arXiv subjects

Jiahan Chen

Publications and source records attributed to Jiahan Chen.

10 recordsLinked to original sources

arXiv2026

Triage: An Adaptive Parallel Window Decoding Scheduler for Real-time Fault-Tolerant Quantum Computation

Fault-tolerant quantum computation (FTQC) critically depends on real-time classical decoding, which is rapidly emerging as a system bottleneck. As quantum systems scale, decoding latency and throughput limitations lead to exponential syndrome backlogs and logical operation stalls. While hardware accelerators and parallel windowing offer pathways to speed up decoding, dynamically deploying a finite pool of decoders across a vast quantum error correction architecture remains an unresolved resource allocation problem. To address this, we formulate FTQC decoding as a constrained dynamic scheduling problem by utilizing a spatio-temporal framework based on slices. We propose Triage, a dual-mode architecture that mitigates operation stalls by adaptively combining a cost-efficient heuristic scheduler with a priority-aware emergency mode to rapidly resolve the causal cone of critical operations. Our evaluation shows that Triage maintains low algorithm stalls and logical error rates even under scarce classical resource constraints. Across various benchmarks, Triage achieves an average logical error rate reduction of 52.6% compared to standard temporal parallelism, enabling an efficient classical control plane for scalable FTQC architectures.

quant-ph↗

arXiv2026

O3LS: Optimizing Lattice Surgery via Automatic Layout Searching and Loose Scheduling

Toward the large-scale, practical realization of quantum computing, quantum error correction is essential. Among various quantum error-correcting codes, the surface code stands out as a leading candidate, and lattice surgery based on surface codes has emerged as a promising technique for fault-tolerant quantum computation (FTQC). However, implementing quantum algorithms using lattice surgery introduces both resource and time overhead. Existing approaches typically focus on large layout designs, with compiler passes aimed primarily at optimizing time overhead. This often overlooks the trade-off between rotation bottlenecks and movement distance, which leads to inefficient resource utilization and prevents further reduction of the quantum computation failure rate. To address these challenges, we introduce O3LS, a framework for optimizing lattice surgery through automatic layout search and loose scheduling. O3LS achieves an optimal balance by automatically generating squeezed data layouts to reduce space requirements and employing loose scheduling algorithms combined with circuit synthesis techniques to reduce time overhead, thereby effectively minimizing overall logical error rates. Numerical results indicate that O3LS can reduce space overhead by 28.0% over standard layouts and 46.7% over sparse layouts without increasing the number of time steps, leading to suppression of logical error rates by up to 16% relative to larger data layout designs. O3LS can also achieve time overhead reductions of 36.07% and 24.76% in compact and standard data layout designs, respectively. It suppresses logical error rates by up to an order of magnitude compared to prior compilers that focus primarily on maximizing parallelism.

quant-ph↗

arXiv2026

Reconstructing Content with Collaborative Attention for Universal Multimodal Representation Learning

Multimodal embedding models, rooted in multimodal large language models (MLLMs), have yielded significant performance improvements across diverse tasks such as retrieval and classification. However, most existing approaches rely heavily on large-scale contrastive learning, with limited exploration of how the architectural and training paradigms of MLLMs affect embedding quality. While effective for generation, the causal attention and next-token prediction paradigm of MLLMs does not explicitly encourage the formation of globally compact representations, limiting their effectiveness as multimodal embedding backbones. To address this, we propose CoCoA, a Content reconstruction pre-training paradigm based on Collaborative Attention for multimodal embedding optimization. Specifically, we restructure the attention flow and introduce an EOS-based reconstruction task, encouraging the model to reconstruct input from the corresponding embeddings. This drives the multimodal model to compress the semantic information of the input into the token, laying the foundations for subsequent contrastive learning. Extensive experiments on MMEB-V1 demonstrate that CoCoA built upon Qwen2-VL and Qwen2.5-VL significantly improves embedding quality. Results validate that content reconstruction serves as an effective strategy to maximize the value of existing data, enabling multimodal embedding models generate compact and informative representations, raising their performance ceiling.

cs.IR↗

arXiv2025

High-dimensional quantum XYZ product codes for biased noise

Three-dimensional (3D) quantum XYZ product can construct a class of non-CSS quantum codes by using three classical codes. However, there has been limited study on their error-correcting performance so far and whether this code construction can be generalized to higher dimension is an open question. In this paper, we first study the error-correcting performance of the 3D Chamon code, which is an instance of the 3D XYZ product of three repetition codes. Second, we show that the 3D XYZ product can be generalized to four dimension and propose four-dimensional (4D) XYZ product code construction, which constructs a class of non-CSS quantum codes by using either four classical codes or two CSS quantum codes. Compared with the 4D homological product, we show that the 4D XYZ product can construct non-CSS codes with higher code dimension or code distance. Third, we consider two instances of the 4D XYZ product, to which we refer as the 4D Chamon code and the 4D XYZ product concatenated code, respectively. Our simulation results show that, the 4D XYZ product can construct non-CSS codes with better error-correcting performance against Pauli-$Z$-biased noise than CSS codes constructed by the 4D homological product. Finally, we present the geometric arrangement of the 4D Chamon code within a 4D cubic lattice, demonstrating that it possesses two key characteristics of fracton models, which strongly suggest that it is a novel 4D fracton model.

quant-ph↗

arXiv2025

Improved Belief Propagation Decoding Algorithms for Surface Codes

Quantum error correction is crucial for universal fault-tolerant quantum computing. Highly accurate and low-time-complexity decoding algorithms play an indispensable role in ensuring quantum error correction works effectively. Among existing decoding algorithms, belief propagation (BP) is notable for its nearly linear time complexity and general applicability to stabilizer codes. However, BP's decoding accuracy without post-processing is unsatisfactory in most situations. This article focuses on improving the decoding accuracy of BP over GF(4) for surface codes. Inspired by machine learning optimization techniques, we first propose Momentum-BP and AdaGrad-BP to reduce oscillations in message updating, breaking the trapping sets of surface codes. We further propose EWAInit-BP, which adaptively updates initial probabilities and provides a 1 to 3 orders of magnitude improvement over traditional BP for planar surface code, toric code, and XZZX surface code without any post-processing method, showing high decoding accuracy even under parallel scheduling. The theoretical $O(1)$ time complexity under parallel implementation and high accuracy of EWAInit-BP make it a promising candidate for high-precision real-time decoders.

quant-ph↗

arXiv2025

LifeIR at the NTCIR-18 Lifelog-6 Task

In recent years, sharing lifelogs recorded through wearable devices such as sports watches and GoPros, has gained significant popularity. Lifelogs involve various types of information, including images, videos, and GPS data, revealing users' lifestyles, dietary patterns, and physical activities. The Lifelog Semantic Access Task(LSAT) in the NTCIR-18 Lifelog-6 Challenge focuses on retrieving relevant images from a large scale of users' lifelogs based on textual queries describing an action or event. It serves users' need to find images about a scenario in the historical moments of their lifelogs. We propose a multi-stage pipeline for this task of searching images with texts, addressing various challenges in lifelog retrieval. Our pipeline includes: filtering blurred images, rewriting queries to make intents clearer, extending the candidate set based on events to include images with temporal connections, and reranking results using a multimodal large language model(MLLM) with stronger relevance judgment capabilities. The evaluation results of our submissions have shown the effectiveness of each stage and the entire pipeline.

cs.IR↗

arXiv2024

KnobTree: Intelligent Database Parameter Configuration via Explainable Reinforcement Learning

Databases are fundamental to contemporary information systems, yet traditional rule-based configuration methods struggle to manage the complexity of real-world applications with hundreds of tunable parameters. Deep reinforcement learning (DRL), which combines perception and decision-making, presents a potential solution for intelligent database configuration tuning. However, due to black-box property of RL-based method, the generated database tuning strategies still face the urgent problem of lack explainability. Besides, the redundant parameters in large scale database always make the strategy learning become unstable. This paper proposes KnobTree, an interpertable framework designed for the optimization of database parameter configuration. In this framework, an interpertable database tuning algorithm based on RL-based differentatial tree is proposed, which building a transparent tree-based model to generate explainable database tuning strategies. To address the problem of large-scale parameters, We also introduce a explainable method for parameter importance assessment, by utilizing Shapley Values to identify parameters that have significant impacts on database performance. Experiments conducted on MySQL and Gbase8s databases have verified exceptional transparency and interpretability of the KnobTree model. The good property makes generated strategies can offer practical guidance to algorithm designers and database administrators. Moreover, our approach also slightly outperforms the existing RL-based tuning algorithms in aspects such as throughput, latency, and processing time.

cs.AI↗

arXiv2024

Recursive expansion of Tanner graph: a method to construct stabilizer codes with high coding rate

Quantum stabilizer codes face the problem of low coding rate. In this article, following the idea of recursively expanding Tanner graph proposed in our previous work, we try to construct new stabilizer codes with high coding rate, and propose XZ-type Tanner-graph-recursive-expansion (XZ-TGRE) code and Tanner-graph-recursive-expansion hypergraph product (TGRE-HP) code. XZ-TGRE code have zero asymptotic coding rate, but its coding rate tends to zero extremely slowly with the growth of code length. Under the same code length, its coding rate is much higher than that of surface code. The coding rate of TGRE-HP is the constant 0.2, which is the highest constant coding rate of stabilizer codes to our best knowledge. We prove that the code distance of XZ-TGRE code scales as $O(log(N))$, and that of TGRE-HP code scales as $O(\log \sqrt{N})$, where $N$ is the code length. Moreover, the code capacity noise threshold of XZ-TGRE code is around 0.078, and that of TGRE-HP code is around 0.096. This articles shows that the idea of recursively expanding Tanner graph might have potential to construct quantum codes with good performance.

quant-ph↗

arXiv2024

CHEAT: A Large-scale Dataset for Detecting ChatGPT-writtEn AbsTracts

The powerful ability of ChatGPT has caused widespread concern in the academic community. Malicious users could synthesize dummy academic content through ChatGPT, which is extremely harmful to academic rigor and originality. The need to develop ChatGPT-written content detection algorithms call for large-scale datasets. In this paper, we initially investigate the possible negative impact of ChatGPT on academia,and present a large-scale CHatGPT-writtEn AbsTract dataset (CHEAT) to support the development of detection algorithms. In particular, the ChatGPT-written abstract dataset contains 35,304 synthetic abstracts, with Generation, Polish, and Mix as prominent representatives. Based on these data, we perform a thorough analysis of the existing text synthesis detection algorithms. We show that ChatGPT-written abstracts are detectable, while the detection difficulty increases with human involvement.Our dataset is available in https://github.com/botianzhe/CHEAT.

cs.CL↗

arXiv2024

Hypergraph product code with 0.2 constant coding rate and high code capacity noise threshold

The low coding rate of quantum stabilizer codes results in formidable physical qubit overhead when realizing quantum error correcting in engineering. In this letter, we propose a new class of hypergraph-product code called TGRE-hypergraph-product code. This code has constant coding rate 0.2, which is the highest constant coding rate of quantum stabilizer codes to our best knowledge. We perform simulations to test the error correcting capability TGRE-hypergraph-product code and find their code capacity noise threshold in depolarizing noise channel is around 0.096.

quant-ph↗