June, 2026; originally announced June 2026.

Comments: Accepted to ACM CAIS '26 Demo Track

Journal ref: ACM Conference on AI and Agentic Systems (CAIS '26), May 26-29, 2026, San Jose, CA, USA

The First Nozzle-Mounted Compton Camera Prompt Gamma Imaging System for In Vivo Proton Therapy Dose Verification

Authors: Farshad Safavi, Stephen W. Peterson, Sina Mossahebi, Ananta Chalise, Vijay R. Sharma, Matthias K. Gobbert, Jerimy C. Polf, Lei Ren

Abstract: This study presents the first clinical integration and experimental demonstration of a nozzle-mounted Compton camera prompt gamma imaging (PGI) system for in vivo proton range verification. Four position-sensitive solid-state Compton camera modules, each containing four cadmium zinc telluride (CdZnTe) detector crystals, were integrated into a modified range shifter mounted directly on the treatmen… ▽ More This study presents the first clinical integration and experimental demonstration of a nozzle-mounted Compton camera prompt gamma imaging (PGI) system for in vivo proton range verification. Four position-sensitive solid-state Compton camera modules, each containing four cadmium zinc telluride (CdZnTe) detector crystals, were integrated into a modified range shifter mounted directly on the treatment nozzle of a clinical proton therapy gantry. This compact fixed-geometry configuration maintained alignment with the proton beam axis throughout irradiation and enabled stable synchronized data acquisition during pencil-beam scanning delivery. The system was evaluated under realistic clinical proton beam delivery conditions using single-energy and spread-out Bragg peak (SOBP) irradiations at gantry angles of 90° and 270°, delivered doses of 2 Gy and 7.5 Gy, and controlled distal range shifts of up to 10 mm. Prompt gamma events were reconstructed into three-dimensional emission distributions using a physics-based Compton scatter reconstruction framework. The system operated reliably during all irradiations and produced reproducible prompt-gamma localization across repeated measurements. Reconstructed emission distributions remained geometrically consistent across gantry angles and demonstrated sensitivity to controlled distal range perturbations, with measurable upstream shifts of the emission hotspot corresponding to reduced proton penetration depth. These results demonstrate the feasibility of a clinically integrated nozzle-mounted quad-camera Compton PGI system for detecting millimeter-scale proton range variations during beam delivery and represent an important step toward clinically deployable prompt gamma-based in vivo treatment verification in proton therapy. △ Less

Submitted 2 June, 2026; originally announced June 2026.

Comments: 10 pages, 9 figures. Submitted to IEEE Transactions on Medical Imaging

Contrastive Augmented Transformer with Domain-specific Enhancement for Robust Multi-scenario Metal Surface Defect Detection

Authors: Yiyao Liu, Wenxiao He, Liyuan Ren, Huan Wang

Abstract: Metal surface defect detection is critical for maintaining product quality in industrial manufacturing. However, it faces significant challenges, including limited annotated data, difficulty in identifying subtle multi-scale defects, and poor generalization across diverse scenarios. To address these issues, this paper proposes a novel Contrastive Augmented Transformer (CAT) framework for robust de… ▽ More Metal surface defect detection is critical for maintaining product quality in industrial manufacturing. However, it faces significant challenges, including limited annotated data, difficulty in identifying subtle multi-scale defects, and poor generalization across diverse scenarios. To address these issues, this paper proposes a novel Contrastive Augmented Transformer (CAT) framework for robust defect detection. CAT employs a hierarchical Swin Transformer backbone and redesigns the feature pyramid network to effectively fuse low-level textures with high-level semantics, enabling precise modeling of subtle and multi-scale defect patterns. To enhance robustness under real-world noise conditions, we propose a domain-specific droplet augmentation algorithm. Furthermore, we incorporate a hard negative mining strategy into the contrastive loss to strengthen the model's discrimination ability in ambiguous defect regions. Experimental results on the KolektorSDD2 dataset demonstrate that CAT achieves a pixel-level AUROC of 99.54%, outperforming existing methods. In addition, CAT exhibits superior generalization and robustness on three unseen datasets, including KSDD1, MTD for tile defects, and MSDD for rail surface defects, demonstrating its potential for wide-scale industrial deployment. △ Less

Submitted 2 June, 2026; v1 submitted 1 June, 2026; originally announced June 2026.

Regime-Adaptive Continual Learning for Portfolio Management

Authors: Chaofan Pan, Lingfei Ren, Linbo Xiong, Yonghao Li, Wei Wei, Xin Yang

Abstract: Financial markets are inherently non-stationary, exhibiting frequent regime shifts and structural changes that render traditional Portfolio Management (PM) approaches ineffective. Existing remedies, such as rolling-window retraining and naive online fine-tuning, are hindered by high computational costs and insufficient knowledge utilization, respectively, resulting in low returns and limited adapt… ▽ More Financial markets are inherently non-stationary, exhibiting frequent regime shifts and structural changes that render traditional Portfolio Management (PM) approaches ineffective. Existing remedies, such as rolling-window retraining and naive online fine-tuning, are hindered by high computational costs and insufficient knowledge utilization, respectively, resulting in low returns and limited adaptability. Continual learning (CL) offers a promising paradigm by enabling trading agents to accumulate and transfer knowledge across sequential tasks. In this paper, we propose \textbf{Re}gime-aware \textbf{C}ontinual \textbf{A}daptive \textbf{P}ortfolio management (\textbf{ReCAP}), a novel framework that integrates CL into PM to address the challenges of dynamic financial environments. ReCAP employs an adaptive regime detection module to segment historical market data into variable-length regimes, enabling regime-specific learning of policy vectors and the construction of a policy library. During continual trading, a regime-gate module adaptively combines policy vectors from the library based on the current market state, facilitating rapid adaptation to newly detected regimes. Only the regime-gate and the current regime's policy vector are continually updated to preserve useful knowledge effectively. Extensive experiments on five real-world datasets demonstrate that ReCAP consistently outperforms popular baselines, achieving superior returns in long-term investment horizons and rapid adaptation to regime shifts. △ Less

Submitted 28 May, 2026; originally announced June 2026.

Comments: Accepted by KDD 2026

Sequential quantum nonlocality sharing under local noisy quantum channels

Authors: Na Li, Chen-Yue Li, Yu-Hong Zheng, Wen-Long Ma, Li-Hang Ren, Yan-Kui Bai

Abstract: Sequential sharing of quantum nonlocality (SSQN) is crucial for device-independent tasks in quantum information processing, wherein relaying the post-measurement qubit through a local quantum channel to a subsequent observer constitutes an essential operational step. Here we present a theoretical analysis of noise robustness of sequential sharing for bipartite Bell and tripartite Mermin nonlocalit… ▽ More Sequential sharing of quantum nonlocality (SSQN) is crucial for device-independent tasks in quantum information processing, wherein relaying the post-measurement qubit through a local quantum channel to a subsequent observer constitutes an essential operational step. Here we present a theoretical analysis of noise robustness of sequential sharing for bipartite Bell and tripartite Mermin nonlocality under the influence of local phase-flip, bit-flip, and depolarizing quantum channels. It is proved that arbitrarily many independent observers can sequentially share the quantum nonlocality of Bell, Greenberger-Horne-Zeilinger, and W states via respective noise-immune channels, whereas such unbound feature of SSQN is lost under other local noisy quantum channels. Furthermore, we demonstrate that the noise-immune channel enabling unbounded SSQN can be switched by employing our newly designed measurement strategies assisted by local unitary operations on the initial entangled states. Moreover, as illustrative examples of noise robustness, we propose two concrete schemes for sharing Bell and Mermin nonlocality with two sequential local observers on one side subject to local noisy channels. Our work establishes a practical framework for realizing the SSQN under noisy quantum channels, and reveals the connection between noise robustness and measurement strategies. △ Less

Submitted 1 June, 2026; v1 submitted 26 May, 2026; originally announced May 2026.

Comments: 19 pages, 7 figures

Latent Recurrent Transformer: Architecture Exploration, Training Strategies, and Scaling Behavior

Authors: Zeyi Huang, Xuehai He, LiLiang Ren, Yiping Wang, Baolin Peng, Hao Cheng, Shuohang Wang, Pengcheng He, Jianfeng Gao, Yong Jae Lee, Yelong Shen

Abstract: We study Latent Recurrent Transformer (LRT), a lightweight augmentation of autoregressive transformers that reuses a high-level source-layer hidden state from the previous token as recurrent memory for the next token. Because this source state is already computed during ordinary decoding, LRT adds a cross-layer recurrent latent pathway across positions without inserting pause tokens or extra depth… ▽ More We study Latent Recurrent Transformer (LRT), a lightweight augmentation of autoregressive transformers that reuses a high-level source-layer hidden state from the previous token as recurrent memory for the next token. Because this source state is already computed during ordinary decoding, LRT adds a cross-layer recurrent latent pathway across positions without inserting pause tokens or extra depth loops, and the standard attention mechanism and KV-cache interface are preserved. To pretrain this recurrence at scale without sequentially unrolling the transformer, we introduce interleaved parallel training: a single full-sequence initialization forward pass builds a shared buffer; then disjoint position subsets are refined in parallel and written back, so that all tokens receive recurrent-memory-aware supervision at roughly 2 times baseline compute. Across nanochat style backbones and a wide range of tokens-per-parameter budgets, LRT improves both language-modeling loss and in-context learning under matched effective compute while adding as little as 0.3% parameters. △ Less

Submitted 26 May, 2026; originally announced May 2026.

Dense2MoE: Pushing the Pareto Frontier of On-Device LLMs via Unified Pruning and Upcycling

Authors: Fengfa Li, Hongjin Ji, Yifeng Ding, Lei Ren, Chen Wei

Abstract: The Mixture of Experts MoE architecture is highly promising for resource constrained on device deployments yet training these models from scratch incurs prohibitive costs Current methods attempt to alleviate this by upcycling dense models into MoEs however they often introduce parameter redundancy that degrades inference efficiency Alternatively standard layer pruning mitigates redundancy but inev… ▽ More The Mixture of Experts MoE architecture is highly promising for resource constrained on device deployments yet training these models from scratch incurs prohibitive costs Current methods attempt to alleviate this by upcycling dense models into MoEs however they often introduce parameter redundancy that degrades inference efficiency Alternatively standard layer pruning mitigates redundancy but inevitably compromises model accuracy To resolve this dilemma we propose Dense2MoE a novel framework that unifies pruning and upcycling through Layer Fusion UpCycling LF UC Guided by hardware Roofline theory Dense2MoE systematically overcomes the inference memory wall by pruning bandwidth heavy attention modules from redundant layers while repurposing their Multi Layer Perceptrons MLPs into MoE experts This structural innovation preserves the models core capabilities and strictly limits active parameters via selective token routing With a modest continual pre training budget Dense2MoE efficiently converts publicly available dense LLMs into on device ready MoE models Extensive experiments demonstrate that Dense2MoE significantly advances the Pareto frontier for on device inference latency versus model accuracy outperforming dense baselines state of the art compression and standard upcycling methods △ Less

Submitted 25 May, 2026; originally announced May 2026.

Comments: 19 pages

Pantheon360: Taming Digital Twin Generation via 3D-Aware 360° Video Diffusion

Authors: Ting-Hsuan Chen, Ying-Huan Chen, Tao Tu, Jie-Ying Lee, Cho-Ying Wu, Fangzhou Lin, Hengyuan Zhang, David Paz, Xinyu Huang, Yuliang Guo, Yu-Lun Liu, Yue Wang, Liu Ren

Abstract: Generating complete digital twins from videos requires precise camera control, global scene coverage, and strict spatial-temporal consistency constraints that remain challenging for perspective video generators due to their limited field of view (FoV). Their narrow FoV forces long or multi-view trajectories, amplifying cross-view inconsistency and temporal drift. We argue that 360° video generatio… ▽ More Generating complete digital twins from videos requires precise camera control, global scene coverage, and strict spatial-temporal consistency constraints that remain challenging for perspective video generators due to their limited field of view (FoV). Their narrow FoV forces long or multi-view trajectories, amplifying cross-view inconsistency and temporal drift. We argue that 360° video generation offers a natural solution: panoramic coverage simplifies trajectory design and provides a strong global context for maintaining coherence. We introduce Pantheon360: Taming Digital Twin Generation via 3D-Aware 360° Video Diffusion, a controllable 360° video generation framework that synthesizes high-fidelity videos from sparse 360° inputs. The key idea is an explicit 3D Cache, reconstructed from the input, which serves as a geometric scaffold for any user-defined camera path. This allows the diffusion model to focus on photorealistic texture refinement while the 3D Cache enforces global geometric consistency. Experiments show that Pantheon360 achieves superior visual quality and unmatched geometric coherence, enabling reliable and flexible 360° scene generation for downstream simulation and digital-twin applications. △ Less

Submitted 15 June, 2026; v1 submitted 25 May, 2026; originally announced May 2026.

Comments: Accepted to CVPR 2026. Project page: https://koi953215.github.io/pantheon360_page/

Lngram: N-gram Conditional Memory in Latent Space

Authors: Yunao Zheng, Guoyang Xia, Xiaojie Wang, Lei Ren

Abstract: Sequence modeling requires both compositional reasoning and local static knowledge retrieval, yet standard Transformers handle both through dense computation. Engram partially decouples retrieval from the backbone, but its token-based keys remain tied to text tokenization and hash compression. We propose Lngram, a latent-space conditional memory module that learns discrete symbols directly from hi… ▽ More Sequence modeling requires both compositional reasoning and local static knowledge retrieval, yet standard Transformers handle both through dense computation. Engram partially decouples retrieval from the backbone, but its token-based keys remain tied to text tokenization and hash compression. We propose Lngram, a latent-space conditional memory module that learns discrete symbols directly from hidden states and performs N-gram lookup over these symbols. This design removes the dependence on tokenizer IDs and naturally extends to non-text modalities. In our evaluated settings, Lngram outperforms Transformer and Engram baselines, consistently reduces perplexity in long-context language modeling, and effectively injects domain knowledge when added post hoc to pretrained models. Joint training with the backbone further surpasses full fine-tuning, while experiments on vision-language and vision-language-action tasks show overall gains. Analyses with LogitLens and CKA suggest that Lngram enables prediction-relevant information to emerge earlier, increasing effective depth with limited inference and memory overhead. Code is available at https://github.com/zyaaa-ux/Lngram. △ Less

Submitted 24 May, 2026; originally announced May 2026.

de Sitter Wavefunction from Quadrangular Polylogarithms: Chain Graphs

Authors: Livia Ferro, Tomasz Lukowski, Lecheng Ren, Marcus Spradlin, Anastasia Volovich, He-Chen Weng, Yao-Qi Zhang

Abstract: We present an explicit formula for the $n$-site chain graph contribution to the cosmological wavefunction for conformally coupled $φ^3$ theory in de Sitter space. Our result relies on the recent finding that the symbol of this function satisfies total compatibility with respect to the $A_{2n-2}$ cluster algebra, and that Rudenko's quadrangular polylogarithms provide, by construction, a complete ba… ▽ More We present an explicit formula for the $n$-site chain graph contribution to the cosmological wavefunction for conformally coupled $φ^3$ theory in de Sitter space. Our result relies on the recent finding that the symbol of this function satisfies total compatibility with respect to the $A_{2n-2}$ cluster algebra, and that Rudenko's quadrangular polylogarithms provide, by construction, a complete basis for such functions. We prove our formula by directly relating a recursive set of differential equations satisfied by these wavefunction coefficients to a recursive coproduct formula for quadrangular polylogarithms. △ Less

Submitted 7 May, 2026; originally announced May 2026.

Comments: 28 pages, 2 figures

Kwai Summary Attention Technical Report

Authors: Chenglong Chu, Guorui Zhou, Guowang Zhang, Han Li, Hao Peng, Hongtao Cheng, Hui Wang, Jian Liang, Jiangxia Cao, Kun Gai, Lingzhi Zhou, Lu Ren, Qi Zhang, Ruiming Tang, Ruitao Wang, Xinchen Luo, Yi Su, Zhiyuan Liang, Ziqi Wang, Boyang Ding, Chengru Song, Dunju Zang, Jiao Ou, Jiaxin Deng, Jijun Shi , et al. (13 additional authors not shown)

Abstract: Long-context ability, has become one of the most important iteration direction of next-generation Large Language Models, particularly in semantic understanding/reasoning, code agentic intelligence and recommendation system. However, the standard softmax attention exhibits quadratic time complexity with respect to sequence length. As the sequence length increases, this incurs substantial overhead i… ▽ More Long-context ability, has become one of the most important iteration direction of next-generation Large Language Models, particularly in semantic understanding/reasoning, code agentic intelligence and recommendation system. However, the standard softmax attention exhibits quadratic time complexity with respect to sequence length. As the sequence length increases, this incurs substantial overhead in long-context settings, leading the training and inference costs of extremely long sequences deteriorate rapidly. Existing solutions mitigate this issue through two technique routings: i) Reducing the KV cache per layer, such as from the head-level compression GQA, and the embedding dimension-level compression MLA, but the KV cache remains linearly dependent on the sequence length at a 1:1 ratio. ii) Interleaving with KV Cache friendly architecture, such as local attention SWA, linear kernel GDN, but often involve trade-offs among KV Cache and long-context modeling effectiveness. Besides the two technique routings, we argue that there exists an intermediate path not well explored: {Maintaining a linear relationship between the KV cache and sequence length, but performing semantic-level compression through a specific ratio $k$}. This $O(n/k)$ path does not pursue a ``minimum KV cache'', but rather trades acceptable memory costs for complete, referential, and interpretable retention of long distant dependency. Motivated by this, we propose Kwai Summary Attention (KSA), a novel attention mechanism that reduces sequence modeling cost by compressing historical contexts into learnable summary tokens. △ Less

Submitted 5 July, 2026; v1 submitted 27 April, 2026; originally announced April 2026.

Comments: update related works

EmbodiedMidtrain: Bridging the Gap between Vision-Language Models and Vision-Language-Action Models via Mid-training

Authors: Yiyang Du, Zhanqiu Guo, Xin Ye, Liu Ren, Chenyan Xiong

Abstract: Vision-Language-Action Models (VLAs) inherit their visual and linguistic capabilities from Vision-Language Models (VLMs), yet most VLAs are built from off-the-shelf VLMs that are not adapted to the embodied domain, limiting their downstream performance. In this work, we propose EmbodiedMidtrain to bridge the gap between VLMs and VLAs. We first characterize the data distribution gap between them, s… ▽ More Vision-Language-Action Models (VLAs) inherit their visual and linguistic capabilities from Vision-Language Models (VLMs), yet most VLAs are built from off-the-shelf VLMs that are not adapted to the embodied domain, limiting their downstream performance. In this work, we propose EmbodiedMidtrain to bridge the gap between VLMs and VLAs. We first characterize the data distribution gap between them, showing that VLA data occupy compact regions that are largely separated from the broader VLM distribution, while the degree of alignment varies substantially both across and within VLM data sources. Then, we build a mid-training data engine that leverages a lightweight learnable proximity estimator to select the most VLA-aligned candidates from a large VLM pool, and mid-trains the VLM on this curated mixture before downstream VLA fine-tuning. Experiments on three robot manipulation benchmarks show that mid-training consistently improves performance across different VLM backbones, achieving results competitive with expert VLAs and off-the-shelf VLMs trained with larger model scale and training budgets. Further analysis reveals that mid-training provides a stronger initialization for VLA fine-tuning, with gains emerging from the earliest steps and widening throughout training. Moreover, the data engine captures both dataset-level and sample-level alignment signals, favoring spatial reasoning over text-centric tasks while preserving the diversity of the VLM data. We will release all code, data and models for future research. △ Less

Submitted 21 April, 2026; originally announced April 2026.

Momentum-constrained Hybrid Heuristic Trajectory Optimization Framework with Residual-enhanced DRL for Visually Impaired Scenarios

Authors: Yuting Zeng, Zhiwen Zheng, Jingya Wang, You Zhou, JiaLing Xiao, Yongbin Yu, Manping Fan, Bo Gong, Liyong Ren

Abstract: Safe and efficient assistive planning for visually impaired scenarios remains challenging, since existing methods struggle with multi-objective optimization, generalization, and interpretability. In response, this paper proposes a Momentum-Constrained Hybrid Heuristic Trajectory Optimization Framework (MHHTOF). To balance multiple objectives of comfort and safety, the framework designs a Heuristic… ▽ More Safe and efficient assistive planning for visually impaired scenarios remains challenging, since existing methods struggle with multi-objective optimization, generalization, and interpretability. In response, this paper proposes a Momentum-Constrained Hybrid Heuristic Trajectory Optimization Framework (MHHTOF). To balance multiple objectives of comfort and safety, the framework designs a Heuristic Trajectory Sampling Cluster (HTSC) with a Momentum-Constrained Trajectory Optimization (MTO), which suppresses abrupt velocity and acceleration changes. In addition, a novel residual-enhanced deep reinforcement learning (DRL) module refines candidate trajectories, advancing temporal modeling and policy generalization. Finally, a dual-stage cost modeling mechanism (DCMM) is introduced to regulate optimization, where costs in the Frenet space ensure consistency, and reward-driven adaptive weights in the Cartesian space integrate user preferences for interpretability and user-centric decision-making. Experimental results show that the proposed framework converges in nearly half the iterations of baselines and achieves lower and more stable costs. In complex dynamic scenarios, MHHTOF further demonstrates stable velocity and acceleration curves with reduced risk, confirming its advantages in robustness, safety, and efficiency. △ Less

Submitted 16 April, 2026; originally announced April 2026.

Comments: 24 pages, 14 figures. arXiv admin note: text overlap with arXiv:2509.15582

Shuffle the Context: RoPE-Perturbed Self-Distillation for Long-Context Adaptation

Authors: Zichong Li, Chen Liang, Liliang Ren, Tuo Zhao, Yelong Shen, Weizhu Chen

Abstract: Large language models (LLMs) increasingly operate in settings that require reliable long-context understanding, such as retrieval-augmented generation and multi-document reasoning. A common strategy is to fine-tune pretrained short-context models at the target sequence length. However, we find that standard long-context adaptation can remain brittle: model accuracy depends strongly on the absolute… ▽ More Large language models (LLMs) increasingly operate in settings that require reliable long-context understanding, such as retrieval-augmented generation and multi-document reasoning. A common strategy is to fine-tune pretrained short-context models at the target sequence length. However, we find that standard long-context adaptation can remain brittle: model accuracy depends strongly on the absolute placement of relevant evidence, exhibiting high positional variance even when controlling for task format and difficulty. We propose RoPE-Perturbed Self-Distillation, a training regularizer that improves positional robustness. The core idea is to form alternative "views" of the same training sequence by perturbing its RoPE indices -- effectively moving parts of the context to different positions -- and to train the model to produce consistent predictions across views via self-distillation. This encourages reliance on semantic signals instead of brittle position dependencies. Experiments on long-context adaptation of Llama-3-8B and Qwen-3-4B demonstrate consistent gains on long-context benchmarks, including up to 12.04% improvement on RULER-64K for Llama-3-8B and 2.71% on RULER-256K for Qwen-3-4B after SFT, alongside improved length extrapolation beyond the training context window. △ Less

Submitted 15 April, 2026; originally announced April 2026.

LogicPoison: Logical Attacks on Graph Retrieval-Augmented Generation

Authors: Yilin Xiao, Jin Chen, Qinggang Zhang, Yujing Zhang, Chuang Zhou, Longhao Yang, Lingfei Ren, Xin Yang, Xiao Huang

Abstract: Graph-based Retrieval-Augmented Generation (GraphRAG) enhances the reasoning capabilities of Large Language Models (LLMs) by grounding their responses in structured knowledge graphs. Leveraging community detection and relation filtering techniques, GraphRAG systems demonstrate inherent resistance to traditional RAG attacks, such as text poisoning and prompt injection. However, in this paper, we fi… ▽ More Graph-based Retrieval-Augmented Generation (GraphRAG) enhances the reasoning capabilities of Large Language Models (LLMs) by grounding their responses in structured knowledge graphs. Leveraging community detection and relation filtering techniques, GraphRAG systems demonstrate inherent resistance to traditional RAG attacks, such as text poisoning and prompt injection. However, in this paper, we find that the security of GraphRAG systems fundamentally relies on the topological integrity of the underlying graph, which can be undermined by implicitly corrupting the logical connections, without altering surface-level text semantics. To exploit this vulnerability, we propose \textsc{LogicPoison}, a novel attack framework that targets logical reasoning rather than injecting false contents. Specifically, \textsc{LogicPoison} employs a type-preserving entity swapping mechanism to perturb both global logic hubs for disrupting overall graph connectivity and query-specific reasoning bridges for severing essential multi-hop inference paths. This approach effectively reroutes valid reasoning into dead ends while maintaining surface-level textual plausibility. Comprehensive experiments across multiple benchmarks demonstrate that \textsc{LogicPoison} successfully bypasses GraphRAG's defenses, significantly degrading performance and outperforming state-of-the-art baselines in both effectiveness and stealth. Our code is available at \textcolor{blue}https://github.com/Jord8061/logicPoison. △ Less

Submitted 3 April, 2026; originally announced April 2026.

ExploreVLA: Dense World Modeling and Exploration for End-to-End Autonomous Driving

Authors: Zihao Sheng, Xin Ye, Jingru Luo, Sikai Chen, Liu Ren

Abstract: End-to-end autonomous driving models based on Vision-Language-Action (VLA) architectures have shown promising results by learning driving policies through behavior cloning on expert demonstrations. However, imitation learning inherently limits the model to replicating observed behaviors without exploring diverse driving strategies, leaving it brittle in novel or out-of-distribution scenarios. Rein… ▽ More End-to-end autonomous driving models based on Vision-Language-Action (VLA) architectures have shown promising results by learning driving policies through behavior cloning on expert demonstrations. However, imitation learning inherently limits the model to replicating observed behaviors without exploring diverse driving strategies, leaving it brittle in novel or out-of-distribution scenarios. Reinforcement learning (RL) offers a natural remedy by enabling policy exploration beyond the expert distribution. Yet VLA models, typically trained on offline datasets, lack directly observable state transitions, necessitating a learned world model to anticipate action consequences. In this work, we propose a unified understanding-and-generation framework that leverages world modeling to simultaneously enable meaningful exploration and provide dense supervision. Specifically, we augment trajectory prediction with future RGB and depth image generation as dense world modeling objectives, requiring the model to learn fine-grained visual and geometric representations that substantially enrich the planning backbone. Beyond serving as a supervisory signal, the world model further acts as a source of intrinsic reward for policy exploration: its image prediction uncertainty naturally measures a trajectory's novelty relative to the training distribution, where high uncertainty indicates out-of-distribution scenarios that, if safe, represent valuable learning opportunities. We incorporate this exploration signal into a safety-gated reward and optimize the policy via Group Relative Policy Optimization (GRPO). Experiments on the NAVSIM and nuScenes benchmarks demonstrate the effectiveness of our approach, achieving a state-of-the-art PDMS score of 93.7 and an EPDMS of 88.8 on NAVSIM. The code is available at https://zihaosheng.github.io/ExploreVLA/. △ Less

Submitted 28 June, 2026; v1 submitted 3 April, 2026; originally announced April 2026.

Comments: Accepted to ECCV 2026. The code is available at https://zihaosheng.github.io/ExploreVLA/

Rethinking Language Model Scaling under Transferable Hypersphere Optimization

Authors: Liliang Ren, Yang Liu, Yelong Shen, Weizhu Chen

Abstract: Scaling laws for large language models depend critically on the optimizer and parameterization. Existing hyperparameter transfer laws are mainly developed for first-order optimizers, and they do not structurally prevent training instability at scale. Recent hypersphere optimization methods constrain weight matrices to a fixed-norm hypersphere, offering a promising alternative for more stable scali… ▽ More Scaling laws for large language models depend critically on the optimizer and parameterization. Existing hyperparameter transfer laws are mainly developed for first-order optimizers, and they do not structurally prevent training instability at scale. Recent hypersphere optimization methods constrain weight matrices to a fixed-norm hypersphere, offering a promising alternative for more stable scaling. We introduce HyperP (Hypersphere Parameterization), the first framework for transferring optimal learning rates across model width, depth, training tokens, and Mixture-of-Experts (MoE) granularity under the Frobenius-sphere constraint with the Muon optimizer. We prove that weight decay is a first-order no-op on the Frobenius sphere, show that Depth-$μ$P remains necessary, and find that the optimal learning rate follows the same data-scaling power law with the "magic exponent" 0.32 previously observed for AdamW. A single base learning rate tuned at the smallest scale transfers across all compute budgets under HyperP, yielding $1.58\times$ compute efficiency over a strong Muon baseline at $6\times10^{21}$ FLOPs. Moreover, HyperP delivers transferable stability: all monitored instability indicators, including $Z$-values, output RMS, and activation outliers, remain bounded and non-increasing under training FLOPs scaling. We also propose SqrtGate, an MoE gating mechanism derived from the hypersphere constraint that preserves output RMS across MoE granularities for improved granularity scaling, and show that hypersphere optimization enables substantially larger auxiliary load-balancing weights, yielding both strong performance and good expert balance. We release our training codebase at https://github.com/microsoft/ArchScale. △ Less

Submitted 4 April, 2026; v1 submitted 30 March, 2026; originally announced March 2026.

UniDAC: Universal Metric Depth Estimation for Any Camera

Authors: Girish Chandar Ganesan, Yuliang Guo, Liu Ren, Xiaoming Liu

Abstract: Monocular metric depth estimation (MMDE) is a core challenge in computer vision, playing a pivotal role in real-world applications that demand accurate spatial understanding. Although prior works have shown promising zero-shot performance in MMDE, they often struggle with generalization across diverse camera types, such as fisheye and $360^\circ$ cameras. Recent advances have addressed this throug… ▽ More Monocular metric depth estimation (MMDE) is a core challenge in computer vision, playing a pivotal role in real-world applications that demand accurate spatial understanding. Although prior works have shown promising zero-shot performance in MMDE, they often struggle with generalization across diverse camera types, such as fisheye and $360^\circ$ cameras. Recent advances have addressed this through unified camera representations or canonical representation spaces, but they require either including large-FoV camera data during training or separately trained models for different domains. We propose UniDAC, an MMDE framework that presents universal robustness in all domains and generalizes across diverse cameras using a single model. We achieve this by decoupling metric depth estimation into relative depth prediction and spatially varying scale estimation, enabling robust performance across different domains. We propose a lightweight Depth-Guided Scale Estimation module that upsamples a coarse scale map to high resolution using the relative depth map as guidance to account for local scale variations. Furthermore, we introduce RoPE-$φ$, a distortion-aware positional embedding that respects the spatial warping in Equi-Rectangular Projections (ERP) via latitude-aware weighting. UniDAC achieves state of the art (SoTA) in cross-camera generalization by consistently outperforming prior methods across all datasets. △ Less

Submitted 8 April, 2026; v1 submitted 27 March, 2026; originally announced March 2026.

A Graphical Coaction for FRW Wavefunction Coefficients

Authors: Andrew McLeod, Andrzej Pokraka, Lecheng Ren

Abstract: We show that the wavefunction of the universe in theories of conformally coupled scalars in power-law Friedmann-Robertson-Walker (FRW) cosmologies satisfies a graphical coaction, by means of which we can understand its complete analytic structure in terms of the acyclic minors of Feynman graphs. Our construction extends to all particle multiplicities and any loop order, and if we isolate certain w… ▽ More We show that the wavefunction of the universe in theories of conformally coupled scalars in power-law Friedmann-Robertson-Walker (FRW) cosmologies satisfies a graphical coaction, by means of which we can understand its complete analytic structure in terms of the acyclic minors of Feynman graphs. Our construction extends to all particle multiplicities and any loop order, and if we isolate certain weight-one contributions, it reproduces the ``kinematic flow'' that encodes the differential equation of the wavefunction coefficients. Similarly, any discontinuity of the wavefunction coefficient is easily extracted from the coaction. △ Less

Submitted 26 March, 2026; originally announced March 2026.

Comments: 5+2 pages

Self-Evolving Multi-Agent Framework for Efficient Decision Making in Real-Time Strategy Scenarios

Authors: Li Ma, Hao Peng, Yiming Wang, Hongbin Luo, Jie Liu, Kongjing Gu, Guanlin Wu, Hui Lin, Lei Ren

Abstract: Large language models (LLMs) have demonstrated exceptional potential in complex reasoning,pioneering a new paradigm for autonomous agent decision making in dynamic settings. However, in Real-Time Strategy (RTS) scenarios, LLMs suffer from a critical speed-quality trade-off. Specifically expansive state spaces and time limits render inference delays prohibitive, while stochastic planning errors und… ▽ More Large language models (LLMs) have demonstrated exceptional potential in complex reasoning,pioneering a new paradigm for autonomous agent decision making in dynamic settings. However, in Real-Time Strategy (RTS) scenarios, LLMs suffer from a critical speed-quality trade-off. Specifically expansive state spaces and time limits render inference delays prohibitive, while stochastic planning errors undermine logical consistency. To address these challenges, we present SEMA (Self-Evolving Multi-Agent), a novel framework designed for high-performance, low-latency decision-making in RTS environments. This collaborative multi-agent framework facilitates self-evolution by adaptively calibrating model bias through in-episode assessment and cross-episode analysis. We further incorporate dynamic observation pruning based on structural entropy to model game states topologically. By distilling high dimensional data into core semantic information, this approach significantly reduces inference time. We also develop a hybrid knowledge-memory mechanism that integrates micro-trajectories, macro-experience, and hierarchical domain knowledge, thereby enhancing both strategic adaptability and decision consistency. Experiments across multiple StarCraft II maps demonstrate that SEMA achieves superior win rates while reducing average decision latency by over 50%, validating its efficiency and robustness in complex RTS scenarios. △ Less

Submitted 24 March, 2026; originally announced March 2026.

Comments: 17 pages, 6 figures. Submitted to SCIS (Science China Information Science)

Reconstruction Matters: Learning Geometry-Aligned BEV Representation through 3D Gaussian Splatting

Authors: Yiren Lu, Xin Ye, Burhaneddin Yaman, Jingru Luo, Zhexiao Xiong, Liu Ren, Yu Yin

Abstract: Bird's-Eye-View (BEV) perception serves as a cornerstone for autonomous driving, offering a unified spatial representation that fuses surrounding-view images to enable reasoning for various downstream tasks, such as semantic segmentation, 3D object detection, and motion prediction. However, most existing BEV perception frameworks adopt an end-to-end training paradigm, where image features are dire… ▽ More Bird's-Eye-View (BEV) perception serves as a cornerstone for autonomous driving, offering a unified spatial representation that fuses surrounding-view images to enable reasoning for various downstream tasks, such as semantic segmentation, 3D object detection, and motion prediction. However, most existing BEV perception frameworks adopt an end-to-end training paradigm, where image features are directly transformed into the BEV space and optimized solely through downstream task supervision. This formulation treats the entire perception process as a black box, often lacking explicit 3D geometric understanding and interpretability, leading to suboptimal performance. In this paper, we claim that an explicit 3D representation matters for accurate BEV perception, and we propose Splat2BEV, a Gaussian Splatting-assisted framework for BEV tasks. Splat2BEV aims to learn BEV feature representations that are both semantically rich and geometrically precise. We first pre-train a Gaussian generator that explicitly reconstructs 3D scenes from multi-view inputs, enabling the generation of geometry-aligned feature representations. These representations are then projected into the BEV space to serve as inputs for downstream tasks. Extensive experiments on nuScenes and argoverse dataset demonstrate that Splat2BEV achieves state-of-the-art performance and validate the effectiveness of incorporating explicit 3D reconstruction into BEV perception. △ Less

Submitted 19 March, 2026; originally announced March 2026.

Comments: Project page at https://vulab-ai.github.io/Splat2BEV/