arXiv subjects

Hao Dai

Publications and source records attributed to Hao Dai.

At least 19 recordsLinked to original sources

arXiv2026

Uncovering Non-Gaussianity through Multi-Copy Symmetries

Gaussian states are fundamental in continuous-variable quantum information, yet characterizing non-Gaussianity remains challenging due to the non-convexity of the Gaussian set. Existing witnesses typically rely on Wigner negativity or other information-theoretic quantities. In this work, we develop a group-theoretic, multi-copy approach to detect non-Gaussianity in bosonic systems. We study passive linear optical transformations that mix copies of a quantum state and analyze their commutation with identical Gaussian unitaries applied to each copy. Orthogonal copy-mixing transformations commute with the symplectic part of the Gaussian action, while the displacement part restricts the symmetry to the stabilizer of the collective mode. This structure yields a family of witnesses satisfied by all single-mode Gaussian states. Fixing the thermal reference parameter via the purity, violation of these identities certifies non-Gaussianity. We illustrate the method with several single-mode examples and present an experimental protocol based on passive interferometry and photon-number-resolved detection, showing that the relevant multi-copy expectation values can be estimated from bounded phase observables. Finally, we extend the construction to multi-mode systems and discuss how the same symmetry framework may lead to quantitative measures of non-Gaussianity.

quant-ph↗

arXiv2026

DualDecoder: Accelerate Long Context LLM Inference by Predictive Prefetch

Long-context inference is becoming a fundamental capability for modern LLM serving, especially driven by emerging agentic applications. Yet it faces a severe memory wall that the KV cache scales proportionally with increasing context length and request concurrency. Existing sparse KV cache methods offload most KV entries to host memory and retrieve only the critical KV entries needed by each decoding step. However, they commonly introduce substantial auxiliary states in GPU memory for KV retrieval management. Our measurements show that these often-overlooked auxiliary states introduce significant memory overhead and become a new bottleneck under high-concurrency workloads. In this paper, we present DualDecoder, a lightweight serving system for long-context LLM inference that enables efficient sparse KV cache retrieval from host memory. Our key insight is that the critical KV entries required for decoding the next token can be accurately predicted from the preceding speculated token. This predictability enables KV retrieval to be proactively prefetched and overlapped with decoding computation, effectively eliminating the GPU memory overhead of auxiliary states. To achieve this prefetching efficiently, DualDecoder leverages a novel dual-token decoding pipeline that accurately identifies critical KV entries with negligible computational overhead, and designs a layer-aware transfer schedule to overlap KV prefetching with model computation and a layer-scoped memory manager to reduce the GPU runtime buffer. Experimental results show that DualDecoder improves decoding throughput by up to 2.62$\times$ over state-of-the-art systems while preserving decoding latency and model quality.

cs.DC↗

arXiv2026

Fracture Risk Prediction in Adults Over 50 Years Old Using DXA and EHR: Comparison of Traditional and Machine Learning Models in Two Large Cohorts

Accurate fracture risk prediction is important for osteoporosis management, but commonly used clinical tools may not fully use information available in electronic health records (EHRs) and dual-energy X-ray absorptiometry (DXA) reports. We developed and externally validated time-to-event fracture prediction models among adults aged 50 years or older with clinically obtained DXA reports in 2 US health care systems. The development cohort was derived from NewYork-Presbyterian/Weill Cornell Medical Center and the external validation cohort from the Indiana Network for Patient Care. Predictors included demographics, lifestyle factors, prior fracture, comorbidities, medication exposures, osteoporosis treatment history, and DXA-derived T-scores extracted from radiology reports. The outcome was time from index DXA to first incident fragility fracture identified from structured diagnosis codes. We evaluated penalized Cox regression, random survival forest, gradient-boosting survival, and XGBoost survival models using 2 prespecified predictor settings and compared discrimination with clinically reported FRAX major osteoporotic fracture probabilities. The development cohort included 11,510 adults, of whom 858 sustained incident fragility fractures; the external validation cohort included 1,932 adults, of whom 180 sustained fractures. In internal validation, the expanded Cox model achieved a mean Harrell C-index of 0.779, compared with 0.653 for FRAX. In external validation, the corresponding Cox model achieved a Harrell C-index of 0.714, compared with 0.590 for FRAX; gradient-boosting survival had the highest external discrimination (0.725). EHR- and DXA-enhanced models showed better discrimination than clinically reported FRAX scores in this DXA-tested population, but calibration assessment, prospective evaluation, and implementation workflow assessment are needed before clinical use.

stat.AP↗

arXiv2026

Next-Generation Agentic Reinforcement Learning Systems Enable Self-Evolving Agents

LLM agents are rapidly being deployed in production, including coding assistants, customer-support chatbots, and scientific research assistants, yet they remain fundamentally static in enterprise deployment. The LLM weights, system prompts, tool repertoires, and in-context harnesses are frozen at deployment time, and any improvement requires a manual loop of human-curated data collection, offline fine-tuning, modification of the agentic paradigm, and re-deployment. Recent work on self-evolving agents, such as OpenClaw for individual users, indicates that the next leap in agent capability will come from agents that continually learn from their own experience. In this paper, we argue that this vision for self-evolving agent deployment is being held back for enterprise-level large-scale agentic service not by reinforcement learning (RL) algorithms but by agentic online RL systems. Specifically, current agentic RL systems and the surrounding observability software stack are inadequate along three essential aspects: (i) there is no standardized agent trajectory data protocol capable of carrying RL learning signals at step granularity across heterogeneous agent paradigms; (ii) there is no enterprise-grade comprehensive data proxy that converts real workloads into governed learning substrates; and (iii) there is no unified agent evolution control plane that automatically decides, based on trajectory statistics, when to update policy weights or evolve the in-context harness. The next generation of agentic RL systems must be co-designed around these three pillars, and we sketch concrete architectures, case studies, and counter-arguments. We instantiate one branch through AReaL2.0, reorganizing existing RL infrastructure into an agent-oriented online RL loop for policy weight updates from deployed workloads.

cs.DC↗

arXiv2026

Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale

Efficient and scalable agentic intelligence requires models that can deliver both low-latency responses and strong reasoning capabilities while remaining practical to train, serve, and deploy. In this report, we present Ling-2.6 and Ring-2.6, a family of models designed to address this challenge at scale. Ling-2.6 is optimized for instant response generation and high capability per output token, whereas Ring-2.6 is tailored for deeper reasoning and more advanced agentic workflows. Instead of training from scratch, we upgrade the Ling-2.0 base model through architectural migration pre-training and large-scale post-training. This upgrade is guided by a unified co-design of model architecture, optimization objectives, serving systems, and agent training environments, enabling improvements in both model capability and deployment efficiency. At the architectural level, we introduce a hybrid linear attention design that integrates Lightning Attention with MLA, improving the efficiency of long-context training and decoding. To further enhance token efficiency, we optimize capability per output token through Evolutionary Chain-of-Thought, Linguistic Unit Policy Optimization, bidirectional preference alignment, and shortest-correct-response distillation. For agentic capabilities, we propose KPop, a reinforcement learning framework designed to support stable training of Ring-2.6-1T on large-scale environment-grounded data. KPop improves training efficiency through asynchronous scheduling across coding, search, tool use, and workflow execution, enabling scalable learning from complex agent-environment interactions. Together, Ling-2.6 and Ring-2.6 provide a practical pathway toward efficient, scalable, and open agentic systems. We open-source all checkpoints in the 2.6 family to support further research and development in practical agentic intelligence.

cs.CL↗

arXiv2026

Towards Bulk Locality: A Systematic Construction of Contact Interactions from Chord Diagrams

Chord diagrams encode boundary correlators in the double-scaled holographic Sachdev-Ye-Kitaev model, but currently capture only a limited class of bulk interactions that yield pure power-law correlators. In this article, we investigate a general construction based on Fock-space flux models with arbitrary periodic lattice size, clarifies how lattice dimensions control probe configurations and bulk contact vertices. Developing a systematic matching scheme and using the chord path integral formalism, we compute three- to six-point contact correlators and reproduce a broad class of AdS$_2$ scalar contact Witten diagrams, including those with logarithmic singularities. The results demonstrate that chord diagrams, in full generality, provide a microscopic description of bulk contact interactions and thereby establish a principled framework for reconstructing bulk locality from boundary data.

hep-th↗

arXiv2026

Domain-Specific Knowledge Graphs in RAG-Enhanced Healthcare LLMs

Large Language Models (LLMs) generate fluent answers but can struggle with trustworthy, domain-specific reasoning. We evaluate whether domain knowledge graphs (KGs) improve Retrieval-Augmented Generation (RAG) for healthcare by constructing three PubMed-derived graphs: $\mathbb{G}_1$ (T2DM), $\mathbb{G}_2$ (Alzheimer's disease), and $\mathbb{G}_3$ (AD+T2DM). We design two probes: Probe 1 targets merged AD T2DM knowledge, while Probe 2 targets the intersection of $\mathbb{G}_1$ and $\mathbb{G}_2$. Seven instruction-tuned LLMs are tested across retrieval sources {No-RAG, $\mathbb{G}_1$, $\mathbb{G}_2$, $\mathbb{G}_1$ + $\mathbb{G}_2$, $\mathbb{G}_3$, $\mathbb{G}_1$+$\mathbb{G}_2$ + $\mathbb{G}_3$} and three decoding temperatures. Results show that scope alignment between probe and KG is decisive: precise, scope-matched retrieval (notably $\mathbb{G}_2$) yields the most consistent gains, whereas indiscriminate graph unions often introduce distractors that reduce accuracy. Larger models frequently match or exceed KG-RAG with a No-RAG baseline on Probe 1, indicating strong parametric priors, whereas smaller/mid-sized models benefit more from well-scoped retrieval. Temperature plays a secondary role; higher values rarely help. We conclude that precision-first, scope-matched KG-RAG is preferable to breadth-first unions, and we outline practical guidelines for graph selection, model sizing, and retrieval/reranking. Code and Data available here - https://github.com/sydneyanuyah/RAGComparison

cs.CL↗

arXiv2025

Spin minimum uncertainty states for refined uncertainty relations

Minimum uncertainty states of the conventional Heisenberg uncertainty relation have been extensively studied and are often regarded as the most classical quantum states from the perspective of uncertainty, providing valuable insight into the nature of quantumness and its potential applications. In this work, we investigate the minimum uncertainty states associated with an information-theoretic refinement of the Heisenberg uncertainty relation in general spin systems. Using two different approaches, the matrix formulation and the Wick symbol representation, we derive explicit expressions for the states that saturate the uncertainty bound. We show that spin coherent states indeed achieve minimum uncertainty, consistent with their conventional identification as the classical states of spin systems. Moreover, we also identify additional classes of minimum uncertainty states beyond the coherent family. Finally, we compare the spin-system results with the previously studied bosonic case and elucidate the origin of the differences between the two settings.

quant-ph↗

arXiv2025

Parallax: Runtime Parallelization for Operator Fallbacks in Heterogeneous Edge Systems

The growing demand for real-time DNN applications on edge devices necessitates faster inference of increasingly complex models. Although many devices include specialized accelerators (e.g., mobile GPUs), dynamic control-flow operators and unsupported kernels often fall back to CPU execution. Existing frameworks handle these fallbacks poorly, leaving CPU cores idle and causing high latency and memory spikes. We introduce Parallax, a framework that accelerates mobile DNN inference without model refactoring or custom operator implementations. Parallax first partitions the computation DAG to expose parallelism, then employs branch-aware memory management with dedicated arenas and buffer reuse to reduce runtime footprint. An adaptive scheduler executes branches according to device memory constraints, meanwhile, fine-grained subgraph control enables heterogeneous inference of dynamic models. By evaluating on five representative DNNs across three different mobile devices, Parallax achieves up to 46% latency reduction, maintains controlled memory overhead (26.5% on average), and delivers up to 30% energy savings compared with state-of-the-art frameworks, offering improvements aligned with the responsiveness demands of real-time mobile inference.

cs.DC↗

arXiv2025

Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale Thinking Model

We present Ring-1T, the first open-source, state-of-the-art thinking model with a trillion-scale parameter. It features 1 trillion total parameters and activates approximately 50 billion per token. Training such models at a trillion-parameter scale introduces unprecedented challenges, including train-inference misalignment, inefficiencies in rollout processing, and bottlenecks in the RL system. To address these, we pioneer three interconnected innovations: (1) IcePop stabilizes RL training via token-level discrepancy masking and clipping, resolving instability from training-inference mismatches; (2) C3PO++ improves resource utilization for long rollouts under a token budget by dynamically partitioning them, thereby obtaining high time efficiency; and (3) ASystem, a high-performance RL framework designed to overcome the systemic bottlenecks that impede trillion-parameter model training. Ring-1T delivers breakthrough results across critical benchmarks: 93.4 on AIME-2025, 86.72 on HMMT-2025, 2088 on CodeForces, and 55.94 on ARC-AGI-1. Notably, it attains a silver medal-level result on the IMO-2025, underscoring its exceptional reasoning capabilities. By releasing the complete 1T parameter MoE model to the community, we provide the research community with direct access to cutting-edge reasoning capabilities. This contribution marks a significant milestone in democratizing large-scale reasoning intelligence and establishes a new baseline for open-source model performance.

cs.CL↗

arXiv2025

ViRN: Variational Inference and Distribution Trilateration for Long-Tailed Continual Representation Learning

Continual learning (CL) with long-tailed data distributions remains a critical challenge for real-world AI systems, where models must sequentially adapt to new classes while retaining knowledge of old ones, despite severe class imbalance. Existing methods struggle to balance stability and plasticity, often collapsing under extreme sample scarcity. To address this, we propose ViRN, a novel CL framework that integrates variational inference (VI) with distributional trilateration for robust long-tailed learning. First, we model class-conditional distributions via a Variational Autoencoder to mitigate bias toward head classes. Second, we reconstruct tail-class distributions via Wasserstein distance-based neighborhood retrieval and geometric fusion, enabling sample-efficient alignment of tail-class representations. Evaluated on six long-tailed classification benchmarks, including speech (e.g., rare acoustic events, accents) and image tasks, ViRN achieves a 10.24% average accuracy gain over state-of-the-art methods.

cs.LG↗

arXiv2025

Continual Generalized Category Discovery: Learning and Forgetting from a Bayesian Perspective

Continual Generalized Category Discovery (C-GCD) faces a critical challenge: incrementally learning new classes from unlabeled data streams while preserving knowledge of old classes. Existing methods struggle with catastrophic forgetting, especially when unlabeled data mixes known and novel categories. We address this by analyzing C-GCD's forgetting dynamics through a Bayesian lens, revealing that covariance misalignment between old and new classes drives performance degradation. Building on this insight, we propose Variational Bayes C-GCD (VB-CGCD), a novel framework that integrates variational inference with covariance-aware nearest-class-mean classification. VB-CGCD adaptively aligns class distributions while suppressing pseudo-label noise via stochastic variational updates. Experiments show VB-CGCD surpasses prior art by +15.21% with the overall accuracy in the final session on standard benchmarks. We also introduce a new challenging benchmark with only 10% labeled data and extended online phases, VB-CGCD achieves a 67.86% final accuracy, significantly higher than state-of-the-art (38.55%), demonstrating its robust applicability across diverse scenarios. Code is available at: https://github.com/daihao42/VB-CGCD

cs.LG↗

arXiv2025

Ring-lite: Scalable Reasoning via C3PO-Stabilized Reinforcement Learning for LLMs

We present Ring-lite, a Mixture-of-Experts (MoE)-based large language model optimized via reinforcement learning (RL) to achieve efficient and robust reasoning capabilities. Built upon the publicly available Ling-lite model, a 16.8 billion parameter model with 2.75 billion activated parameters, our approach matches the performance of state-of-the-art (SOTA) small-scale reasoning models on challenging benchmarks (e.g., AIME, LiveCodeBench, GPQA-Diamond) while activating only one-third of the parameters required by comparable models. To accomplish this, we introduce a joint training pipeline integrating distillation with RL, revealing undocumented challenges in MoE RL training. First, we identify optimization instability during RL training, and we propose Constrained Contextual Computation Policy Optimization(C3PO), a novel approach that enhances training stability and improves computational throughput via algorithm-system co-design methodology. Second, we empirically demonstrate that selecting distillation checkpoints based on entropy loss for RL training, rather than validation metrics, yields superior performance-efficiency trade-offs in subsequent RL training. Finally, we develop a two-stage training paradigm to harmonize multi-domain data integration, addressing domain conflicts that arise in training with mixed dataset. We will release the model, dataset, and code.

cs.CL↗

arXiv2024

Measurement-device-independent resource characterization protocols

Measurement-device-independent (MDI) quantum information processing tasks are important subroutines in quantum information science because they are robust against any type of measurement imperfections. In this work, we propose a framework of MDI resource characterization protocols that unifies and generalizes these tasks. We show that resources that do not increase under local operation and shared randomness can be characterized with any untrusted measurement, and we provide a general procedure to convert such resource characterization tasks into MDI protocols. We then focus on applying our framework to two cases that satisfy the criteria: the resource theory of bipartite and multipartite entanglement, and the resource theory of quantum memories. We demonstrate several MDI characterization protocols for these resources. These protocols are either novel or generalize existing ones from the literature. We also show that MDI quantum key distribution can be viewed as an MDI quantification protocol for quantum memory.

quant-ph↗

arXiv2024

CAAP: Class-Dependent Automatic Data Augmentation Based On Adaptive Policies For Time Series

Data Augmentation is a common technique used to enhance the performance of deep learning models by expanding the training dataset. Automatic Data Augmentation (ADA) methods are getting popular because of their capacity to generate policies for various datasets. However, existing ADA methods primarily focused on overall performance improvement, neglecting the problem of class-dependent bias that leads to performance reduction in specific classes. This bias poses significant challenges when deploying models in real-world applications. Furthermore, ADA for time series remains an underexplored domain, highlighting the need for advancements in this field. In particular, applying ADA techniques to vital signals like an electrocardiogram (ECG) is a compelling example due to its potential in medical domains such as heart disease diagnostics. We propose a novel deep learning-based approach called Class-dependent Automatic Adaptive Policies (CAAP) framework to overcome the notable class-dependent bias problem while maintaining the overall improvement in time-series data augmentation. Specifically, we utilize the policy network to generate effective sample-wise policies with balanced difficulty through class and feature information extraction. Second, we design the augmentation probability regulation method to minimize class-dependent bias. Third, we introduce the information region concepts into the ADA framework to preserve essential regions in the sample. Through a series of experiments on real-world ECG datasets, we demonstrate that CAAP outperforms representative methods in achieving lower class-dependent bias combined with superior overall performance. These results highlight the reliability of CAAP as a promising ADA method for time series modeling that fits for the demands of real-world applications.

cs.LG↗

arXiv2024

Open Set Dandelion Network for IoT Intrusion Detection

As IoT devices become widely, it is crucial to protect them from malicious intrusions. However, the data scarcity of IoT limits the applicability of traditional intrusion detection methods, which are highly data-dependent. To address this, in this paper we propose the Open-Set Dandelion Network (OSDN) based on unsupervised heterogeneous domain adaptation in an open-set manner. The OSDN model performs intrusion knowledge transfer from the knowledge-rich source network intrusion domain to facilitate more accurate intrusion detection for the data-scarce target IoT intrusion domain. Under the open-set setting, it can also detect newly-emerged target domain intrusions that are not observed in the source domain. To achieve this, the OSDN model forms the source domain into a dandelion-like feature space in which each intrusion category is compactly grouped and different intrusion categories are separated, i.e., simultaneously emphasising inter-category separability and intra-category compactness. The dandelion-based target membership mechanism then forms the target dandelion. Then, the dandelion angular separation mechanism achieves better inter-category separability, and the dandelion embedding alignment mechanism further aligns both dandelions in a finer manner. To promote intra-category compactness, the discriminating sampled dandelion mechanism is used. Assisted by the intrusion classifier trained using both known and generated unknown intrusion knowledge, a semantic dandelion correction mechanism emphasises easily-confused categories and guides better inter-category separability. Holistically, these mechanisms form the OSDN model that effectively performs intrusion knowledge transfer to benefit IoT intrusion detection. Comprehensive experiments on several intrusion datasets verify the effectiveness of the OSDN model, outperforming three state-of-the-art baseline methods by 16.9%.

cs.LG↗

arXiv2023

Robust Estimation of Nonlinear Properties of Quantum Processes

Accurate and robust estimation of quantum process properties is crucial for quantum information processing and quantum many-body physics. Combining classical shadow tomography and randomized benchmarking, Helsen et al. introduced a method to estimate the linear properties of quantum processes. In this work, we focus on the estimation protocols of nonlinear process properties that are robust to state preparation and measurement errors. We introduce two protocols, both utilizing random gate sequences but employing different post-processing methods, which make them suitable for measuring different nonlinear properties. The first protocol offers a robust and sound method to estimate the out-of-time-ordered correlation, as demonstrated numerically in an Ising model. The second protocol estimates unitarity, effectively characterizing the incoherence of quantum channels. We expect the two protocols to be useful tools for exploring quantum many-body physics and characterizing quantum processes.

quant-ph↗

arXiv2023

Approximate quantum error correction, covariance symmetry, and their relation

To perform reliable quantum computation, quantum error correction is indispensable. In certain cases, continuous covariance symmetry of the physical system can make exact error correction impossible. In this work we study the approximate error correction and covariance symmetry from the information-theoretic perspective. For general encoding and noise channels, we define a quantity named infidelity to characterize the performance of the approximate quantum error correction and quantify the noncovariance of an encoding channel with respect to a general Lie group from the asymmetry measure of the corresponding Choi state. In particular, when the encoding channel is isometric, we derive a trade-off relation between infidelity and noncovariance. Furthermore, we calculate the average infidelity and noncovariance measure for a type of random code.

quant-ph↗