Search arXiv⌕ Search

arXiv subjects

Jun Du

Publications and source records attributed to Jun Du.

At least 19 recordsLinked to original sources
arXiv2026

Sequence-to-Sequence Refinement for EEND-VC Diarization with a Shared Encoder

End-to-end neural diarization with vector clustering (EEND-VC) separates local segmentation from global speaker assignment, leaving speaker confusion (SC) despite strong segmentation. We propose shared-encoder sequence-to-sequence (S2S) refinement, which reuses DiariZen's self-supervised learning (SSL) representations for speaker-conditioned decoding and prior-constrained assignment. We study frozen-feature refinement with WavLM and structurally pruned w2v-BERT 2.0, and extend the latter to joint training of local segmentation and refinement heads. The heads learn separate layer mixtures while their losses update one acoustic encoder. Across eight domains, frozen-feature refinement reduces macro diarization error rate (DER) from 12.95% to 12.40% with WavLM and from 12.35% to 12.08% with w2v-BERT 2.0, with relative SC reductions of 17.4% and 14.6%. It adds approximately 15% to system real-time factor. The joint system further lowers macro DER to 11.90%, improving seven of eight domains over non-joint refinement. Error decomposition separates SC-driven refinement gains from detection-driven joint-system gains.

eess.AS↗
arXiv2026

A Stable, High-Order Time-Stepping Scheme for the Drift-Diffusion Model in Modern Solar Cell Simulation

This paper presents a one-dimensional transient drift-diffusion simulator for advanced solar cells, integrating a structure-preserving finite-volume spatial discretization with Scharfetter-Gummel-type fluxes and a high-order, L-stable implicit Runge-Kutta (Radau IIA) temporal integrator. The scheme ensures local charge conservation, handles sharp material interfaces, achieves second-order spatial and fifth-order temporal convergence, and exhibits excellent computational efficiency and numerical stability. Its accuracy is verified against the classical depletion approximation in p-n junction and validated against four independent software implementations for an organic photovoltaic device under a common protocol. The framework's extensibility is demonstrated by incorporating exciton kinetics in organic solar cells, capturing multi-timescale dynamics, and by modeling mobile ions in perovskite solar cells, reproducing characteristic J-V hysteresis without empirical parameters. This work provides a robust, high-order numerical foundation for simulating coupled charge, exciton, and ion transport in next-generation photovoltaic devices.

physics.app-ph↗
arXiv2026

RMS-AQA: A Two-Stage Spatial Audio Question Answering Benchmark for Real-World Domestic Environments

Embodied assistants in domestic environments must infer what happened, where and when it occurred, and how to respond. To address this, we introduce RMS-AQA, a spatial audio question answering (SAQA) benchmark for real-world domestic environments. The benchmark features a two-stage question-answering (QA) format to comprehensively assess the ability of audio-language models (ALMs) to first ground audible sound events and subsequently perform complex spatio-temporal reasoning based on that grounding. To maximize acoustic realism, our dataset combines authentic real-world first-order Ambisonics (FOA) recordings with high-fidelity synthetic data generated using measured room impulse responses (RIRs). Furthermore, we provide a lightweight spatial plug-in that injects FOA-format data into frozen audio-language backbones. Experimental results reveal that the primary challenges stem from concurrent sources, far distance, and sim-to-real domain gap between RIR-synthesized and authentic recordings.

cs.SD↗
arXiv2026

VASPO: Self-Distillation with Privileged Value Estimation

On-policy self-distillation provides dense supervision for reasoning models via a privileged self-teacher. However, this supervision reflects local token preferences rather than the expected return of an unguided student continuation. Recent methods combine this signal with GRPO. They use GRPO for update directions and the teacher for signal scaling. However, we demonstrate a more effective integration within PPO. Specifically, we identify complementary positional dynamics: privileged information enhances a critic's value-prediction accuracy toward the response's end, whereas self-teacher guidance is strong initially, drops mid-generation, and rebounds finally. These insights motivate Value-Augmented Self-Distillation Policy Optimization (VASPO), wherein a self-teacher and critic evaluate identical privileged information at student-generated prefixes. The critic computes token-level PPO advantages from predicted continuation returns, while the self-teacher modulates selected positive advantages via bounded probability ratios. Across six mathematical reasoning benchmarks, VASPO improves mean accuracy over PPO by 2.59 and 1.81 percentage points on Qwen3-4B and OLMo3-7B variants, respectively, and achieves the highest mean accuracy among evaluated methods on Knights and Knaves. Our code is available at https://github.com/cii030/VASPO.

cs.CL↗
arXiv2026

Device Model Linking Singlet and Triplet Spin-State Properties to Organic Photovoltaic Device Characteristics

This work proposes an exciton-kinetics coupled drift-diffusion (EKDD) model that bridges device-level charge transport with singlet and triplet excited-state kinetics. State populations and reversible photophysical transitions are solved self-consistently, enabling transient and steady-state device response calculations. We investigate how LE-CT, CT-CS, and LE-T energy offsets, together with triplet-transfer kinetics, govern charge generation, recombination, voltage loss, and extraction in organic photovoltaics. We identify CT-CS offset as a primary, largely overlooked factor controlling charge generation and voltage loss. A positive offset makes CT3-CS separation exothermic and improves generation, but lowers CS energy and increases voltage loss; a negative offset makes separation endothermic and highly temperature-sensitive. The model reproduces triplet-mediated recombination in PM6:Y6 and material-dependent triplet recycling in D18:NFA blends. We propose design rules for balanced generation and voltage loss via two charge-transfer channels: the LE-CT-CS pathway favors a moderate LE-CT offset of about 0.15-0.20 eV and a weakly exothermic CT-CS offset of about 0.05-0.075 eV; the T-CT3-CS pathway requires the same CT-CS alignment plus a small CT-T offset, no more than about 0.1 eV, corresponding to a recycling factor K greater than about 10^-2, to maintain fast T-to-CT3 recycling. CT-CS alignment is the main determinant of the charge-generation-voltage-loss trade-off, while triplet-recycling kinetics determines whether this balance survives at the device level. This work provides a unified theoretical description from exciton generation to free charge collection, offering a convenient tool for transient and steady-state analysis in organic photovoltaics.

physics.app-ph↗
arXiv2026

An Embodied Multiagent Framework Based on Token Communications for Cooperative ISAC

The emerging low-altitude economy demands unmanned aerial vehicle (UAV)-enabled integrated sensing and communication (ISAC) for reliable connectivity and environmental awareness. In particular, embodied UAV agents offer a promising means of supporting autonomous operations through a closed loop linking perception, decision-making, and physical actions. However, each UAV has access only to local observations, and effective cooperation requires exchanging local states and intentions. Directly sharing such information can incur substantial signaling overhead and hinder timely coordination in dynamic environments. To deal with this problem, this paper investigates a cooperative ISAC network of embodied UAV agents and formulates a joint token communications (TokCom) and physical control problem to minimize total propulsion energy subject to communication and sensing rate requirements. Then, we propose a state--intent TokCom (SI-TokCom) framework driven by multi-agent embodied policy learning. Specifically, separate pretrained codebooks enable compact exchanges of local states and intentions, while UAV agents jointly learn to select and compose tokens and determine physical actions based on local observations and received tokens. Simulation results show that SI-TokCom achieves 98.9\% and 99.1\% of the centralized baseline's communication and sensing rates, respectively. Compared with the local baseline, it improves the corresponding rates by 5.0\% and 43.6\%, respectively, with essentially unchanged propulsion energy. These results highlight the potential of TokCom for communication-efficient cooperation among embodied UAV agents in ISAC systems.

eess.SP↗
arXiv2026

KUAISHOU Explorer LLM-Rec Challenge 2026: Reasoning Generative Recommendation

Generative recommendation, has been attracted a surge of attentions in industrial and academic research community, towards to build more smart system to build next-generation recommender. Under the significant developing wave of large language model, our team have been developed Semantic ID based OneRec/OneRec-V2. These models have been widely deployed in production and demonstrate the scaling potential of the autoregressive next-item prediction paradigm for industrial recommender systems. Building on the success of OneRec, we further explored a series of models, including OneRec-Think, OpenOneRec, and OneReason, that connect item Semantic IDs with natural language in a unified representation space and seek to unlock the potential of natural-language chain-of-thought (CoT) reasoning for recommendation. However, our preliminary works found that introducing reasoning CoT does not always improve the recommendation performance. To address this issue, OneReason strengthens the semantic alignment between items and language, introduces structured template-based supervision for interest reasoning, and applies advanced reinforcement learning techniques to make reasoning more beneficial to recommendation. As a frontier topic to building recommendation foundation models, we believe this topic has significant research value and hope to encourage more researchers to explore it together. To this end, together with the SIGIR 2026 community, we organized the KUAISHOU Explorer LLM-Rec Challenge 2026: Reasoning Generative Recommendation.

cs.IR↗
arXiv2026

Towards Training-free Automatic Proxy Discovery via Large Language Models for Mixed Precision Quantization

Mixed-Precision Quantization (MPQ) liberates Deep Neural Networks (DNNs) from the Out-Of-Memory (OOM) bottleneck and has garnered increasing research attention. However, conventional methods either rely on costly differentiable optimization search, which is neither efficient nor flexible, or learn a quantized DNN from a proxy (e.g., HAWQ) manually designed by human experts, which is labor-intensive and requires extensive expert knowledge. Can we design a proxy without involving any human experts or training? In this paper, we provide an affirmative answer by proposing a novel Large Language Model (LLM)-driven Training-free Automatic Proxy (dubbed TAP) discovery framework. It reforms the design paradigm of MPQ by utilizing LLMs and evolutionary search strategies to automatically find superior TAP tailored for MPQ. In addition, to bridge the gap between black-box LLMs and the challenging MPQ task, we introduce a lightweight Direct Preference Optimization (DPO)-based strategy controller that dynamically reweights the selection probabilities of the three prompt templates for evolutionary search strategies according to fitness signals, without fine-tuning the LLM. This forms a task-aware feedback loop that improves proxy generation across evolutions. Extensive experiments on mainstream benchmarks demonstrate that TAP achieves state-of-the-art performance. Finally, we believe that our TAP will significantly contribute to the MPQ community by providing a new perspective on LLM-driven design algorithms.

cs.LG↗
arXiv2026

PalmLeaf-VQA: A Multi-Script Visual Question Answering Benchmark for Historical Palm-Leaf Manuscript Understanding Across Diverse Regions

Historical manuscripts remain largely absent from modern vision-language benchmarks, leaving open how well multimodal large language models (MLLMs) handle culturally diverse, degraded, and non-Latin document images. We introduce \textbf{PalmLeaf-VQA}, a multi-script visual question answering benchmark for historical palm-leaf manuscript understanding across South and Southeast Asian traditions. PalmLeaf-VQA contains \textbf{923 curated manuscript images} and \textbf{7,384 question--answer pairs} from eight collection groups: Balinese, Grantha, Jathakam, Kambaramayanam, Kannada, Khmer, Sundanese, and Tamil. Unlike recognition-oriented resources, the benchmark targets manuscript-aware visual reasoning over preservation-relevant cues, including physical condition, line structure, material and coating, binding holes, margins, symbols, drawings, and localized visual artifacts. We evaluate recent proprietary and open-weight MLLMs under open-answer and constrained-answer prompting and provide fine-grained analysis across collections, question categories, and task types. The strongest evaluated model reaches only \textbf{58.00\% exact-match accuracy} on the held-out test split, revealing substantial limitations in current MLLMs for rare-script, degraded-layout, and preservation-oriented document understanding. PalmLeaf-VQA provides a standardized benchmark for advancing culturally grounded and layout-aware multimodal document analysis.

cs.CV↗
arXiv2026

PEARL: Training Socratic Tutors with Pedagogically Aligned Reinforcement Learning

Large Language Models (LLMs) show strong potential as educational tutors. Existing approaches typically train them to solve problems and provide correct answers, but this problem-solving-centered paradigm overlooks key requirements of effective tutoring: progressive guidance and the coordination of multiple pedagogical objectives across multi-turn interactions. Developing such tutors remains challenging because student behavior varies substantially with individual knowledge states, pedagogical effectiveness depends on multiple factors beyond final-answer correctness, and coordinating these objectives over tutor-student interactions is inherently difficult. To address these challenges, we propose PEARL, a PEdagogically Aligned Reinforcement Learning framework for training Socratic tutoring agents. First, we introduce a controllable student simulator that disentangles latent cognitive states from response generation, enabling simulation of diverse abilities and misconceptions. Second, we develop a pedagogically aligned reward model that jointly assesses pedagogical quality and objective correctness. Finally, we propose a stable multi-objective reinforcement learning approach that balances competing pedagogical objectives during tutor training. Experiments across multiple benchmarks show that PEARL performs competitively against tutoring-specific open-source systems and leading proprietary LLMs.

cs.LG↗
arXiv2026

The SLT 2026 SmartGlasses Challenge: Benchmarking Egocentric Multi-Talker Speech Recognition and Understanding with Audio-Language Models

Recent advances in large language models (LLMs) and multimodal LLMs (MLLMs) have created new opportunities for wearable speech interfaces, with smart glasses providing an egocentric platform for continuous audio sensing and assistance. However, speech recognition and understanding in this setting remain challenging because of dynamic acoustic conditions, speaker overlap, and the spatial ambiguity introduced by wearer-centered recording geometry. To support systematic evaluation in this setting, we introduce the IEEE SLT 2026 SmartGlasses Challenge for egocentric multi-speaker speech processing. The challenge consists of two tracks, Dyadic Dialogue Understanding and Multi-party Meeting Understanding, and jointly evaluates Time-Stamped Speaker-Attributed Automatic Speech Recognition (TSA-ASR) and Spoken Language Understanding (SLU). It is built on a 106-hour four-channel egocentric speech dataset containing 714 sessions collected in real-world scenarios. This paper describes challenge tasks, dataset construction, submissions, and summarizes the main findings from the shared evaluation. The results show that heavy speaker overlap remains a major factor affecting TSA-ASR performance, while paralinguistic acoustic understanding continues to be difficult for current audio-language models in complex SLU settings. Further details can be found on the official challenge website.

eess.AS↗
arXiv2026

ReDiPPO: Reference-Guided Value Calibration and Discrepancy-Aware Token Reweighting for Mathematical Reasoning

Reinforcement learning has emerged as an effective paradigm for enhancing the mathematical reasoning capabilities of large language models. Among existing policy optimization methods, Proximal Policy Optimization (PPO) remains particularly appealing because its learned critic can, in principle, provide token-level credit assignment. However, in mathematical reasoning tasks characterized by long reasoning horizons and sparse outcome rewards, reliable token-level credit assignment remains challenging. The standard critic often fails to accurately evaluate intermediate reasoning states, resulting in noisy advantage estimates and suboptimal policy updates. In this paper, we propose ReDiPPO, a Reference-guided and Discrepancy-aware PPO framework for mathematical reasoning. ReDiPPO introduces a reference-guided critic that uses reference answers as training-time privileged signals to provide more accurate value estimation. Meanwhile, it retains a standard critic and quantifies the token-level reference-standard discrepancy between the standard value estimate and the reference-guided value estimate. This discrepancy serves as an indicator of difficult reasoning states and is used to reweight the corresponding token-level advantages during PPO optimization. Extensive experiments on diverse mathematical reasoning benchmarks demonstrate that ReDiPPO improves value-estimation accuracy and consistently outperforms strong policy optimization baselines, including PPO, DAPO, and GSPO, in final reasoning performance. Our code is available on https://github.com/cii030/ReDiPPO.

cs.AI↗
arXiv2026

KAT-Coder-V2.5 Technical Report

We present KAT-Coder-V2.5, a coding-focused agentic model trained to act autonomously inside real, executable repositories rather than as a single-turn code generator. Its capability is bottlenecked less by model scale than by the scarcity of reproducible environments, verifiable rewards, and high-value trajectories, which we address with an end-to-end agentic post-training framework. AutoBuilder reconstructs multilingual repositories into sandboxed environments with fail-to-pass and pass-to-pass verification at scale, from which we regenerate self-contained task specifications, recover near-miss trajectories, and distill supervision through process-aware filtering, while KwaiClawEnv synthesizes large-scale tool-use trajectories from executable services and real task seeds. We further scale reinforcement learning with harness randomization, a reliability-hardened sandbox, an asymmetric actor--critic PPO with hindsight-augmented value estimation, and a harness-oriented reward framework, and unify SWE, Agent-Claw, and WebCoding experts via Multi-Teacher On-Policy Distillation. Across six software-engineering and agentic benchmarks, KAT-Coder-V2.5 delivers the best agentic tool-use result on PinchBench and ranks second only to the frontier Opus 4.8 on repository-level software engineering. Our service is available at https://streamlake.com/product/kat-coder.

cs.SE↗
arXiv2026

Resource-Aware LLM Reasoning for Mobile Edge General Intelligence

The rapid advancement of large language models (LLMs) has enabled an emergence of agentic artificial intelligence (AI) with powerful reasoning and autonomous decision-making capabilities. This integration with edge computing has led to the development of Mobile Edge General Intelligence (MEGI), which brings real-time, privacy-preserving reasoning to the network edge. However, deploying LLM-based agentic AI reasoning in MEGI environments poses significant challenges due to the high computational demands of reasoning and the limited resources of edge devices. To address these challenges, we propose a joint optimization framework for efficient LLM reasoning deployment in MEGI. First, we systematically review enhancement methods to identify mechanisms suitable for edge adaptation. Subsequently, we present a distributed framework that synergizes reasoning enhancement via adaptive CoT prompting with scalable deployment through a distributed MoE architecture. An important innovation of this approach involves modeling reasoning depth as a dynamic network resource variable, which is optimized jointly with expert activation and transmission power. This mechanism allows the system to dynamically regulate expert networks and reasoning complexity according to task requirements and device capabilities. Experimental evaluations in mobile edge environments demonstrate that the proposed framework effectively balances reasoning quality and resource efficiency. The results show that with less than one second of additional inference time, both accuracy and latency satisfaction rate can reach 90\%, validating the practical viability of deploying sophisticated LLM reasoning in resource-constrained MEGI systems.

cs.AI↗
arXiv2026

EmoInstruct-TTS: Dual-Path Instruction-Guided Emotional Speech Synthesis

Instruction-based controllable speech synthesis enables users to specify emotions through natural language. However, existing approaches often rely on coarse emotion labels and lack explicit modeling of fine-grained intensity. We propose EmoInstruct-TTS, a dual-path instruction-guided framework for emotional speech synthesis. We introduce Emotion2embed, a supervised semantic-acoustic emotion embedding covering 48 emotional states, including fine-grained categories and intensity levels. To infer embeddings from free-form instructions, we design an Instruction-Conditioned Emotion Flow Model (ICE-Flow) that generates acoustically grounded emotion representations. The inferred embeddings are integrated into an LLM-based synthesis pipeline to provide explicit emotional control while preserving semantic planning. Experiments show improved emotional controllability and speech naturalness over strong baselines.

cs.CL↗
arXiv2026

YARD: Y-Architecture Register Decoding for Efficient Hallucination Mitigation in Large Vision-Language Models

Contrastive decoding (CD) seeks to mitigate hallucinations in Large Vision-Language Models (LVLMs) by contrasting the output distributions of a standard model and a visually degraded model. However, existing training-free CD methods suffer from sub-optimal degraded branches: completely dropping visual tokens is too extreme and induces language hallucinations, while corrupting input images offers coarse control over visual evidence and suffers from high inference latency due to requiring two full forward passes. To address these dilemmas, we propose YARD, a training-free Y-Architecture Register Decoding framework. Motivated by the observation that reliable text-to-vision grounding predominantly emerges in the middle decoder layers, YARD constructs the degraded branch internally by sharing shallow-layer computations and branching exactly at this critical stage. For the degraded branch, YARD replaces patch-level visual tokens with register tokens, which preserve global image semantics but lack fine-grained local evidence. This image-aware yet locally under-grounded design provides a faithful contrastive signal without extreme modality mismatch, while the Y-architecture strictly avoids a costly second forward pass. Extensive experiments on generative and discriminative hallucination benchmarks demonstrate that YARD consistently achieves state-of-the-art hallucination mitigation across multiple LVLMs, alongside a significant reduction in inference latency.

cs.CV↗
arXiv2026

Antiferromagnetic Ordering Enhanced Magnetic Damping in Mn2Au/CoFeB Bilayers

Antiferromagnets (AFMs) hold significant potential for spintronic devices owing to their insensitivity to external magnetic fields and the absence of stray fields. Beyond these inherent advantages, an AFM can manipulate the magnetic dynamics of a ferromagnet (FM) layer in AFM/FM bilayers, whereas the mechanism of such manipulation remains controversial. Here, we investigate the magnetic dynamics of AFM/FM Mn2Au/CoFeB bilayers via Ferromagnetic Resonance (FMR). It is found that the Néel temperature of 2-nm-thick Mn2Au is as low as ~40 K, in sharp contrast to that of bulk Mn2Au, which exceeds 1000 K. In the Mn2Au(2 nm)/CoFeB(4 nm) bilayer, the magnetic damping $α$ of the CoFeB layer increases from 0.013 to 0.047 as temperature decreases from 160 K to 10 K, accompanied by a synchronous increase in the exchange coupling field H_rot. Such an increase in $α$ is attributed to the enhanced spin angular momentum transfer from CoFeB to Mn2Au, mediated through AFM-FM exchange coupling between Mn2Au and CoFeB, which is enhanced by the Mn2Au antiferromagnetic ordering as the temperature decreases. Our study provides deeper insights into AFM/FM dynamics and spintronic storage technology.

cond-mat.mtrl-sci↗
arXiv2026

Unlocking Complex Visual Generation via Closed-Loop Verified Reasoning

Despite rapid advancements, current text-to-image (T2I) models predominantly rely on a single-step generation paradigm, which struggles with complex semantics and faces diminishing returns from parameter scaling. While recent multi-step reasoning approaches show promise, they are hindered by ungrounded planning hallucinations lacking verification, monolithic post-hoc reflection, long-context optimization instabilities, and prohibitive inference latency. To overcome these bottlenecks, we propose the Closed-Loop Visual Reasoning (CLVR) framework, a comprehensive system that deeply couples visual-language logical planning with pixel-level diffusion generation. CLVR introduces an automated data engine with step-level visual verification to synthesize reliable reasoning trajectories, and proposes Proxy Prompt Reinforcement Learning (PPRL) to resolve long-context optimization instabilities by distilling interleaved multimodal histories into explicit reward signals for accurate causal attribution. Furthermore, to mitigate the severe latency bottleneck caused by iterative denoising, we propose $Δ$-Space Weight Merge (DSWM), a theoretically grounded method that fuses alignment weights with off-the-shelf distillation priors, reducing the per-step inference cost to just 4 NFEs without requiring expensive re-distillation. Extensive experiments demonstrate that CLVR outperforms existing open-source baselines across multiple benchmarks and approaches the performance of proprietary commercial models, unlocking general test-time scaling capabilities for complex visual generation.

cs.CV↗