arXiv subjects

Meng Chen

Publications and source records attributed to Meng Chen.

At least 19 recordsLinked to original sources

arXiv2026

What Helps Denoising Can Hurt Memory: Rethinking History in Long Video Generation

Autoregressive video generators produce long videos block by block, using historical context to maintain temporal consistency. After denoising, a committing pass encodes each clean block into keys and values (KV) that later blocks read, and the same clean block can be encoded into different KV depending on the history it reads. We therefore study historical context not only as a condition for denoising but also as an input to committing. Older historical context that supports denoising can harm subsequent generation when read during committing. At fixed context size and temporal positions, older historical context lowers mean clip-to-clip Subject Consistency. Across 5 generator configurations supporting longer contexts, consistency rises, plateaus, and then declines as committing context expands. Motivated by these findings, we propose Commit Forcing, a training-free rule that organizes committing into a chain. During committing, each block reads only the preceding block's stored KV, so older historical context reaches new entries only through the chain. We apply this rule to 6 pretrained generators. On 60-second VBench, Commit Forcing improves Quality and Total on all 6 generators, by up to 1.27 Quality points and 1.10 Total points. Long-range consistency gains extend to 240-second videos. In a blinded comparison on 120-second LongLive videos, Commit Forcing achieves win rates of 83.5% for subject and scene consistency and 82.9% for overall preference against the base generator, excluding ties.

cs.CV↗

arXiv2026

Sharp Noether-type inequalities for $3$-folds of intermediate Kodaira dimensions

For a smooth projective $3$-fold $V$ of intermediate Kodaira dimension, we establish Noether-type inequalities relating the Iitaka volume $\operatorname{Ivol}(V)$ and the geometric genus $p_g(V)$. Specifically, for a smooth projective $3$-fold $V$ of Kodaira dimension $2$, we show that $\operatorname{Ivol}(V)\geq \frac{3}{4}p_g(V)-\frac{5}{4};$ while if the Kodaira dimension of $V$ is $1$, then $\operatorname{Ivol}(V)\geq p_g(V)-1.$ Both inequalities are sharp. In the Kodaira dimension $2$ case, we prove that the equality holds only if the bicanonical system of $V$ induces the Iitaka fibration and the canonical system of $V$ is composed with a pencil of surfaces whose general member has Iitaka volume $\frac12$ provided that $p_g(V)\geq 4$.

math.AG↗

arXiv2026

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↗

arXiv2026

LLM-Powered Socially Assistive Robot-Delivered Cognitive Behavioral Therapy Exercises: an Exploratory Study with University Students

Mental health is a significant healthcare challenge, and cognitive behavioral therapy (CBT) is a widely used therapeutic method for treating anxiety and depression. However, traditional CBT often requires access to trained clinicians and can be cost-prohibitive or logistically difficult for many individuals. To address these barriers, we developed a low-cost socially assistive robot (SAR) that uses a large language model (LLM) to guide the user through interactive at-home CBT exercises. In this exploratory study, 38 university students completed CBT exercises across a 15-day period using one of three modalities: with a robot (using an LLM for dialogue), a chatbot (using the same LLM for dialogue), or traditional CBT worksheets. We measured weekly therapeutic outcomes, changes in pre-/post-session anxiety measures, and adherence to completing CBT exercises. Our findings indicate that self-reported general psychological distress significantly decreased over the study period in the robot and worksheet conditions but not in the chatbot condition. Additionally, the SAR enabled significant single-session improvements on more days than the other two conditions combined. Mixed-effects modeling further suggested that the robot and chatbot conditions better reduced post-session anxiety for those with elevated levels of anxiety. Our findings suggest that SAR-guided, LLM-powered CBT may be an effective method for supporting therapeutic progress and decreasing user anxiety immediately after completing the CBT exercise. The findings underscore the potential for combining AI-driven personalization with socially assistive robotics to create accessible, scalable, and engaging mental health interventions.

cs.RO↗

arXiv2026

Accelerating Dense LLMs via L0-regularized Mixture-of-Experts

Large language models (LLMs) achieve strong performance but suffer from slow and costly inference. Existing acceleration methods often lead to noticeable performance degradation, while Mixture-of-Experts (MoE) models require extensive computational resources. In this paper, we propose L0-MoE, a lightweight MoE approach using L0-regularization to accelerate dense LLMs nearly without performance loss. Our method introduces a cluster confusion matrix for domain-aware dataset curation and applies dynamic batching for efficient training. Experiments show that L0-MoE achieves up to 2.5x speedup over dense models while maintaining competitive performance, outperforming existing LLM acceleration baselines.

cs.AI↗

arXiv2026

POSPAN: Position-Constrained Span Masking for Language Model Pre-training

Span-level masked language modeling (MLM) has shown to be advantageous to pre-trained language models over the original single-token MLM, as entities/phrases and their dependencies are critical to language understanding. Previous works only consider span length with some discrete distributions, while the dependencies among spans are ignored, i.e., assuming that the positions of masked spans are uniformly distributed. In this paper, we present POSPAN, a general framework to allow diverse position-constrained span masking strategies via the combination of span length distribution and position constraint distribution, which unifies all existing span-level masking methods. To verify the effectiveness of POSPAN in pre-training, we evaluate it on the datasets from several NLU benchmarks. Experimental results indicate that the position constraint is capable of enhancing span-level masking broadly, and our best POSPAN setting consistently outperforms its span-length-only counterparts and vanilla MLM. We also conduct theoretical analysis for the position constraint in masked language models to shed light on the reason why POSPAN works well, demonstrating the rationality and necessity of POSPAN.

cs.LG↗

arXiv2026

DigitalCoach: Communication and Grounding Gaps in Human and Agentic Computer Use Coaching

Agents are increasingly capable of automating software tasks, but can they teach humans how to use software themselves? We introduce DigitalCoach, a multimodal dataset of 72 human expert-novice computer use coaching sessions consisting of 22,752 dialogue turns grounded in 28.1 hours of screen and input event recordings across five software applications. We use DigitalCoach to evaluate whether state-of-the-art models can teach humans how to use computers. Automated evaluation shows that models differ from humans in how they coach: models provide more direct instructions, but fewer explanations, error diagnoses, and knowledge-check questions. When we fix the coaching method, models produce utterances similar to human references yet poorly grounded in visual context. Interactive evaluation confirms that model coaches cause learners to passively follow instructions without deeper engagement and fall short in visual grounding. DigitalCoach lays a foundation for collaborative and proactive computer use coaching agents. Data and code are available at https://project-digital-coach.vercel.app.

cs.CL↗

arXiv2026

The Shrinking Landscape of Linguistic Diversity in the Age of Large Language Models

Language is far more than a communication tool; it encodes a wealth of information about a person's identity, psychological state, and social context, providing valuable insights for diverse fields including psychology, marketing, and healthcare. Across three studies spanning seven datasets in different domains and over 880,000 texts, we show that the widespread adoption of large language models (LLMs) as writing assistants is linked to declines in linguistic diversity, interfering with the societal and psychological insights language provides. While core content is retained when LLMs polish and rewrite texts, LLMs also homogenize writing styles, reducing writing-complexity variance by a statistically significant 21-50% across datasets and models (p <= .05), and amplify patterns associated with dominant characteristics while suppressing others, emphasizing conformity over individuality. These trends hold across different LLMs, prompts, and contexts, with potential implications for diagnostic processes, personalization efforts, hiring assessments, and cultural preservation.

cs.CL↗

arXiv2026

Transport-defined photodetection through electrically selectable nonequilibrium carrier transport

Broadband optical fields can change in both intensity and spectral distribution, but a fixed-response detector maps this evolving information onto a single electrical signal. Here we demonstrate transport-defined photodetection, in which electrical bias selects how photoexcited carriers are redistributed, escape and are collected, creating complementary response functions within one shared active region. In a GaAs/AlGaAs semiconductor ratchet, light-driven ratchet transport defines the response at 0 V, whereas the spectral evolution at -2 V is consistent with field-assisted hot-carrier transport. The overlapping states span a measured spectral range of 0.4-94.5 μm at 5 K, provide state-dependent calculated detection floors and support joint infrared operation at 30 K. Direct optical beat notes at 7.432 GHz in the mid-infrared and 17.103 GHz in the terahertz demonstrate optical-to-electrical conversion. Their common-path outputs recover an imposed spatial temperature gradient and the transient field-of-view-integrated effective radiation temperature of laser-excited graphite. These results establish post-photoexcitation transport as a function-defining design variable for semiconductor photodetectors, complementing structure-defined, field-tuned and optically encoded approaches to reconfigurable photodetection.

physics.optics↗

arXiv2026

Who's Keeping Score? Interactive Steering of LLM-Powered Scoring with Attune

Large language models (LLMs) are increasingly used to score text records at scale (e.g., rating candidate resumes on a 1-5 scale). However, existing LLM-powered approaches do not account for the fact that effective scoring requires both holistic understanding of records and locally consistent judgments across similar ones. We present Attune, a mixed-initiative system for steerable LLM-powered scoring. Given a task description and scoring range, Attune performs pairwise comparisons across records to develop a global understanding first, and then resolves these comparisons into consistent score assignments-deriving scoring criteria and rules bottom-up in the process. These serve as shared representations of scoring logic that users can inspect and edit. Based on insights from a formative study (n = 12), Attune's interface introduces novel steering interactions that allow users to deterministically refine scoring logic. Users can provide examples, directly edit criteria, rules, or target distributions, and give natural language feedback-with all refinements compiling into constraints that guide re-scoring. We validate our approach through a technical evaluation across three workloads and a user study with domain experts (n = 8) in healthcare, law, education, and AI evaluation.

cs.HC↗

arXiv2026

Coordinated Primary Frequency Regulation and Grid-Forming Control for Wind Turbine Generators

Conventional grid-forming (GFM) control strategies often treat the DC source as an unconstrained link, creating mismatches when applied to the wind turbine generators (WTGs). Focusing on primary frequency regulation, this paper systematically investigates the mismatch between the GFM-WTGs behavior and the droop-based primary frequency regulation. To address this issue, a novel coordination strategy between WTG primary frequency regulation and GFM control is proposed. By establishing well-designed relationships among the power-tracking coefficient, power set-point, and frequency deviation, the proposed strategy enables GFM-WTGs to participate consistently in primary frequency regulation within predefined frequency limits while maintaining appropriate power points and effectively utilizing the allowable power reserve. Furthermore, the proposed method preserves the control structure and dynamic performance of conventional GFM control and inherently adapts to varying wind-speed conditions. Comparative case studies under different operating scenarios demonstrate the effectiveness and superiority of the proposed strategy.

eess.SY↗

arXiv2026

TaskArtisan: Designing Composable Generative Widgets for LLM-Assisted Analysis

People increasingly use chatbots such as ChatGPT for everyday analysis tasks. While chatbots unify many analysis functions (e.g., scripts, visualizations, summaries), long conversations become hard to navigate, making it difficult to revisit prior steps or reuse successful workflows. LLMs now generate high-fidelity GUI code that enables people to create customized analysis tools beyond text. Yet, what new opportunities generative UIs bring to analysis work remain unclear. We interviewed six professionals about analysis with chatbots, analyzed publicly shared LLM-generated GUI tools, and conducted a comparison study (N=12) between a chatbot and TaskArtisan, a technology probe that enables people to create and assemble generative analysis UI widgets for sequential and fan-out composition. We find that GUI improved clarity and visual presentation but also introduced rigidity and additional prompting challenges. We summarize the trade-offs into a provisional design framework (malleability, specification, interoperability) to inform future generative UI in LLM-assisted analysis workflows.

cs.HC↗

arXiv2026

LaDA-Band: Language Diffusion Models for Vocal-to-Accompaniment Generation

Vocal-to-accompaniment (V2A) generation, which aims to transform a raw vocal recording into a fully arranged accompaniment, inherently requires jointly addressing an accompaniment trilemma: preserving acoustic authenticity, maintaining global coherence with the vocal track, and producing dynamic orchestration across a full song. Existing open-source approaches typically make compromises among these goals. Continuous-latent generation models can capture long musical spans but often struggle to preserve fine-grained acoustic detail. In contrast, discrete autoregressive models retain local fidelity but suffer from unidirectional generation and error accumulation in extended contexts. We present LaDA-Band, an end-to-end framework that introduces Discrete Masked Diffusion to the V2A task. Our approach formulates V2A generation as Discrete Masked Diffusion, i.e., a global, non-autoregressive denoising formulation that combines the representational advantages of discrete audio codec tokens with full-sequence bidirectional context modeling. This design improves long-range structural consistency and temporal synchronization while preserving crisp acoustic details. Built on this formulation, LaDA-Band further introduces a dual-track prefix-conditioning architecture, an auxiliary replaced-token detection objective for weakly anchored accompaniment regions, and a two-stage progressive curriculum to scale Discrete Masked Diffusion to full-song vocal-to-accompaniment generation. Extensive experiments on both academic and real-world benchmarks show that LaDA-Band consistently improves acoustic authenticity, global coherence, and dynamic orchestration over existing baselines, while maintaining strong performance even without auxiliary reference audio. Codes and audio samples are available at https://github.com/Duoluoluos/TME-LaDA-Band .

cs.SD↗

arXiv2026

High-Precision Hybrid FA-PSO Based Inversion of Building Material Parameters for Fundamental Wireless Performance Evaluation

In this paper, we propose an inversion method based on the firefly particle swarm optimization (FA-PSO) algorithm to estimate the permittivity, conductivity, and thickness of building materials using the free-space method. To improve convergence efficiency and robustness, an adaptive firefly algorithm (FA) is employed to systematically optimize the hyperparameters of the particle swarm optimization (PSO). By optimizing the parameters of the Gaussian distribution used for population initialization, the accuracy of parameter estimation is gradually improved. Furthermore, we derive the Cramer-Rao lower bound (CRLB) for the permittivity, conductivity, and thickness under a complex Gaussian noise model, which serves as a theoretical benchmark for evaluating the estimation accuracy of the FA-PSO algorithm. Numerical results indicate that for relatively thin materials, the estimation accuracy of the proposed method approaches this theoretical lower bound, confirming the effectiveness of the inversion framework. This study accurately extracts the electromagnetic properties of building materials, providing strong support for evaluating their wireless performance.

cs.NI↗

arXiv2026

Observational Signatures of Rotating Ayón-Beato-García Black Holes: Shadows, Accretion Disks and Images

We investigate the shadows, accretion disks, and observational images of rotating Ayón-Beato--Garc\'ıa (ABG) black holes with mass $M$, spin $a$, and nonlinear-electrodynamics (NLED) charge parameter $ζ$, treating photons as neutral test particles on null geodesics of the background metric so as to obtain the purely geometric shadow. The shadow shrinks with increasing $ζ$ and develops a ``D''-shaped morphology for near-extremal spin. The thermal disk properties follow from the Novikov--Thorne model with inner edge at the innermost stable circular orbit (ISCO), whereas the images use a separate, phenomenological optically thin emission model extending to the horizon. The image asymmetry and redshift maps depend strongly on $(a, ζ)$ and the inclination. Comparing the geometric shadow diameters with Event Horizon Telescope observations of M87$^{*}$ and Sgr A$^{*}$ gives the indicative estimates $ζ\lesssim 0.21\,M$ and $ζ\lesssim 0.43\,M$, respectively; since $ζ$ characterizes each black hole individually, these are quoted separately rather than combined, and the Kerr limit $ζ= 0$ remains fully consistent with both.

gr-qc↗

arXiv2026

Exceptionally high carrier mobility in hexagonal diamond

Hexagonal diamond (h-diamond), or Lonsdaleite, is a promising wide-bandgap semiconductor known for its high thermal conductivity and hardness. Based on \textit{ab initio} calculations, we demonstrate its exceptionally high carrier mobilities. At room temperature, the hole mobilities along the $\perp c$ and $\parallel c$ directions are 6000 and 6024 cm$^{2}$V$^{-1}$s$^{-1}$, respectively, while the corresponding electron mobilities reach 12339 and 28473 cm$^{2}$V$^{-1}$s$^{-1}$. These values are significantly superior to those of most known semiconductors, including cubic diamond. The small effective masses in h-diamond are comparable to those in the cubic phase, which cannot explain its substantially higher mobilities. Instead, two underlying mechanisms are uncovered. First, selection rules enforced by the symmetry of h-diamond significantly suppress scattering, particularly for transverse acoustic phonons, which predominate in the cubic phase around room temperature. Secondly, the spatial mismatch between the electronic wavefunctions and phonon-induced scattering potentials leads to real-space electron-phonon decoupling, which manifests as the suppression of out-of-plane polarised longitudinal acoustic scattering for holes, and a systematic weakening of acoustic scattering for electrons.

cond-mat.mtrl-sci↗

arXiv2026

AutoResearchClaw: Self-Reinforcing Autonomous Research with Human-AI Collaboration

Automating scientific discovery requires more than generating papers from ideas. Real research is iterative: hypotheses are challenged from multiple perspectives, experiments fail and inform the next attempt, and lessons accumulate across cycles. Existing autonomous research systems often model this process as a linear pipeline: they rely on single-agent reasoning, stop when execution fails, and do not carry experience across runs. We present AutoResearchClaw, a multi-agent autonomous research pipeline built on five mechanisms: structured multi-agent debate for hypothesis generation and result analysis, a self-healing executor with a \textsc{Pivot}/\textsc{Refine} decision loop that transforms failures into information, verifiable result reporting that prevents fabricated numbers and hallucinated citations, human-in-the-loop collaboration with seven intervention modes spanning full autonomy to step-by-step oversight, and cross-run evolution that converts past mistakes into future safeguards. On ARC-Bench, a 25-topic experiment-stage benchmark, AutoResearchClaw outperforms AI Scientist v2 by 54.7%. A human-in-the-loop ablation across seven intervention modes reveals that precise, targeted collaboration at high-leverage decision points consistently outperforms both full autonomy and exhaustive step-by-step oversight. We position AutoResearchClaw as a research amplifier that augments rather than replaces human scientific judgment. Code is available at https://github.com/aiming-lab/AutoResearchClaw.

cs.AI↗

arXiv2026

Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation

We present Hunyuan3D 2.0, an advanced large-scale 3D synthesis system for generating high-resolution textured 3D assets. This system includes two foundation components: a large-scale shape generation model -- Hunyuan3D-DiT, and a large-scale texture synthesis model -- Hunyuan3D-Paint. The shape generative model, built on a scalable flow-based diffusion transformer, aims to create geometry that properly aligns with a given condition image, laying a solid foundation for downstream applications. The texture synthesis model, benefiting from strong geometric and diffusion priors, produces high-resolution and vibrant texture maps for either generated or hand-crafted meshes. Furthermore, we build Hunyuan3D-Studio -- a versatile, user-friendly production platform that simplifies the re-creation process of 3D assets. It allows both professional and amateur users to manipulate or even animate their meshes efficiently. We systematically evaluate our models, showing that Hunyuan3D 2.0 outperforms previous state-of-the-art models, including the open-source models and closed-source models in geometry details, condition alignment, texture quality, and etc. Hunyuan3D 2.0 is publicly released in order to fill the gaps in the open-source 3D community for large-scale foundation generative models. The code and pre-trained weights of our models are available at: https://github.com/Tencent/Hunyuan3D-2

cs.CV↗