Searcharxiv⌕ Search

arXiv subjects

Jiaqi Huang

Publications and source records attributed to Jiaqi Huang.

At least 19 recordsLinked to original sources

Classification of ${\rm GL}_1(D)\times {\rm GL}_{n-1}(D)$-distinguished representations

Let $D$ be a quaternion division algebra over a non-Archimedean local field $F$ of characteristic zero. We prove the conjecture of Dagar, Sharma, and Verma classifying irreducible smooth representations of $\mathrm{GL}_n(D)$ distinguished by $\mathrm{GL}_1(D)\times\mathrm{GL}_{n-1}(D)$ for $n>2$. These are precisely the trivial representation and the irreducible normalized parabolic inductions $\one_{n-2}\timesτ$, where $τ$ is an infinite-dimensional irreducible representation of $\mathrm{GL}_2(D)$ distinguished by $\mathrm{GL}_1(D)\times\mathrm{GL}_1(D)$. The proof combines a choice of cuspidal inducing factor with orbit conditions and multisegment methods.

math.RT↗

Semantic Validation of Packer Identification Tools: Characterization, Repair, and Downstream Impact

Packer identification tools are essential to malware analysis, yet their semantic correctness remains largely unverified. Incorrect packer labels can trigger inappropriate unpacking and degrade downstream malware classification. This paper presents a framework for systematically exposing, characterizing, and repairing semantic errors in packer identification tools. Our key insight is to leverage unpackers as executable semantic contracts, enabling automated validation of packer predictions without requiring manually labeled packer ground truth. We develop a pipeline that normalizes heterogeneous tool outputs, localizes faulty decision logic, and repairs identification errors through complementary signatures, heuristic rules, and static unpacker-derived detection predicates. Our empirical evaluation reveals widespread identification failures, with an average packer family identification recall of only 44.8% and an average packedness detection recall of only 63.6%. The proposed two-stage repairs achieve an approximately 100.2% relative improvement in average packer family identification recall, while increasing packedness detection recall to 99.0%-99.5%. We further demonstrate that semantic errors can propagate through unpacking failures and adversely affect downstream malware classification. Correcting these errors improves average classification precision by 13.6 percentage points.

cs.CR↗

Noise suppression of quantum frequency conversion in thin-film lithium niobate waveguides via adapted poling

Quantum frequency conversion (QFC) enables optical interfaces between dissimilar quantum systems, but suppressing noise while maintaining high conversion efficiency remains challenging. Here, we demonstrate that adapted poling in thin-film lithium niobate (TFLN) waveguides reduces the pump-power requirement while shifting the dominant noise mechanism and its power scaling. In 2-cm-long waveguides, adapted poling enables maximum conversion efficiency at an on-chip pump power of $19\pm0.8$~mW, reducing the required pump power by 8~dB and the total conversion noise by 14~dB relative to conventional periodic poling at their respective maximum-conversion operating points. Controlled measurements with different pump conditions and device configurations reveal a transition from predominantly quadratic noise scaling in periodically poled devices to predominantly linear scaling in adapted-poling devices. This transition accompanies a shift from noise dominated by cascaded second-harmonic generation and Stokes spontaneous Raman scattering to direct anti-Stokes spontaneous Raman scattering. These results establish adapted poling as an effective approach to suppressing nonlinear noise, enabling low-noise integrated frequency converters for quantum networks.

physics.optics↗

Comparative Analysis of Blockchain Systems

Blockchain is a type of decentralized distributed database. Unlike traditional relational database management systems, it does not require management or maintenance by a third party. All data management and update processes are open and transparent, solving the trust issues of centralized database management systems. Blockchain ensures network-wide consistency, consensus, traceability, and immutability. Under the premise of mutual distrust between nodes, blockchain technology integrates various technologies, such as P2P protocols, asymmetric encryption, consensus mechanisms, and chain structures. Data is distributed and stored across multiple nodes, maintained by all nodes, ensuring transaction data integrity, undeniability, and security. This facilitates trusted information sharing and supervision. The basic principles of blockchain form the foundation for all related research. Understanding the working principles is essential for further study of blockchain technology. There are many platforms based on blockchain technology, and they differ from one another. This paper will analyze the architecture of blockchain systems at each layer, focusing on the principles and technologies of blockchain platforms such as Bitcoin, Ethereum, and Hyperledger Fabric. The analysis will cover their scalability and security and highlight their similarities, differences, advantages, and disadvantages.

cs.CR↗

ECHO-2: A Large-Scale Distributed Rollout Framework for Cost-Efficient Reinforcement Learning

Reinforcement learning (RL) is a critical stage in post-training large language models (LLMs), involving repeated interaction between rollout generation, reward evaluation, and centralized learning. Distributing rollout execution offers opportunities to leverage more cost-efficient inference resources, but introduces challenges in wide-area coordination and policy dissemination. We present ECHO-2, a distributed RL framework for post-training with remote inference workers and non-negligible dissemination latency. ECHO-2 combines centralized learning with distributed rollouts and treats bounded policy staleness as a user-controlled parameter, enabling rollout generation, dissemination, and training to overlap. We introduce an overlap-based capacity model that relates training time, dissemination latency, and rollout throughput, yielding a practical provisioning rule for sustaining learner utilization. To mitigate dissemination bottlenecks and lower cost, ECHO-2 employs peer-assisted pipelined broadcast and cost-aware activation of heterogeneous workers. Experiments on GRPO post-training of LLMs ranging from 4B to 32B parameters under real wide-area bandwidth regimes show that ECHO-2 significantly improves cost efficiency while preserving RL reward comparable to strong baselines.

cs.LG↗

EndoPrior-GS: Dynamic Endoscopic Reconstruction with a Joint Texture Prior

Dynamic endoscopic reconstruction is fundamental to robotic surgery and computer-assisted interventions. While 3D Gaussian Splatting (3DGS) realises real-time rendering, its application to deformable intraoperative environments remains constrained by spurious geometry and varying illuminations. To address these limitations, we introduce EndoPrior-GS, a novel pipeline that explicitly couples frame-extracted vision heuristics and estimated depth maps. EndoPrior-GS derives a joint texture prior from a tool-filtered valid tissue mask, a non-specular photometric filter, and anatomical structural salience, yielding a probability map that guides primitive initialisation and subsequent density control. The prior is further extended to the temporal domain through a texture-aware term that dynamically weighs pairwise primitive contributions during training. We conduct extensive experiments on benchmark datasets EndoNeRF and SCARED, and the obtained results show that our method EndoPrior-GS reduces Flow Error by 27.7% and 25.8% over the representative approaches while preserving competitive rendering quality and real-time rendering speed. Our project website is available at https://jiaqi-huang-77.github.io/EndoPrior-GS/.

cs.CV↗

SequenceO1: End-to-End Ultra-Long (100K) Sequence Modeling in Recommendation with Low-Rank Caching

Modeling long-term user behavior is central to sequential recommendation and billion-scale industrial recommender systems, yet production ranking models operate under strict latency, memory, communication, and training-throughput constraints. At the 100K scale, the challenge extends beyond attention complexity: raw sequence features must be stored, transferred, and repeatedly processed during training and online serving. Existing approaches based on history truncation, multi-stage behavior retrieval, compressed lifelong histories, or train-short/infer-long extrapolation either weaken end-to-end optimization or retain substantial length-dependent cost. We present SequenceO1, an end-to-end framework for ultra-long user behavior sequence modeling, deployed at full traffic on Douyin with histories of up to 100K interactions. SequenceO1 follows a compress-then-reason design. Its Sketch Attention (SA) uses learnable prototypes and prototype-wise normalization to compress the raw history into a fixed-size, target-agnostic user representation. Target-conditioned Stacked Target-to-History Cross Attention (STCA) then models complementary time scales: a recent 10K suffix for short-term interests and the compact sketch for long-term preferences. To make training and inference practical, SequenceO1 combines low-rank user representation caching, multi-request user-level batching, pipeline lift, and a fused FlashSA kernel to amortize feature storage, communication, and computation across targets, training instances, and consecutive requests. Production experiments show consistent offline and online gains, while the compact cached sketch retains most of the benefit of directly scaling end-to-end sequence ranking to 100K. These results provide a practical model-system approach to efficient attention, sequence compression, and scalable long-sequence and long-context recommendation systems.

cs.IR↗

DynaKRAG: A Unified Framework for Learnable Evidence Control in Multi-Hop Retrieval-Augmented Generation

Multi-hop retrieval-augmented generation (RAG) acquires evidence sequentially, with each document contributing supporting facts, bridge entities, query refinements, or sufficient evidence for answering. Evidence acquisition can involve iterative retrieval, query reformulation, evidence assessment, and sufficiency checking. We introduce DynaKRAG, a unified evidence-action framework that learns a shared state-conditioned policy for coordinating these operations. At each step, a deterministic validity layer constructs the executable action set, a learned continuation gate selects between answer generation and further evidence acquisition, and a learned advantage scorer ranks feasible evidence operations by their predicted gain relative to immediate answer generation. The selected operation updates the shared state and may enable additional operations. Across HotpotQA, 2Wiki, and MuSiQue with Qwen2.5-7B, GPT-4o-mini, and Llama-3.1-8B, DynaKRAG ranks first among the compared methods in both EM and F1 for all nine dataset--backbone pairs. Relative to matched-backbone baseline method, DynaKRAG improves F1 in every pair while achieving total-token efficiency gains of 10.1--34.3\% and retrieval-call efficiency gains of 15.1--43.4\%, establishing Pareto dominance under these measures. With Qwen2.5-7B, terminal evidence compression further improves answer quality across all three datasets while reducing the context passed to final answer generation by 54.4\%--71.5\%. These results demonstrate that unified, state-conditioned evidence control supports strong answer quality, efficient retrieval, and compact answer-generation contexts.

cs.CL↗

In-context superposition: human-like working memory interference in large language models

Intelligent systems must maintain and manipulate task-relevant information online to adapt to dynamic environments. This capacity, known as working memory, is fundamental to human reasoning. Yet, human working memory is strikingly limited, maintaining only three to four items in a brain with billions of neurons. Surprisingly, large language models (LLMs), despite different substrates and direct access to prior context through attention, exhibit similar working memory limitations. Why should such different systems face analogous constraints? We propose that working memory limitations reflect a general trade-off of shared representations: representational compression and reuse support efficient learning and generalization, but also cause simultaneously active representations to interfere. We show a two-layer transformer trained on a working memory task can solve it perfectly, but diverse trained LLMs exhibit human-like limitations: performance declines with memory load, while retrieval is biased by recency and stimulus statistics. Mirroring humans, working memory performance in LLMs is also associated with broader model capability. Mechanistically, we show that LLMs encode multiple memories in entangled representations --- a condition we call \emph{in-context superposition} --- and progressively suppress competing content while aligning the target with the readout. Moreover, a causal intervention that suppresses interfering information improves performance. Together, these findings suggest that working memory capacity reflects the ability to select task-relevant information under interference, a computational challenge shared by biological and artificial systems.

cs.LG↗

Learning Visual Spatial Planning from Symbolic State via Modality-Gap-Aware Self-Distillation

While Vision-Language Models excel at general multimodal understanding, they still struggle with visual spatial planning. We attribute this limitation to a perception--reasoning modality gap. Visual planning requires models to infer latent state structures from pixels and then reason over the recovered structure to produce valid actions, whereas symbolic planning directly leverages explicit representation. This discrepancy introduces two sequential bottlenecks: visual state recovery at the perception stage and multi-step planning at the reasoning stage. To address this, we propose MGSD, a two-stage modality-gap-aware self-distillation framework. First, a cold-start grounding stage establishes reliable visual state recovery before on-policy training. Second, a symbol-guided on-policy self-distillation stage transfers the privileged teacher's planning behavior to the student through token-level supervision on student-generated prefixes. Crucially, symbolic information is used only during training, while inference relies exclusively on visual inputs. Experiments on visual planning benchmarks show that MGSD consistently improves performance across different model scales, raising the macro average by 19.3% and 18.4%, respectively. The resulting models substantially reduce the gap to the upper bounds obtained with symbolic inputs. Ablation studies and diagnostic analyses further confirm that the gains arise from improvements in both visual state recovery and optimal-path reasoning. These results demonstrate that MGSD strengthens not only the recovery of actionable states from visual observations but also the ability to plan over the inferred structures. Code is available at https://github.com/Oranger-l/MGSD.

cs.AI↗

Theoretical Analysis of Engression and Reverse Markov Engression

Engression is a recently proposed and effective framework for conditional distribution learning. Its multi-step Reverse Markov extension further improves generative flexibility by decomposing complex conditional sampling into sequential reverse transitions. Despite their strong empirical performance, rigorous finite-sample statistical guarantees for these methods remain unavailable. In this paper, under deep neural network parameterizations, we establish nonasymptotic convergence bounds for Engression by directly controlling the Energy Distance between the learned and target conditional distributions. For the Reverse Markov framework, we further develop an Energy-Distance-based chain rule that enables a rigorous analysis of error propagation across reverse steps. Our analysis yields corresponding excess-risk bounds that are near-optimal up to logarithmic factors relative to the classical minimax rate over a general Hölder class.

stat.ME↗

On-Policy Replay for Continual Supervised Fine-Tuning

Continual supervised fine-tuning (SFT) is the de facto recipe for adapting large language models (LLMs) to a stream of downstream tasks, but it suffers from catastrophic forgetting of earlier capabilities. Recent work shows that on-policy signals -- training on the model's own outputs -- reduce forgetting more reliably than off-policy supervision. Existing on-policy methods route this signal through a new training objective (e.g., self-distillation losses with a teacher copy), inheriting an extra forward pass, schedule sensitivity, and stylistic drift from the teacher.We instead route the on-policy signal through the training data source. Our method, On-Policy Replay (OPR), rolls out the most recent checkpoint on a small budget of historical prompts, filters the generations by a task reward, and replays the surviving (prompt, model response) pairs as ordinary SFT examples. There is no teacher, no auxiliary loss, and no on-the-fly distillation. Across three 7--8B instruction-tuned backbones (Qwen2.5-7B-Instruct, Qwen3-8B, Llama3.1-8B-Instruct) on the TRACE continual-learning benchmark, OPR consistently reduces forgetting; on the sharpest stress test (Qwen2.5-7B-Instruct, Sequential SFT BWT -13.93), OPR lifts BWT to -0.65 at a 10% replay budget and to -2.29 at a 1% budget -- a 46% reduction in |BWT| over a tuned Vanilla Replay baseline, with 42--46% reductions observed across all three backbones. We give a KL-shrinkage interpretation that places OPR and prior on-policy distillation methods on a single axis, and we present a counterintuitive finding that explains why Vanilla Replay is already a strong baseline: low-score replay is uniformly worse than Vanilla Replay, demonstrating that the active ingredient in OPR is the on-policy distribution, not the response quality alone.Our code is available at https://github.com/Yancey2024/OnPolicyReplay.

cs.LG↗

Divide-and-Conquer Inference for Large-Scale Visual Recognition with Multimodal Large Language Models

Multimodal Large Language Models (MLLMs) have demonstrated strong capabilities across a wide range of vision language tasks. However, when applied to large scale image classification, their performance degrades significantly as the label space expands a phenomenon we define as Performance Collapse in Long Sequence Recognition. Through an information theoretic analysis, we reveal that this collapse stems from a fundamental conflict between the escalating information entropy and the prominent attention dilution and decay within attention mechanisms, which impairs the model's ability to maintain a sufficient signal-to-noise ratio when processing extremely long prompts. To mitigate this, we propose Divide-and-Conquer Inference (DCI), a novel test-time scaling strategy for visual recognition with MLLMs. DCI recursively decomposes complex global classification tasks into multiple simpler, localized subproblems and employs a dynamic pruning mechanism to compress the search space. This method effectively improves the local signal to noise ratio and model accuracy by mitigating the inherent weight dilution issues in long-sequence inference. Moreover, while traditional self-attention incurs a prohibitive quadratic computational complexity, DCI achieves more favorable scaling behavior and substantially accelerates inference in large scale classification scenarios. Extensive experiments on benchmarks such as ImageNet-1K and ImageNet-21K demonstrate that DCI consistently improves classification accuracy. This enables lightweight open-source models to rival or even surpass frontier closed-source giants without any additional training or fine-tuning. As a model-agnostic, plug-and-play paradigm, DCI offers an efficient approach for scaling the inferential precision of MLLMs in large-scale scenarios.

cs.CV↗

EagleVision: A Multi-Task Benchmark for Cross-Domain Perception in High-Speed Autonomous Racing

High-speed autonomous racing presents extreme perception challenges, including large relative velocities and substantial domain shifts from conventional urban-driving datasets. Existing benchmarks do not adequately capture these high-dynamic conditions. We introduce EagleVision, a unified LiDAR-based multi-task benchmark for 3D detection and trajectory prediction in high-speed racing, providing newly annotated 3D bounding boxes for the Indy Autonomous Challenge dataset (14,893 frames) and the A2RL Real competition dataset (1,163 frames), together with 12,000 simulator-generated annotated frames, all standardized under a common evaluation protocol. Using a dataset-centric transfer framework, we quantify cross-domain generalization across urban, simulator, and real racing domains. Urban pretraining improves detection over scratch training (NDS 0.72 vs. 0.69), while intermediate pretraining on real racing data achieves the best transfer to A2RL (NDS 0.726), outperforming simulator-only adaptation. For trajectory prediction, Indy-trained models surpass in-domain A2RL training on A2RL test sequences (FDE 0.947 vs. 1.250), highlighting the role of motion-distribution coverage in cross-domain forecasting. EagleVision enables systematic study of perception generalization under extreme high-speed dynamics. The dataset and benchmark are publicly available at https://avlab.io/EagleVision

cs.RO↗

Quantum Borcherds-Bozec Superalgebras

We introduce quantum Borcherds-Bozec superalgebras. We present and prove various results of the quantum superalgebras including a bilinear form, higher Serre relation, quasi-R-matrix, character formula for the irreducible highest weight modules. We also prove the category of integrable representations is semi-simple.

math.QA↗

SAM-R1: Leveraging SAM for Reward Feedback in Multimodal Segmentation via Reinforcement Learning

Leveraging multimodal large models for image segmentation has become a prominent research direction. However, existing approaches typically rely heavily on manually annotated datasets that include explicit reasoning processes, which are costly and time-consuming to produce. Recent advances suggest that reinforcement learning (RL) can endow large models with reasoning capabilities without requiring such reasoning-annotated data. In this paper, we propose SAM-R1, a novel framework that enables multimodal large models to perform fine-grained reasoning in image understanding tasks. Our approach is the first to incorporate fine-grained segmentation settings during the training of multimodal reasoning models. By integrating task-specific, fine-grained rewards with a tailored optimization objective, we further enhance the model's reasoning and segmentation alignment. We also leverage the Segment Anything Model (SAM) as a strong and flexible reward provider to guide the learning process. With only 3k training samples, SAM-R1 achieves strong performance across multiple benchmarks, demonstrating the effectiveness of reinforcement learning in equipping multimodal models with segmentation-oriented reasoning capabilities.

cs.CV↗

Fixed Effects as Generated Regressors

Many economic models feature moment conditions that involve latent variables. When the latent variables are individual fixed effects in an auxiliary panel data regression, we construct orthogonal moments that eliminate first-order bias induced by estimating the fixed effects. Machine Learning methods and Empirical Bayes methods can be used to improve the estimate of the nuisance parameters in the orthogonal moments. We establish a central limit theorem based on the orthogonal moments without relying on exogeneity assumptions between panel data residuals and the cross-sectional moment functions. In a simulation study where the exogeneity assumption is violated, the estimator based on orthogonal moments has smaller bias compared with other estimators relying on that assumption. An empirical application on experimental site selection demonstrates how the method can be used for nonlinear moment conditions.

econ.EM↗

Implicit Strategic Optimization: Rethinking Long-Horizon Decision-Making in Adversarial Poker Environments

Training large language model (LLM) agents for adversarial games is often driven by episodic objectives such as win rate. In long-horizon settings, however, payoffs are shaped by latent strategic externalities that evolve over time, so myopic optimization and variation-based regret analyses can become vacuous even when the dynamics are predictable. To solve this problem, we introduce Implicit Strategic Optimization (ISO), a prediction-aware framework in which each agent forecasts the current strategic context and uses it to update its policy online. ISO combines a Strategic Reward Model (SRM) that estimates the long-run strategic value of actions with iso-grpo, a context-conditioned optimistic learning rule. We prove sublinear contextual regret and equilibrium convergence guarantees whose dominant terms scale with the number of context mispredictions; when prediction errors are bounded, our bounds recover the static-game rates obtained when strategic externalities are known. Experiments in 6-player No-Limit Texas Hold'em and competitive Pokemon show consistent improvements in long-term return over strong LLM and RL baselines, and graceful degradation under controlled prediction noise.

cs.LG↗