Searcharxiv⌕ Search

arXiv subjects

Xinyu Chen

Publications and source records attributed to Xinyu Chen.

At least 19 recordsLinked to original sources
arXiv2026

InfiniLoRA: Disaggregated Multi-LoRA Serving for Large Language Models

LoRA enables efficient customization of LLMs and is widely used in multi-tenant and multi-task serving. However, emerging model architectures such as MoE significantly increase LoRA memory cost, making existing coupled LoRA serving designs poorly scalable and prone to tail-latency inflation. We present InfiniLoRA, a disaggregated LoRA serving system that decouples LoRA execution from base-model inference. InfiniLoRA introduces a shared LoRA Server with parallelism-aware execution, SLO-driven provisioning, and critical-path optimizations, including GPU-initiated communication and hardware-specialized LoRA kernels. Experiments show that InfiniLoRA can achieve an average $3.05\times$ increase in serviceable request rate under strict latency SLOs, and improve the percentage of LoRA adapters satisfying the SLO requirement by 54.0\%.

cs.DC↗
arXiv2026

Polynomial-Time Minimax Prediction for Stationary Renewal Processes

The minimax next-symbol Kullback-Leibler risk for stationary binary renewal processes is known to have order n^(-1/2). We give a polynomial-time predictor attaining this order for an arbitrary unknown finite-mean interarrival law. The construction uses a classical type-uniform renewal code. Its contribution is an exact dynamic program for censored prefix marginals, avoiding enumeration of gap histograms or binary completions. Combined with an established prediction reduction, it gives risk at most [log M_(n+1)+log(n+2)]/n, where M_N counts renewal types. Shared suffix updates reduce the total computation to O(n^3) arithmetic work and polynomial bit complexity, without tail, mixing, known-mean, or bounded-support assumptions. The statistical rate and its lower bound are prior results; the proposed advance concerns computation.

cs.IT↗
arXiv2026

Spatial Token Interleaving and Two-Pass BP Decoding for Semantic Image Transmission

Semantic error control coding (SECC) can correct the residual errors of a standard channel code with a foundation model that supplies a prior on the source content, but existing SECC schemes are designed for text. In this paper, we develop SECC for image transmission with spatial token interleaving and a two-pass belief propagation (BP) schedule. At the transmitter, a vector-quantized (VQ) tokenizer converts an image into tokens, which are transmitted in multiple low-density parity-check (LDPC) code blocks of fifth-generation New Radio (5G NR). In raster order, a block that BP fails to decode corrupts consecutive token rows, so the interleaver reorders the tokens such that adjacent tokens are carried by different blocks, without additional transmitted bits. At the receiver, after the first BP pass, token recovery fuses the token probabilities given by BP with the predictions of the autoregressive model Emu3-Gen, and the second BP pass decodes each block again with the fused distributions as priors on the information bits. On Kodak24 at 0.5 channel uses per pixel and a signal-to-noise ratio (SNR) of 0.15 dB, the second pass removes 92% of the token errors left by token recovery, and interleaving reduces the final token error rate by 88%. At medium to high SNR, the proposed receiver exceeds deep joint source--channel coding baselines with the same channel uses by at least 1.2 dB in peak signal-to-noise ratio.

cs.IT↗
arXiv2026

A Mixed-Precision CNN-LSTM Accelerator Featuring Long-Kernel Memory Reuse and Dynamic Quantization for Sleep Staging

This letter presents a mixed-precision CNN-LSTM accelerator for EEG-based sleep staging. To efficiently execute long temporal convolutions, a serpentine dataflow propagates activations between neighboring processing elements, reducing memory accesses by 6.2 times and 2.25 times compared with output- and weight-stationary dataflows, respectively. A hybrid quantization scheme applies runtime dynamic quantization to the LSTM and MLP layers and assigns mixed precision on a layer-by-layer basis, reducing model size by 4.16 times. Combined with a bit-flexible MAC architecture and channel-grouped memory scheduling, the accelerator supports INT2/4/8/16 inference. Fabricated in 65 nm CMOS, the accelerator consumes 0.40 mW and achieves a 258 times speedup over an iso-frequency 32b RISC-V processor. Evaluated across all 62 subjects in the public MASS-SS3 dataset using 31-fold leave-two-subject-out cross-validation, the mixed-precision approach with hybrid quantization maintains 5-class accuracy within 0.15% of the floating-point baseline when optimized for general sleep staging and N3 sensitivity within 0.51% when optimized for deep-sleep detection. The developed techniques are broadly applicable to EEG-based DL workloads.

cs.AR↗
arXiv2026

SemantiScene-Edit: Expert-Synthesized Supervision and Dual-Path Source Conditioning for Instruction-Based Video Editing

Instruction-guided video editing requires executing a requested change while preserving the source video's appearance and temporal structure. We present SemantiScene-Edit, a video editor that couples a multimodal encoder with a pretrained video diffusion transformer. The encoder jointly processes source frames and the editing instruction, and learnable queries convey their semantic context to the generator through a trainable connector and cross-attention. A complementary visual pathway concatenates source-video latents with noisy target latents before patch embedding, supplying dense spatial and temporal information for content preservation. We train the model with flow matching and a three-stage image--video curriculum comprising semantic alignment, joint adaptation, and high-resolution refinement. To support training, we construct over 1.2 million instruction-conditioned video-editing records covering object addition, removal, replacement, and attribute modification. We also construct a 100-case benchmark for evaluating instruction execution, source preservation, temporal consistency, and visual quality. Comparative experiments show that SemantiScene-Edit effectively follows editing instructions while retaining source content. At inference, the model requires only a source video and a natural-language instruction.

cs.AI↗
arXiv2026

Low Depth Distributed Quantum Algorithms for Unordered Database Search

Grover's algorithm accelerates unstructured database search quadratically compared to classical algorithms. In the NISQ era, distributed quantum computing can decrease circuit depth and reduce noise. In this paper, an algorithm for constructing query operators for subfunctions is proposed. By dividing the target string of the search problem into several substrings and integrating the query operator of each subfunction, a low-depth distributed exact quantum search algorithm is designed. The contributions of this paper are as follows: (1) The proposed distributed algorithm has a lower circuit depth and can mitigate error accumulation compared to distributed quantum search algorithms; (2) The target can be accurately located by the proposed distributed algorithm; (3) Experiments conducted with the quantum software MindQuantum confirm the effectiveness and feasibility of the proposed distributed algorithm. Moreover, the introduction of noise to the circuit during these experiments indicates that the algorithm possesses an inherent capacity for noise resistance.

quant-ph↗
arXiv2026

DEFLECT: Temporal Counterfactual Preference Learning for Delay-Robust Asynchronous VLAs

Vision-Language-Action (VLA) policies increasingly rely on asynchronous inference to hide model latency behind ongoing robot motion. During inference, the robot continues executing its current action chunk while the next chunk is generated from the observation available at inference start. By the time the new chunk is executed, the robot or scene may have changed, so an action that fits the prediction-time state can be misaligned with the execution-time state. Training-time methods modify the policy input or supervised target based on the expected execution delay, but do not directly supervise how actions should change as the scene evolves between prediction and execution. Runtime repair methods improve chunk continuity or modify actions during deployment, but can still leave actions aligned with the prediction-time state rather than the execution-time state. Our key insight is that execution-time observations are unavailable when the online prediction is made but are available in recorded trajectories during offline training, making the prediction-execution mismatch observable during training. We propose DEFLECT, an offline post-training framework that turns this temporal contrast into preference supervision for a policy acting from stale inputs. Across Kinetix, LIBERO, and three real-robot tasks, DEFLECT improves delay robustness across the evaluated VLA policies and settings. At the longest supported delay, it improves success by 4.2 percentage points over VLASH on Kinetix and by 22.7 points over naive asynchronous execution with GR00T N1.7 on LIBERO.

cs.RO↗
arXiv2026

PACT-VLA: Speculative Action-Chunk Verification with Pre-Commit Observations

Vision-language-action (VLA) policies are commonly deployed with asynchronous inference in real-world settings, where the next action chunk is generated while the robot is still executing the previous one. By the time the new chunk is committed, the robot and scene have changed, so the chunk is conditioned on a stale observation and task success drops sharply as this inference delay grows. Generating multiple candidate chunks could let newer observations inform the final choice, but existing multi-candidate methods either rank candidates with the stale inference-start observation or repeatedly switch between candidates during execution, which can make the executed action sequence discontinuous. Our key insight is that inference delay can be used as decision time. While candidate generation is still underway, candidate selection can reuse the VLA itself to choose the complete action chunk that best fits a later observation. We propose PACT-VLA, a training-free framework that generates candidates from the inference-start observation and selects one using this pre-commit observation. Its scoring function, FlowRes, measures the residual between each candidate's implied flow direction and the velocity predicted by the frozen policy under the pre-commit observation, requiring no separate verifier. A pipelined implementation overlaps observation processing with candidate generation and batches scoring, so selection adds no control-step delay on the evaluated hardware. At a four-step delay, PACT-VLA improves average success by 15.0 percentage points over RTC on Kinetix and by 5.2 points over single-candidate asynchronous execution with GR00T N1.7 on LIBERO, and it further improves existing delay-robust methods. On three real-robot tasks, it raises average success from 60% to 82% over naive asynchronous execution with the same frozen $π_{0.5}$ checkpoint.

cs.RO↗
arXiv2026

MSCR: Jointly Balancing Utilization Across Modalities and Discovering Synergistic Information

In multimodal learning, modality imbalance often lets one modality dominate optimization and keeps the model from fully exploiting cross-modal interactions. Balance-oriented methods such as the Multimodal Competition Regularizer (MCR) perform well by alleviating this imbalance during training. These dynamic balancing methods, however, mainly adjust modality contributions and do not consider how imbalance affects the formation of synergistic information during fusion. Synergy is a major source of multimodal advantage, because it is predictive gain that cannot be recovered from unimodal evidence alone. We propose the Multimodal Synergistic Competition Regularizer (MSCR), a fusion method that jointly promotes balanced modality utilization and the discovery of synergistic information. MSCR has two complementary objectives. An anti-dominance objective suppresses excessive unimodal influence and leaves room for synergistic information, and a synergy objective encourages the fused predictor to outperform a non-synergistic unimodal reference. An adaptive gate further concentrates the regularization on samples that need stronger collaborative modeling. Experiments on five benchmarks show that MSCR consistently improves performance over MCR and strengthens synergistic capability. It raises conditional recovery, the accuracy of the fused prediction on samples that the unimodal predictors misclassify, as well as the share of the model's output that InterSHAP attributes to cross-modal interactions. Its gain in conditional recovery persists when the fusion module is Transformer-based. Code is available at https://github.com/chenxinyu991-collab/MSCR.

cs.AI↗
arXiv2026

RAM: Reachability Across Morphologies

Many stages of the robotic lifecycle, from morphology synthesis to operation, rely fundamentally on the reachable workspace. However, current methods for approximating workspaces are slow, imprecise, or tied to a single morphology. We introduce Reachability Across Morphologies (RAM): a morphology-conditioned, implicit neural representation that acts as a fast, differentiable surrogate for pose reachability, generalising to unseen morphologies while inherently accounting for self-collisions. RAM is trained on a large-scale dataset of $3\cdot10^{10}$ samples generated solely from forward kinematics, which we publish alongside the model. Experiments show that RAM reaches 84\% balanced accuracy at nanosecond inference, 15 percentage points higher than a generative inverse kinematics baseline that also generalises across morphologies, while running three orders of magnitude faster. This speed carries over to downstream tasks, accelerating morphology and trajectory optimisation by one to two orders of magnitude.

cs.RO↗
arXiv2026

Explicit Bounds for Restricted Invertibility with Prescribed Group Cardinalities

We prove an explicit lower spectral bound for independent random vectors with prescribed group multiplicities. Let $v_{ij}\in\mathbb C^d$, $1\le i\le l$, $1\le j\le k_i$, be independent random vectors with finite support and $\mathbb E v_{ij}v_{ij}^*=A_i$, where $\sum_{i=1}^l A_i=I_d$ and $\operatorname{Tr}(A_i)\geε$. Set $k=\max_i k_i$, $n=\max_i\operatorname{rank}(A_i)$, and $K=\sum_{i=1}^l k_i$. If $n>(6+4\sqrt2)k$ and $2\sqrt{nk}-k<ε\le n/2$, then, with positive probability, \[ λ_K\left(\sum_{i=1}^l\sum_{j=1}^{k_i}v_{ij}v_{ij}^*\right) \ge n-(\sqrt{k}+\sqrt{n-ε})^2. \] As a consequence, we obtain a restricted invertibility result selecting exactly $k_i$ vectors from the $i$th group of a Parseval frame. The proof combines interlacing families, a reciprocal transformation of mixed characteristic polynomials, and a multivariate barrier estimate that accommodates repeated differentiation in the same variable.

math.FA↗
arXiv2026

DRelay: Global Draft Context for Prefix-Aware Parallel Speculative Decoding Repair

Parallel drafting reduces the drafting overhead of speculative decoding for large language models (LLMs), but its gains remain limited by the accepted prefix length. Even when the correct token is present in the candidate pool, a single early selection error prevents subsequent predictions from being used. We propose DRelay, which uses global information from the entire draft block to perform prefix-aware selective repair of candidate selections before target-model verification. DRelay bases its decisions on candidate correlations and the selected path: a global reader extracts predictive information across positions for each candidate. While a causal selector combines candidate-level information extracted by the global read with the tokens selected at preceding positions to determine whether the native choice at the current position is consistent with the global evidence and the selected prefix. It then decides whether to retain or replace the token, thereby repairing early errors and extending the accepted prefix. We further jointly train the draft backbone and the selector, combining candidate-support learning with a repair objective, while weighting the repair loss according to each block position's potential contribution to the consecutive accepted prefix. Across eight diverse benchmarks on an H800 GPU, DRelay consistently improves both average acceptance length and end-to-end decoding performance over DFlash, Domino, and DSpark. Under SGLang serving, DRelay improves average end-to-end speedup over DFlash, Domino, and DSpark by 14.7%-16.8%, 8.7%-9.3%, and 8.1%-9.3%, respectively.

cs.AI↗
arXiv2026

OneStreamer: Unifying Perception, Memory, and Proactive Response in Streaming Video Interaction

Streaming video LLMs must retain evidence before its relevance to future tasks is known and respond when sufficient evidence becomes available. The challenge is to form reusable factual memory without compromising real-time perception. We introduce OneStreamer, which jointly learns query-independent evidence recording and task response through a shared proactive generation process. Its Proactive Hierarchical Caption Memory (PHCM) produces time-grounded local-detail captions and summaries of completed events. Streaming caption targets supervise the interpretation of observed video prefixes during training. At inference, model-generated records complement a recent visual window, providing reusable factual context without revisiting historical visual features. Proactive State Transition Learning (PSTL) reduces the dominance of repeated waiting states by preserving supervision at all output anchors and selecting representative state-change and state-persistence tokens. We further develop a streaming data synthesis pipeline that aligns output content and timing with available evidence. Combining the resulting streaming captions and QA with cleaned open-source data yields OneStreamer-1M, a broad-coverage streaming video interaction dataset with over one million records spanning diverse tasks. Our 4B model achieves the best results among the compared methods across all eight evaluated streaming video understanding benchmarks. Ablations show that retaining generated captions improves historical QA without degrading real-time perception. PSTL also outperforms dense state supervision while supervising only 27.5% of annotated state tokens. Together, these results support proactive generation as a shared learning interface connecting perception, memory formation, and timely response in streaming video interaction.

cs.CV↗
arXiv2026

PostCam: Camera-Controllable Novel-View Video Generation with Query-Shared Cross-Attention

We propose PostCam, a streamlined framework for novel-view video generation that achieves superior detail preservation and precise camera trajectory editing in dynamic scenes. Current methods often struggle with a trade-off between pose-based control, which lacks visual detail, and rendering-based guidance, which is overly sensitive to geometric accuracy. Despite recent hybrid attempts, achieving precise motion and visual consistency remains challenging due to the lack of effective cross-modal alignment. We argue that robust control stems from the deep alignment of multimodal signals rather than increased input complexity. Our core contribution is the Query-Shared Cross-Attention mechanism, which projects 6-DoF poses and rendered features into a unified latent space. This allows the model to spontaneously achieve intrinsic consistency between motion cues and pixel-level guidance during denoising. Experiments demonstrate that PostCam maintains high-fidelity visual details while outperforming state-of-the-art methods by 20% in trajectory precision, exhibiting superior robustness in complex dynamic scenes. Our project webpage is publicly available at: https://cccqaq.github.io/PostCam.github.io/

cs.CV↗
arXiv2026

Learn Before You Judge: Progressive Knowledge-to-Decision Alignment for Explainable Hateful Meme Detection

Hateful memes spread abusive content through implicit interactions between images and text, posing serious threats to the safety of online communities. In recent years, multimodal large language models have been widely used for hateful meme detection and are increasingly adopted to generate explainable detection results. However, we find that existing explain-then-detect methods often couple explanation generation and label prediction within the same training process. This coupling causes interference between task objectives, leading to limited detection performance and even worse results than simple SFT baselines. To address these challenges, we propose ProKDA, a progressive knowledge-to-decision alignment method for explainable hateful meme detection. Inspired by the human annotation training process, ProKDA first uses an agentic background knowledge construction pipeline to obtain external knowledge related to meme understanding. It then adopts a three-stage training strategy that sequentially performs background knowledge learning, hatefulness detection learning, and hatefulness boundary alignment. Unlike prior explain-then-detect methods that jointly optimize both tasks, ProKDA focuses on a single training objective at each stage. This design reduces interference between the two tasks and progressively transforms background knowledge into robust detection decisions. Experiments on three public hateful meme benchmarks show that ProKDA achieves state-of-the-art detection performance and provides accurate, explainable, and evidence-supported decisions for hateful meme moderation. Project page: https://meizhiyuan88666.github.io/prokda.

cs.CL↗
arXiv2026

FPGN: Redefining Ultra-Fast Programmable Gate-based Neural Acceleration with Differentiable LUTs

Achieving nanosecond-scale inference latency for deep neural networks (DNNs) has become a primary architectural concern for latency-critical applications. While Field-Programmable Gate Arrays (FPGAs) offer a promising substrate for low-latency inference, conventional FPGA accelerators remain arithmetic-centric, using LUTs primarily as building blocks for numerical operators and peripheral logic. In contrast, recent LUT-native neural networks treat LUTs as learnable neurons, revealing promising theoretical potential to exploit their intrinsic logic expressivity. However, existing methods are largely confined to algorithmic optimizations, failing to translate this theoretical potential into high-performance FPGA accelerators. Specifically, their differentiable formulations do not faithfully match FPGA LUT primitives, their physically-unaware topologies compromise routability and timing closure, and their lack of automated optimization flow hinders systematic design space exploration (DSE) and efficient hardware implementation. In this paper, we propose FPGN, an end-to-end physically-aware framework that closes the gap between LUT-native learning and latency-optimized FPGA implementation. FPGN addresses these challenges through (i) a hardware-aligned differentiable formulation for training FPGA-native LUT neurons, (ii) a structured LUT-native topology with a streaming hardware architecture to improve routing locality and timing closure, and (iii) a latency-driven compiler that leverages high-fidelity analytical Quality of Results models to automate DSE and hardware generation. Experiments show that FPGN achieves up to 205$\times$ latency reduction compared to representative FPGA-based BNN accelerators and up to 30$\times$ higher LUT efficiency than prior differentiable LUT-native networks, while maintaining competitive inference accuracy.

cs.AR↗
arXiv2026

Can We Trust Video Hallucination Detectors? VidHalLoc for Evaluating the Evaluators

Video-language models and video agents can produce hallucinations that conflict with spatiotemporal evidence. Existing benchmarks mainly evaluate model hallucinations, and heterogeneous mechanisms make detector reliability difficult to compare. We introduce VidHalLoc, a benchmark that evaluates hallucination detection methods under a unified diagnostic evaluation protocol using 2,000 adversarial hallucination samples across Video Question Answering and Video Captioning tasks, spanning Ontology and Dynamic hallucination categories. To construct VidHalLoc efficiently, we introduce VideoHALO, a Harness Engineering-informed multi-agent workflow that decomposes data construction into four executable stages supported by a memory system and a communication protocol. Evaluation of fifteen methods reveals that the four dedicated detectors peak at an Overall accuracy of only 34.63%, indicating limited reliability across video hallucination types [Dataset Repository: https://huggingface.co/datasets/wesfggfd/VidHalLoc].

cs.CV↗
arXiv2026

Who Pays for Open Review? Visible Author Reputation and Its Effect on Ratings

An OpenReview bug in November 2025 broke anonymity at several conferences and prompted calls for open review, which motivate us to ask what shifting from blind to open would mean for authors. Analyzing over 18,000 reviewed submissions to ICLR 2026, split into de facto open and blind groups by arXiv preprint timing, we find that ratings rise with author reputation under both mechanisms, with a steeper slope under open review that is statistically significant, and that the open-blind difference is concentrated at the borderline ratings. The pattern holds across five reputation proxies (including institution, h-index, and citation count), three author-aggregation rules, and five definitions of the open window. A controlled simulation with five AI models as reviewers, holding the manuscript fixed and varying the author reputation, reproduces the effect. With claude-opus-5 as the reviewer, for example, rating rises by 0.5 points as the author moves from low to high reputation.

cs.DL↗