arXiv subjects

Jin Qin

Publications and source records attributed to Jin Qin.

At least 19 recordsLinked to original sources

arXiv2026

Phase-resolved QPO Analysis of GX 339-4: Improved Technique and Consistent Behaviors between QPOs and Broadband Noise

The nature of low-frequency quasi-periodic oscillations (QPOs) in black hole X-ray binaries remains unclear, and their relationship with the accompanying broadband noise (BBN) is still under debate. Here, we propose an improved variational mode decomposition (VMD) technique. Compared with the original algorithm that requires iterative, case-by-case parameter tuning, the new algorithm automatically and consistently determines the relevant VMD parameters based on the QPO central frequency and width measured from the power spectral density (PSD). This enables a more robust phase determination for QPOs and can also be applied to the study of BBN. We found that, for low-frequency type-C QPOs without significant harmonics in the black hole X-ray binary GX 339-4, the spectral properties of the QPOs and BBN are statistically consistent with each other: (1) the photon index is positively correlated with count rate as a function of phase, and (2) the PSD ratio spectra across different energy bands show no statistically significant QPO-like structures near the QPO frequencies, indicating that both components share the same energy dependence. These suggest that QPOs and BBN may be driven by the same physical processes. The QPO models based on geometric modulation struggle to account for the results, while those invoking corona oscillations are favored.

astro-ph.HE↗

arXiv2026

StrataCL: Fabric-Native Communication Library for Production Supernodes

Modern distributed AI workloads run across hundreds of accelerators, making communication a major bottleneck. Existing communication libraries remain largely buffer-centric because user and communication buffers are managed separately, causing redundant data copies or costly user-buffer registration. This paper presents StrataCL, a zero-redundancy and fabric-native communication library for production supernodes. StrataCL introduces registration-on-allocation to realize user-buffer direct communication, and designs communication operators with workload-balanced NPU-core partitioning and NPU-driven SDMA offloading to exploit supernode architecture features. On the Huawei CloudMatrix384, StrataCL improves collective bus bandwidth by up to 1.6x and improves MoE dispatch/combine bus bandwidth by up to 1.4x. Across three production workloads, StrataCL improves LLM inference throughput by 1.9x, reduces P99 TTFT by 2.2x, and reduces LLM and Recsys training iteration time by 1.4x and 1.3x, respectively.

cs.DC↗

arXiv2026

Beyond Retrieval: A Multitask Benchmark and Model for Code Search

Code search has usually been evaluated as first-stage retrieval, even though production systems rely on broader pipelines with reranking and developer-style queries. Existing benchmarks also suffer from data contamination, label noise, and degenerate binary relevance. In this paper, we introduce \textsc{CoREB}, a contamination-limited, multitask \underline{co}de \underline{r}etrieval and r\underline{e}ranking \underline{b}enchmark, together with a fine-tuned code reranker, that goes beyond retrieval to cover the full code search pipeline. \textsc{CoREB} is built from counterfactually rewritten LiveCodeBench problems in five programming languages and delivered as timed releases with graded relevance judgments. We benchmark eleven embedding models and five rerankers across three tasks: text-to-code, code-to-text, and code-to-code. Our experiments reveal that: \circone code-specialised embeddings dominate code-to-code retrieval (${\sim}2{\times}$ over general encoders), yet no single model wins all three tasks; \circtwo short keyword queries, the format closest to real developer search, collapse every model to near-zero nDCG@10; \circthree off-the-shelf rerankers are task-asymmetric, with a 12-point swing on code-to-code and no baseline net-positive across all tasks; \circfour our fine-tuned \textsc{CoREB-Reranker} is the first to achieve consistent gains across all three tasks. The data and model are released.

cs.SE↗

arXiv2026

Tessera: Unlocking Heterogeneous GPUs through Kernel-Granularity Disaggregation

Disaggregation maps parts of an AI workload to different types of GPUs, offering a path to utilize modern heterogeneous GPU clusters. However, existing solutions operate at a coarse granularity and are tightly coupled to specific model architectures, leaving much room for performance improvement. This paper presents Tessera, the first kernel disaggregation system to improve performance and cost efficiency on heterogeneous GPUs for large model inference. Our key insight is that kernels within a single application exhibit diverse resource demands, making them the most suitable granularity for aligning computation with hardware capabilities. Tessera integrates offline analysis with online adaptation by extracting precise inter-kernel dependencies from PTX to ensure correctness, overlapping communication with computation through a pipelined execution model, and employing workload-aware scheduling with lightweight runtime adaptation. Extensive evaluations across five heterogeneous GPUs and four model architectures, scaling up to 16 GPUs, show that Tessera improves serving throughput and cost efficiency by up to 2.3x and 1.6x, respectively, compared to existing disaggregation methods, while generalizing to model architectures where prior approaches do not apply. Surprisingly, a heterogeneous GPU pair under Tessera can even exceed the throughput of two homogeneous high-end GPUs at a lower cost.

cs.DC↗

arXiv2026

Exploring the $S_8$ Tension: Insights from the CatNorth 1.5-Million Quasar Candidates

The parameter $S_8$, a key probe of cosmic structure growth, exhibits a persistent $\sim3\sigma$ tension between high-redshift measurements from cosmic microwave background (CMB) anisotropies and low-redshift weak gravitational lensing observations. This discrepancy may indicate either unaccounted systematic effects or new physics beyond the standard $\Lambda$CDM cosmology. In this work, we constrain $S_8$ using the high purity CatNorth 1.5 million quasar candidate catalog and the {\it Planck} DR4 CMB lensing data across the broad redshift ranges through auto-correlation and cross-correlation analyses. To address the spatial incompleteness, we develop a machine-learning-based selection function that effectively suppresses the systematics-induced power spectrum excess on large scales. Our robust low-redshift measurements at $z<1.5$ yield $S_8 = 0.844^{+0.058}_{-0.056}$, consistent with the {\it Planck} 2018 CMB anisotropies constraints of $S_8=0.834\pm0.016$ but lower than the $0.879^{+0.055}_{-0.055}$ reported by a previous work using the Quaia quasar candidate catalog. However, for high-redshift faint quasars at $z>1.5$, we find a lower value of $S_8=0.724^{+0.058}_{-0.054}$, likely due to the sample incompleteness and/or the foreground contamination. Further tests on the volume-limited samples exhibit a consistent trend: $S_8 = 0.835^{+0.053}_{-0.049}$ for $z < 2$, $0.824^{+0.061}_{-0.062}$ for $0.4 < z < 1.5$, and a lower value of $0.789^{+0.062}_{-0.062}$ for the higher redshift range of $1.5 < z < 2.5$. While future data may refine these results, our current measurements based on a large sample of quasar candidates show less evidence of the $S_8$ tension.

astro-ph.CO↗

arXiv2026

Hummingbird: SLO-Oriented GPU Preemption at Microsecond-scale

Existing GPU-sharing techniques, including spatial and temporal sharing, aim to improve utilization but face challenges in simultaneously ensuring SLO adherence and maximizing efficiency due to the lack of fine-grained task scheduling on closed-source GPUs. This paper presents Hummingbird, an SLO-oriented GPU scheduling system that overcomes these challenges by enabling microsecond-scale preemption on closed-source GPUs while effectively harvesting idle GPU time slices. Comprehensive evaluations across diverse GPU architectures reveal that Hummingbird improves the SLO attainment of high-priority tasks by 9.7x and 3.5x compared to the state-of-the-art spatial and temporal-sharing approaches. When compared to executing exclusively, the SLO attainment of the high-priority task, collocating with low-priority tasks on Hummingbird, only drops by less than 1%. Meanwhile, the throughput of the low-priority task outperforms the state-of-the-art temporal-sharing approaches by 2.4x. Hummingbird demonstrates significant effectiveness in ensuring the SLO while enhancing GPU utilization.

cs.DC↗

arXiv2025

C2LLM Technical Report: A New Frontier in Code Retrieval via Adaptive Cross-Attention Pooling

We present C2LLM - Contrastive Code Large Language Models, a family of code embedding models in both 0.5B and 7B sizes. Building upon Qwen-2.5-Coder backbones, C2LLM adopts a Pooling by Multihead Attention (PMA) module for generating sequence embedding from token embeddings, effectively 1) utilizing the LLM's causal representations acquired during pretraining, while also 2) being able to aggregate information from all tokens in the sequence, breaking the information bottleneck in EOS-based sequence embeddings, and 3) supporting flexible adaptation of embedding dimension, serving as an alternative to MRL. Trained on three million publicly available data, C2LLM models set new records on MTEB-Code among models of similar sizes, with C2LLM-7B ranking 1st on the overall leaderboard.

cs.CL↗

arXiv2025

On chip plasmonic slit cavity platform for room temperature strong coupling with deterministically positioned colloidal quantum dots

Strong coupling between quantum emitters and optical cavities is essential for quantum information processing, high-purity single-photon sources, and nonlinear quantum devices. Achieving this regime at room temperature in a compact, deterministic on-chip platform-critical for integration with nanoelectronic circuitry and scalable device architectures-remains a major challenge, mainly due to the difficulty of fabricating cavities with ultra-small mode volumes and precisely positioning quantum emitters. Here, we demonstrate a robust quantum plasmonic device in which colloidal quantum dots (Qdots) are strongly coupled to plasmonic slit cavities using a dielectrophoresis-based positioning technique with real-time photoluminescence (PL) feedback, providing directly resolvable coupled structures that enable parallel device fabrication and straightforward integration with additional optical elements such as waveguides. Our measurements reveal clear PL resolved Rabi splitting at room temperature with pre characterized cavities, with variations across devices that scale with the average number of coupled Qdots. While electrical tuning via the quantum-confined Stark effect is enabled by integrated electrodes, its impact is largely overshadowed by room-temperature spectral diffusion. Our results pave the way for scalable, electrically tunable quantum plasmonic platforms, offering new opportunities for integrated quantum photonic circuits, active light-matter interactions, and room-temperature quantum technologies.

physics.optics↗

arXiv2025

Intrinsic Maximum Light Absorption in Laser-Field-Driven Growth of Highly Ordered Silicon Nanowire Arrays

We provide direct experimental evidence for a state-selection principle in a far-from-equilibrium system. Using the laser-driven growth of silicon nanowires as a uniquely clean and quantifiable platform, we show that a long-range ordered array emerges as the system spontaneously selects the periodicity that maximizes its collective light absorption. This establishes a direct, measurable link between a maximum dissipation/absorption principle and emergent structural order. Our results thus offer a concrete test for models of non-equilibrium self-organization.

cond-mat.mes-hall↗

arXiv2025

QiMeng-MuPa: Mutual-Supervised Learning for Sequential-to-Parallel Code Translation

The rise of GPU-based high-performance computing (HPC) has driven the widespread adoption of parallel programming models such as CUDA. Yet, the inherent complexity of parallel programming creates a demand for the automated sequential-to-parallel approaches. However, data scarcity poses a significant challenge for machine learning-based sequential-to-parallel code translation. Although recent back-translation methods show promise, they still fail to ensure functional equivalence in the translated code. In this paper, we propose \textbf{QiMeng-MuPa}, a novel \textbf{Mu}tual-Supervised Learning framework for Sequential-to-\textbf{Pa}rallel code translation, to address the functional equivalence issue. QiMeng-MuPa consists of two models, a Translator and a Tester. Through an iterative loop consisting of Co-verify and Co-evolve steps, the Translator and the Tester mutually generate data for each other and improve collectively. The Tester generates unit tests to verify and filter functionally equivalent translated code, thereby evolving the Translator, while the Translator generates translated code as augmented input to evolve the Tester. Experimental results demonstrate that QiMeng-MuPa significantly enhances the performance of the base models: when applied to Qwen2.5-Coder, it not only improves Pass@1 by up to 28.91% and boosts Tester performance by 68.90%, but also outperforms the previous state-of-the-art method CodeRosetta by 1.56 and 6.92 in BLEU and CodeBLEU scores, while achieving performance comparable to DeepSeek-R1 and GPT-4.1. Our code is available at https://github.com/kcxain/mupa.

cs.SE↗

arXiv2025

Large Processor Chip Model

Computer System Architecture serves as a crucial bridge between software applications and the underlying hardware, encompassing components like compilers, CPUs, coprocessors, and RTL designs. Its development, from early mainframes to modern domain-specific architectures, has been driven by rising computational demands and advancements in semiconductor technology. However, traditional paradigms in computer system architecture design are confronting significant challenges, including a reliance on manual expertise, fragmented optimization across software and hardware layers, and high costs associated with exploring expansive design spaces. While automated methods leveraging optimization algorithms and machine learning have improved efficiency, they remain constrained by a single-stage focus, limited data availability, and a lack of comprehensive human domain knowledge. The emergence of large language models offers transformative opportunities for the design of computer system architecture. By leveraging the capabilities of LLMs in areas such as code generation, data analysis, and performance modeling, the traditional manual design process can be transitioned to a machine-based automated design approach. To harness this potential, we present the Large Processor Chip Model (LPCM), an LLM-driven framework aimed at achieving end-to-end automated computer architecture design. The LPCM is structured into three levels: Human-Centric; Agent-Orchestrated; and Model-Governed. This paper utilizes 3D Gaussian Splatting as a representative workload and employs the concept of software-hardware collaborative design to examine the implementation of the LPCM at Level 1, demonstrating the effectiveness of the proposed approach. Furthermore, this paper provides an in-depth discussion on the pathway to implementing Level 2 and Level 3 of the LPCM, along with an analysis of the existing challenges.

cs.AR↗

arXiv2025

Plasmonic Su-Schrieffer-Heeger chains with strong coupling amplitudes

Plasmonic many-particle systems with precisely tuned resonances and coupling strengths can exhibit emergent collective properties governed by universal principles. In one-dimensional chains with alternating couplings, known as Su-Schrieffer-Heeger (SSH) systems, this includes the formation of topologically protected mid-gap modes whose intensities localize at the chain's ends. This subwavelength localization at optical frequencies is crucial for achieving strong coupling of mid-gap modes to two-level systems under ambient conditions, extending topological protection to hybrid light-matter states. Here, we have fabricated SSH chains from plasmonic nanoslit resonators with strong inter-resonator coupling. The alternating distance between the nanoslit resonators is controlled with sub-nanometer precision, enabling accurate prediction and experimental observation of topologically protected mid-gap modes via photoemission electron microscopy (PEEM). Our results open the path towards experimental realizations of two-dimensional photonic metasurfaces exhibiting higher-order topological modes that can be strongly coupled to single emitters and quantum materials at ambient conditions.

physics.optics↗

arXiv2025

The CatSouth Quasar Candidate Catalog for the Southern Sky and a Unified All-Sky Catalog Based on Gaia DR3

The Gaia DR3 has provided a large sample of more than 6.6 million quasar candidates with high completeness but low purity. Previous work on the CatNorth quasar candidate catalog has shown that including external multiband data and applying machine-learning methods can efficiently purify the original Gaia DR3 quasar candidate catalog and improve the redshift estimates. In this paper, we extend the Gaia DR3 quasar candidate selection to the southern hemisphere using data from SkyMappper, CatWISE, and VISTA surveys. We train an XGBoost classifier on a unified set of high-confidence stars and spectroscopically confirmed quasars and galaxies. For sources with available Gaia BP/RP spectra, spectroscopic redshifts are derived using a pre-trained convolutional neural network (RegNet). We also train an ensemble photometric redshift estimation model based on XGBoost, TabNet, and FT-Transformer, achieving an RMSE of 0.2256 and a normalized median absolute deviation of 0.0187 on the validation set. By merging CatSouth with the previously published CatNorth catalog, we construct the unified all-sky CatGlobe catalog with nearly 1.9 million sources at $G<21$, providing a comprehensive and high-purity quasar candidate sample for future spectroscopic and cosmological investigations.

astro-ph.GA↗

arXiv2025

Constraints on Baryon Density from the Effective Optical Depth of High-Redshift Quasars

We present constraints on the baryonic matter density parameter, $Ω_b$, within the framework of the $Λ$CDM model. Our analysis utilizes observational data on the effective optical depth from high-redshift quasars. To parameterize the photoionization rate $Γ_{-12}$, we employ a Bézier polynomial. Additionally, we approximate the Hubble parameter at high redshifts as $H(z)\approx 100hΩ_m^{1/2} (1+z)^{3/2}$ km s$^{-1}$ Mpc$^{-1}$. Confidence regions are obtained with $h=0.701\pm0.013$ and $Ω_m = 0.315$, optimized by the Planck mission. The best-fit values are $Ω_b =0.043^{+0.005}_{-0.006}$ and $Ω_b = 0.045^{+0.004}_{-0.006}$, corresponding to an old data set and a new data set, respectively. And we test the non-parametric form of $Γ_{-12}$, obtaining $Ω_b = 0.048^{+0.001}_{-0.003}$. These results are consistent with the findings of Planck at the 1 $σ$ confidence level. Our findings underscore the effectiveness of quasar datasets in constraining $Ω_b$, eliminating the need for independent photoionization rate data. This approach provides detailed cosmic information about baryon density and the photoionization history of the intergalactic medium.

astro-ph.CO↗

arXiv2025

A new algorithm for detecting X-ray shots in Cyg X-1

The short-term X-ray variability of Cyg X-1 can be interpreted as random occurrence of mini-flares known as the shots, whose physical nature is still unclear. We propose a new algorithm for shot identification in the X-ray light curve, based on baseline detection and template fitting. Compared with previous techniques, our algorithm allows us to detect shots with lower amplitudes and shorter time separations. With NICER observations, we find that, after correction for detection sensitivity, both the shot amplitude and recurrence rate are positively scaled with the mean count rate, while the recurrence rate has a much higher dependence on the count rate. These suggest that a higher mass accretion rate will drive more and slightly larger shots. We also find that the abrupt hardening near the shot peak found in previous studies is attributed to different shot profiles in different energy bands; there is no need to involve a rapid physical process to suddenly harden the emitting spectrum.

astro-ph.HE↗

arXiv2023

KV Inversion: KV Embeddings Learning for Text-Conditioned Real Image Action Editing

Text-conditioned image editing is a recently emerged and highly practical task, and its potential is immeasurable. However, most of the concurrent methods are unable to perform action editing, i.e. they can not produce results that conform to the action semantics of the editing prompt and preserve the content of the original image. To solve the problem of action editing, we propose KV Inversion, a method that can achieve satisfactory reconstruction performance and action editing, which can solve two major problems: 1) the edited result can match the corresponding action, and 2) the edited object can retain the texture and identity of the original real image. In addition, our method does not require training the Stable Diffusion model itself, nor does it require scanning a large-scale dataset to perform time-consuming training.

cs.CV↗

arXiv2023

Anticrossing of a plasmonic nanoresonator mode and a single quantum dot at room temperature

Room-temperature strong coupling of a single quantum emitter and a single resonant plasmonic mode is a key resource for quantum information processing and quantum sensing at ambient conditions. To beat dephasing, ultrafast energy transfer is achieved by coupling single emitters to a plasmonic nanoresonator with an extremely small mode volume and optimal spectral overlap. Typically, normal mode splittings in luminescence spectra of single-emitter strongly-coupled systems are provided as evidence for strong coupling and to obtain rough estimates of the light-matter coupling strength g. However, a complete anticrossing of a single emitter and a cavity mode as well as the characterization of the uncoupled constituents is usually hard to achieve. Here, we exploit the light-induced oxygen-dependent blue-shift of individual CdSe/ZnS semiconductor quantum dots to tune their transition energy across the resonance of a scanning plasmonic slit resonator after characterizing both single emitter and nano resonator in their uncoupled states. Our results provide clear proof of single-emitter strong light-matter coupling at ambient condition as well as a value for the Rabi splitting at zero detuning 100 meV, consistent with modeling, thereby opening the path towards plexitonic devices that exploit single-photon nonlinearities at ambient conditions.

physics.optics↗

arXiv2023

Characterization of loss mechanisms in a fluxonium qubit

Using a fluxonium qubit with in situ tunability of its Josephson energy, we characterize its energy relaxation at different flux biases as well as different Josephson energy values. The relaxation rate at qubit energy values, ranging more than one order of magnitude around the thermal energy $k_B T$, can be quantitatively explained by a combination of dielectric loss and $1/f$ flux noise with a crossover point. The amplitude of the $1/f$ flux noise is consistent with that extracted from the qubit dephasing measurements at the flux sensitive points. In the dielectric loss dominant regime, the loss is consistent with that arises from the electric dipole interaction with two-level-system (TLS) defects. In particular, as increasing Josephson energy thus decreasing qubit frequency at the flux insensitive spot, we find that the qubit exhibits increasingly weaker coupling to TLS defects thus desirable for high-fidelity quantum operations.

quant-ph↗