arXiv subjects

Li Li

Publications and source records attributed to Li Li.

At least 19 recordsLinked to original sources

arXiv2026

DM-ASR: Diarization-aware Multi-speaker ASR with Unified Offline and Online Decoding with Large Language Models

Multi-speaker automatic speech recognition (ASR) aims to recognize not only what was said, but also who said it and when. Recent Speech-LLMs have shown promising results in structured multi-speaker transcription, but jointly modeling speaker attribution, temporal information, and lexical content remains challenging and often requires substantial model capacity and training data. To address this challenge, we propose DM-ASR, a Diarization-aware Multi-speaker ASR framework that reformulates multi-speaker recognition as diarization-conditioned multi-turn dialogue generation. By leveraging diarization-derived speaker-time cues as explicit structural priors, DM-ASR decomposes transcription into a sequence of conditioned queries while preserving mixed acoustic information and conversational context. We further introduce optional word-level timestamp prediction for fine-grained temporal grounding and investigate the complementary roles of diarization cues and language-model predictions. The proposed formulation also enables unified offline and causal sentence-level online decoding through flexible acoustic context and dialogue history management. Experiments on Mandarin and English benchmarks demonstrate competitive structured transcription performance with relatively compact models and training data, while effectively supporting both offline and online recognition.

eess.AS↗

arXiv2026

IGDA-ICMH: an Importance Guided Feature and Network Dual Adaptation Framework towards Image Compression for Machine and Human Vision

With the surge of machine vision applications, image compression for both machine and human vision (ICMH) has become an important research area. Existing studies mainly focus on task-specific adaptation or multi-task joint optimization, yet they often overlook the unequal contributions of different features and network layers to downstream machine vision tasks. To address this issue, we first quantitatively investigate the impact of different feature and network layers on machine vision performance, providing a principled understanding of their importance. Based on this analysis, an Importance Guided feature and network Dual Adaptation (IGDA) framework is developed for ICMH. It adopts implicit neural representation as the basic architecture and designs an efficient adaptation structure with both feature and network weight adaptation. An Importance-Guided Network Adaptation (IGNA) module is developed with a low-rank adaptation strategy to selectively adjust key network layers according to their importance in different tasks. An Importance-Guided Feature Adaptation (IGFA) module is developed with a low-rank and Kronecker product-based transform to explore the non-local property of the features in order to reduce the number of adaptation parameters. Extensive experiments on representative machine vision tasks, including image classification and object detection, demonstrate that our IGDA-ICMH framework achieves significantly better performance than the existing implicit neural representation-based methods for ICMH.

eess.IV↗

arXiv2026

Hardware-Aware Conditional Code in Android Apps

Android OS runs on a highly diverse set of devices, spanning entry-level phones and flagships with substantially different hardware characteristics, including memory capacity, processors, displays, cameras, and sensors. Through APIs, apps can query device information at runtime, such as memory, hardware features, and manufacturer, among other device properties, allowing developers to execute different code paths on different devices. Existing studies cover only specific compatibility bugs or brand-specific behavior; however, the broader use of hardware information in app logic has not been comprehensively explored. We address this question with a large-scale empirical study of hardware-related if statements in Android apps. We statically analyze 50,000 apps to identify the if statements whose conditions depend on the values of hardware-related APIs, and extract their conditions together with the code of their branches. Our results show that 95% of successfully analyzed apps contain at least one such statement, with an average of 74 per app, which suggests that hardware-related logic is widespread. We then apply an LLM-assisted open-coding process to the extracted code to build a structured taxonomy of the purposes behind hardware-related conditional code. The resulting characterization shows that developers use hardware information for feature availability checks, UI adaptation, emulator detection, device tiering, and resource management, among others.

cs.SE↗

arXiv2026

Real-Time Neural Video Compression with Unified Intra and Inter Coding

Neural video compression (NVC) technologies have advanced rapidly in recent years, yielding state-of-the-art schemes such as DCVC-RT that offer superior compression efficiency to H.266/VVC and real-time encoding/decoding capabilities. Nonetheless, existing NVC schemes have several limitations, including inefficiency in dealing with disocclusion and new content, interframe error propagation and accumulation, among others. To eliminate these limitations, we borrow the idea from classic video coding schemes, which allow intra coding within inter-coded frames. With the intra coding tool enabled, disocclusion and new content are properly handled, and interframe error propagation is naturally intercepted without the need for manual refresh mechanisms. We present an NVC framework with unified intra and inter coding, where every frame is processed by a single model that is trained to perform intra/inter coding adaptively. Moreover, we propose a simultaneous two-frame compression design to exploit interframe redundancy not only forwardly but also backwardly. Experimental results show that our scheme outperforms DCVC-RT by an average of 12.1% BD-rate reduction, delivers more stable bitrate and quality per frame, and retains real-time encoding/decoding performances. Code and models will be released.

cs.CV↗

arXiv2026

Emergent Scaling Symmetry and Universal Interior Criticality of Scalarized Black Holes

The interiors of spontaneously scalarized charged black holes exhibit a tower of universality classes, including the previously reported $1/2$ scaling. We show that their near-critical collapse is governed by an emergent scaling symmetry at the would-be Cauchy horizon. This symmetry renders the interior dynamics autonomous, model-independent, and exactly integrable, realizing first-kind self-similarity rather than Choptuik-type second-kind criticality. It also yields a ``seesaw amplification'': an arbitrarily small exterior hair produces an arbitrarily large interior response. The amplification obeys an amplitude identity relating three independently measurable quantities. The singularity exponent equals the critical exponent of the exterior scalar charge. These relations connect the exterior phase transition to the approach to the interior singularity through the scaling dynamics near the would-be Cauchy horizon.

hep-th↗

arXiv2026

PixelMoE: Scaling Pixel-Space Diffusion with LLM-Style Mixture-of-Experts

Pixel-space diffusion models provide an appealing alternative to latent-space diffusion models by directly modeling raw images without relying on a learned tokenizer. However, scaling pixel-space models has predominantly followed dense strategies, tightly coupling model capacity with active computation. In this work, we show that standard LLM-style Mixture-of-Experts (MoE) architectures can be effectively transferred to pixel-space diffusion. Building on this finding, we introduce PixelMoE, a pixel-space generative framework that sparsely scales model capacity with standard LLM-style MoE, without diffusion-specific routing mechanisms or auxiliary routing objectives. Specifically, PixelMoE combines standard Token-Choice (TC) routing with fine-grained routed experts for input-adaptive sparse computation, a shared expert for capturing common representations, and a bias-based load-balancing strategy. To complement sparse capacity scaling, PixelMoE further introduces Pixel-Space Hierarchical Flow Matching, which varies spatial resolution along the denoising trajectory from global structure formation to fine-grained detail refinement. At comparable active parameter counts, PixelMoE consistently outperforms its dense counterparts across model scales, demonstrating effective sparse capacity scaling in pixel space. On class-conditional ImageNet-1K, PixelMoE achieves FID scores of 1.47 at 256x256 and 1.53 at 512x512 without representation alignment or perceptual losses, comparing favorably with leading pixel-space and latent-space generative models. Code and pretrained models are available at https://github.com/KlingAIResearch/PixelMoE and https://huggingface.co/KlingTeam/PixelMoE, respectively.

cs.CV↗

arXiv2026

S2SND-v2: Unified Online and Offline Speaker Diarization with Speaker-Adapted w2v-BERT 2.0

This paper presents S2SND-v2, a unified online and offline speaker diarization framework built on S2SND that improves acoustic feature extraction and the use of the resulting representations. For acoustic feature extraction, we introduce a speaker-adapted w2v-BERT 2.0 front end and compare full and pruned construction routes. To better use these features, overlap-aware activity-query masking improves speaker embedding extraction under overlap, while late-speaker training aligns unknown-speaker training with online inference conditions. Combining these improvements, the pruned system achieves competitive average diarization error rates of 14.52% online and 12.03% offline across eight test sets.

cs.SD↗

arXiv2026

AutoTailor: Automatic and Efficient Adaptive Model Deployment for Diverse Edge Devices

On-device machine learning (ML) has become a fundamental component of emerging mobile applications. Adaptive model deployment delivers efficient inference for heterogeneous device capabilities and performance requirements through customizing neural architectures. SuperNet-based approaches offer a promising solution by generating numerous model variants from a pre-trained ML model. However, applying SuperNet in existing frameworks requires tedious model-aware development and time-consuming hardware-aware profiling, which limits their practical adoption. We present AutoTailor, the first framework to enable automated, end-to-end SuperNet-based adaptive model deployment for edge devices. Unlike manual SuperNet construction, AutoTailor employs a computation graph-guided compilation approach to automatically transform user-provided ML models into SuperNets. To support efficient specialization, AutoTailor incorporates learning-free latency and accuracy predictors, enabling low-cost yet accurate performance prediction. Our extensive evaluations demonstrate that AutoTailor reduces the lines of code for SuperNet construction by 10.8--34.5$\times$, reduces predictor profiling time by 2.2--11.5$\times$, and achieves up to 15.60\% absolute accuracy improvement and 60.03\% latency reduction compared to state-of-the-art approaches across diverse models and devices. AutoTailor's source code is publicly available at \url{https://github.com/liu-mengyang/AutoTailor}.

cs.LG↗

arXiv2026

The fractional Calderón problem on finite Cayley graphs

We study a Calderón-type inverse problem for the spectral fractional Laplacian on finite Cayley graphs. For a known self-adjoint operator, recovery from a single measurement reduces to two requirements: full rank of the boundary (exterior)-to-interior coupling and nonvanishing of the interior solution. We verify both requirements on Cartesian products of arbitrary finite groups with even cycles, using consecutive cyclic halves as the accessible and inaccessible sets. Fractionalization increases the coupling rank, and one fixed nonnegative nonzero input determines every nonnegative potential. Our argument relies on a discrete maximum principle and an explicit resolvent formula.

math.AP↗

arXiv2026

Experimental Realization of a C-Shunt Flux-Qubit Coupler for Controlled-Z Gates between Transmons

Fast entangling gates and low-interaction idling require distinct operating conditions in superconducting quantum processors. We experimentally realize a transmon-C-shunt-transmon (TCT) device in which a flux-tunable, three-junction capacitively shunted flux-qubit coupler connects two transmon data qubits. The C-shunt coupler provides a flux-tunable, positively anharmonic spectrum for tailoring the interactions between the data qubits. Conditional Ramsey and swap measurements identify an idle region with near-zero residual ZZ interaction strength and suppressed transverse exchange. At this bias, simultaneous single-qubit interleaved randomized benchmarking yields X-gate fldelities above 99.7%, consistent with individual operation. A 45 ns nonadiabatic controlled-Z (CZ) gate achieves an interleaved randomized benchmarking fldelity estimate of 99.23(15)%. Numerical simulations identify coupler energy relaxation as an important limitation on CZ-gate fldelity and indicate that extending the coupler lifetime can improve gate performance. These results demonstrate the experimental feasibility of C-shunt flux-qubit couplers for combining low-interaction idling with fast entangling operations between transmon qubits.

quant-ph↗

arXiv2026

An LLM-powered Agentic Recommendation System for Connected TV Content Discovery

Recommendation systems, from traditional multi-stage to recent unified generative architectures, face challenges in incorporating diverse contextual signals, such as trending topics, breaking news, cultural events, and cross-surface user activities, into their ranking pipelines. These systems are designed to consume structured behavioral signals with consistent schemas, and lack the reasoning capability to naturally process unstructured or heterogeneously formatted contextual information. Incorporating such signals typically requires feature engineering, bespoke data pipelines, and carefully tuned heuristics. In this paper, we present an LLM-powered agentic recommendation system designed for Connected TV (CTV) content discovery that addresses these limitations. Our system leverages the reasoning capabilities of large language models to naturally process and synthesize diverse signals across varying schemas and structures, eliminating much of the manual integration inherent in traditional ranking and retrieval systems. Recognizing that current LLM-based solutions still fall short of traditional machine learning models in several recommendation tasks, including retrieval efficiency, personalization precision, and scalability, we adopt an agentic architecture that orchestrates specialized components, allowing each sub-task to be handled by the most suitable method, whether LLM-based or traditional ML. The main contribution of this work is our engineering approach to successfully overcoming the practical limitations of enabling LLM for recommendation, particularly inference latency. We share insights from our work and discuss the trade-offs and lessons learned in building a hybrid system that combines the flexibility of LLMs with the performance of established recommendation techniques.

cs.IR↗

arXiv2026

Decision Uncertainty-Oriented Watermarking (DUOW): Securing Multi-Agent Routing Against Policy Cloning

Routing policies in Multi-Agent Systems (MAS) represent valuable intellectual property, yet their exposure through execution logs creates significant risks of unauthorized policy cloning. Existing agent watermarking methods protect intellectual property by modifying internal generation processes, such as perturbing token sampling distributions or injecting structural mutations into action trajectories, which may compromise task fidelity or suffer from limited transferability across heterogeneous platforms. To address these limitations, we propose Decision Uncertainty-Oriented Watermarking (DUOW), a lightweight framework that operates at the external decision layer without modifying model parameters or generated outputs. DUOW introduces a plug-and-play scoring mechanism that evaluates candidate agents from semantic and operational perspectives, identifies functionally redundant routing decisions, and selectively redirects choices among equivalent alternatives to embed ownership signals. By coupling watermark behaviors with utility-preserving routing decisions, DUOW enables cloned policies to retain ownership signals during policy imitation while increasing the difficulty of watermark removal. A confidence-driven adaptive threshold dynamically determines embedding opportunities, balancing watermark capacity and task fidelity under diverse routing contexts. Extensive evaluations on four multi-agent routing benchmarks demonstrate that DUOW preserves high task utility, achieves reliable ownership verification under up to 90% log erasure, and introduces negligible computational overhead, with an average per-step latency of 0.043 ms and total embedding overhead below 0.54 ms.

cs.CR↗

arXiv2026

Unified Generative Video Compression with Autoregressive and Bidirectional Video Diffusion

Video diffusion models can reconstruct visually compelling content from severely compressed representations at ultra-low bitrates, but they impose a rigid deployment trade-off. Bidirectional (BI) models achieve high quality through dense full-sequence attention, yet require the entire sequence before reconstruction and incur substantial latency. Autoregressive (AR) models enable low-delay streaming, but their restricted temporal context reduces compression efficiency. This tension is particularly relevant when a video stream must be decoded in real time yet is also stored for later, quality-oriented playback, making AR and BI models natural choices for the live and replay stages, respectively. To support both stages, we present UniGVC, a Unified Generative Video Codec that supports two decoding modes from a single bitstream. UniGVC-BI employs a BI-DiT with dense attention for quality-oriented random-access decoding, whereas UniGVC-AR uses windowed AR-DiT for low-delay streaming. Both paths share a hierarchical latent codec that recurrently produces the same conditions with ultra-low bitrates, allowing UniGVC-BI to serve as a conditional teacher for single-step distillation of UniGVC-AR. A tiny conditional video decoder further reduces streaming latency while preserving most of the reconstruction quality. On standard 1080p benchmarks, UniGVC-BI achieves the best overall rate-distortion-perception trade-off among the evaluated methods and produces temporally stable video at rates as low as 0.0005 bits per pixel. Meanwhile, UniGVC-AR remains competitive while decoding 1080p video at 36.85 FPS, over $250\times$ faster than UniGVC-BI. Code and models will be released later.

cs.CV↗

arXiv2026

Beyond the Embedding Bottleneck: Adaptive Retrieval-Augmented 3D CT Report Generation

Automated radiology report generation from 3D CT volumes often suffers from incomplete pathology coverage. We provide empirical evidence that this limitation stems from a representational bottleneck: contrastive 3D CT embeddings encode discriminative pathology signals, yet exhibit severe dimensional concentration, with as few as 2 effective dimensions out of 512. Corroborating this, scaling the language model yields no measurable improvement, suggesting that the bottleneck lies in the visual representation rather than the generator. This bottleneck limits both generation and retrieval; naive static retrieval fails to improve clinical efficacy and can even degrade performance. We propose \textbf{AdaRAG-CT}, an adaptive augmentation framework that compensates for this visual bottleneck by introducing supplementary textual information through controlled retrieval and selectively integrating it during generation. On the CT-RATE benchmark, AdaRAG-CT achieves state-of-the-art clinical efficacy, improving Clinical F1 from 0.420 (CT-Agent) to 0.480 (+6 points); ablation studies confirm that both the retrieval and generation components contribute to the improvement. Code is available at https://github.com/renjie-liang/Adaptive-RAG-for-3DCT-Report-Generation.

cs.CV↗

arXiv2026

Auditable Claims about AI Agents

Organizations make claims about their AI agents: a person approves every external email, every action is logged, an evaluation shows the agent is safe to deploy. Article 12 of the EU AI Act requires high-risk systems to allow the automatic recording of events but does not say which records settle a given claim. The position is one sentence: to be checked, a claim about an agent must first name its policy, its scope, the records that would settle it, and who writes them. Adapting the preconditions of an assurance engagement, we call a claim auditable when these elements and a decision rule are fixed before any verdict and the records are obtainable. This extends the Policy Checkability dimension of our Auditable Agents framework from single actions to claims. Agents add three conditions: coverage by an independent record, authorization bound to each action's arguments, and completeness beyond integrity. Under an explicit model, we prove that support is impossible without each wherever its hypotheses hold. A claim-check table applies the method to six common claims, anchored in current NIST, IETF, and OWASP drafts. A worked case follows one claim through five evidence states. We close with a practice box and steps for operators, buyers, auditors, and standard setters.

cs.AI↗

arXiv2026

When and What to Prune? Stage-Aware Visual Token Pruning for Efficient VLA

Visual token pruning is an effective way to accelerate vision-language models and is especially useful for vision-language-action (VLA) inference, where many visual tokens must be processed before predicting robot actions. Existing pruning methods usually estimate which tokens can be pruned based on attention scores or feature diversity, retaining tokens that are either highly attended or visually different from others. However, most of them use fixed pruning schedules, such as pruning once at a preset layer or pruning at uniformly spaced layers. Such schedules can be risky for VLA models, because the model may not know which visual regions matter for the action in early layers. Tokens that look unimportant at first may become useful after the model combines visual observations with the language instruction. In this work, we propose SAPrune, a training-free visual token pruning framework for efficient VLA inference. Instead of pruning at fixed layers, SAPrune uses a small calibration set to observe how action-to-visual attention changes across layers, and chooses pruning layers only after the attention pattern becomes more reliable. At each selected layer, SAPrune applies a dual-path pruning rule: one path protects strongly attended visual tokens from pruning, while the other prevents useful surrounding context from being discarded. Experiments on LIBERO, SIMPLER, and real-world robotic tasks show that SAPrune prunes 87.5% of visual tokens and achieves up to 1.718x inference speedup while maintaining competitive task success rates.

cs.CV↗

arXiv2026

Source-Free Detection and Impact Analysis of Compiler Optimization Problems in Mobile Applications

Mobile apps frequently suffer from frame drops, overheating, and excessive power consumption. While developers optimize algorithms and debug code, a critical bottleneck often goes unnoticed: native libraries compiled with low optimization levels (O0/O1 instead of O2/O3). Because these libraries execute without functional errors, the resulting performance degradation remains hidden in production apps. We present \textsc{OptDetect}, a source-free framework that detects compiler optimization problems directly from app binaries. \textsc{OptDetect} handles mixed optimization levels through binary disassembly, chunk-level classification, and weighted score aggregation, achieving 93.0\% accuracy on controlled datasets and 81.9\% on real-world datasets. Applying \textsc{OptDetect} to 21,972 native libraries from 830 top-ranked Google Play apps, we find that 30.5\% of libraries use low optimization levels, affecting 91.7\% of apps. Through case studies on 12 production apps, fixing detected issues reduces CPU instructions by 10-63\% (median: 20.5\%) for commercial apps and 15-58\% (median: 32\%) for open-source apps. Performance complaints decrease in 5 of 6 commercial apps, and ratings increase in 5 of 6. Further investigation reveals that widely-used third-party libraries are themselves distributed at low optimization levels, with 49.7\% of 1,073 libraries in a major repository exhibiting this problem. These findings show that compiler optimization problems are common, source-free detectable, and practically consequential in mobile app ecosystems.

cs.SE↗

arXiv2026

Cloud-native and Distributed Systems for Efficient and Scalable Large Language Models -- A Research Agenda

The rapid rise of Large Language Models (LLMs) has revolutionized various artificial intelligence (AI) applications, from natural language processing to code generation. However, the computational demands of these models, particularly in training and inference, present significant challenges. Traditional systems are often unable to meet these requirements, necessitating the integration of cloud-native and distributed architectures. This paper explores the role of cloud platforms and distributed systems in supporting the scalability, efficiency, and optimization of LLMs. We discuss the complexities of LLM deployment, including data management, resource optimization, and the need for microservices, autoscaling, and hybrid cloud-edge solutions. Additionally, we examine emerging research trends, such as serverless inference, quantum computing, and federated learning, and their potential to drive the next phase of LLM innovation. The paper concludes with a roadmap for future developments, emphasizing the need for continued research, standardization, and cross-sector collaboration to sustain the growth of LLMs in both research and enterprise applications.

cs.DC↗