arXiv subjects

Yifan Zhang

Publications and source records attributed to Yifan Zhang.

At least 73 records · Page 4Linked to original sources

arXiv2026

StepAudio 3 Gen Technical Report

We introduce StepAudio 3 Gen, a general-purpose audio generation model that supports zero-shot text-to-speech (TTS), voice design, vocal generation, sound effects, music, vibe speech, and mixtures of multiple audio types within a unified framework. At its core, StepAudio 3 Gen is a discrete autoregressive generator that models audio directly over residual vector quantization (RVQ) tokens, departing from the diffusion Transformer-based continuous generation paradigm prevalent in recent general audio models. Its StepAudio Tokenizer represents general audio at 12.5 Hz in a shared $16 \times 2048$ residual code space, jointly quantizing semantic and waveform-level acoustic features so that each code layer preserves both types of information. For generation, the backbone predicts the first codebook along the time axis using autoregressive modeling, while a lightweight causal Transformer completes the remaining fifteen codebooks along the codebook axis. Our study further identifies three key design principles: (1) interference-aware progressive pretraining for acquiring audio capabilities while preserving the textual abilities of the large language model, (2) RVQ Adaptor for effectively incorporating multi-codebook acoustic representations, and (3) discrete autoregressive modeling over a shared representation across general audio domains. With progressive pretraining, multi-task instruction training, and supervised fine-tuning, StepAudio 3 Gen achieves state-of-the-art performance on both TTS and voice design, while retaining strong generation capabilities across speech, vocals, sound effects, and music. Audio samples are available at https://stepaudiollm.github.io/step-audio-3-gen/.

cs.SD↗

arXiv2026

When Ad Networks Misbehave: Understanding Risks of Semi-Drive-By Splash Ads

We investigate the mobile splash ads ecosystem, i.e., full-screen advertisements shown at app launch, where monetization relies on interaction signals that are difficult to verify end-to-end. This setting is especially sensitive because incidental touches and sensor-driven callbacks are common yet easy to misattribute as engagement. Prior work has largely framed mobile ad fraud as a publisher-side problem, while some studies attribute fraudulent operations to embedded ad libraries. Yet an important risk remains underexplored: ad SDKs control how interaction signals are interpreted, measured, and reported, creating an opportunity to reinterpret ambiguous user or device signals as valid advertising interactions. We uncover a previously less-known form of fraud at the ad-network layer in which splash ads are triggered not by intentional user actions but by incidental or indirect interactions, which we term semi-drive-by splash ads. By translating non-ad interactions into billable engagement events, ad networks can inflate performance metrics, overcharge advertisers, and erode user trust. To expose this behavior in the wild, we design AdHive, an automated honeypot-like analysis framework that induces evasive splash-ad delivery and landing behaviors under realistic device conditions. AdHive reproduces human-like activity through LLM-generated usage traces and sensor dynamics, enabling execution paths that remain hidden in conventional analysis environments. Our large-scale measurement across thousands of popular Android applications shows that semi-drive-by splash ads are widespread and are often triggered by subtle signals such as minor sensor variations. We further confirm real-world impact by working with one of China's largest advertisers, identifying multiple ad networks engaging in this fraud and leading to enforced repayments of about 4 million Yuan (approximately US$600,000).

cs.CR↗

arXiv2026

MedQA-MM: Shortcuts Behind Medical Visual Reasoning

A benchmark score credits final answers, but not the route by which an item can be answered. In medical multimodal multiple-choice questions (MCQs), this distinction matters because a correct answer can be supported by the intended image finding or by benchmark-preserved cues in the wording of answers, non-visual clinical text, visible image text, artificial annotations, or device/context artifacts. We call the resulting score-level overinterpretation reasoning inflation. Here, a route is an observable input path that can support answer selection, not a claim about the model's hidden cognition. Across six medical multimodal MCQ datasets, we separate candidate cues from behavioral evidence through prompt- and image-side audits, modality ablations, and matched repairs that preserve the medical target and answer key. In a 13-configuration open-model panel, full-input accuracy is 62.63%, while text-only and options-only settings achieve 53.96% and 29.71%, respectively. Removing length-gap, absolute/conspicuous, and spatial/prepositional cues lowers accuracy by 6.58, 3.50, and 4.77 percentage points. We also construct MedQA-MM, a 1,000-item shortcut-mitigated subset, where text-only and options-only accuracy fall to 5.21% and 12.33%. This does not imply that models never use images; it shows that medical image-reasoning claims require route-level evidence.

cs.CV↗

arXiv2026

Direct Realization of Near-Ideal Carbyne in Ultrathin Boron Nitride Nanotubes

Carbyne, the sp-hybridized one-dimensional allotrope of carbon, is predicted to be the stiffest known material, with electronic and optical properties set by a single structural parameter, the bond length alternation. However, its intrinsic properties have never been measured: chains synthesized through molecular chemistry carry endgroup and finite-length perturbations that persist even in the longest molecules available, while chains grown inside carbon nanotubes strongly couple to the host, which renormalizes their vibrational frequency by up to 110 cm$^{-1}$ in a diameter-dependent manner. Here, we show that encapsulating and thermally converting hydrogen-capped polyynes inside ultrathin boron nitride nanotubes, structural analogues of carbon nanotubes but electrically insulating, yields carbyne chains in a near-ideal regime, where endgroup, finite length, and host-guest perturbations are reduced to secondary effects. Statistical Raman spectroscopy across 245 locations returns a vibrational frequency distribution an order of magnitude narrower than in carbon nanotubes, an anharmonicity consistent with the universal law for carbyne-like materials, and a bond length alternation matching correlated calculations for the free chain. No photoluminescence is detected, despite the transparent host, as expected for the dipole-forbidden emission of an unperturbed carbyne chain. Boron nitride nanotubes give experimental access to carbyne in its near-ideal form.

cond-mat.mes-hall↗

arXiv2026

GeoContext: One Context Ladder, Two Failure Modes in Vision-Language Geolocation: Flat Reliance on User-Provided Location Context and False Confirmation of Location Claims

Visual geolocation benchmarks typically ask a model where an image was captured without accounting for the location context that users often provide. We introduce GeoContext, a resource supporting two complementary tasks: GeoHint, open-ended localization given a true but coarse location hint, and GeoVerify, binary verification of whether an image was taken within 150 m of a claimed place. GeoContext constructs a context ladder by stratifying nearby reference points according to distance and referenceability, allowing the image to remain fixed while the supplied context varies. The benchmark covers 109 sites in 30 cities and evaluates five vision-language models using 21,933 GeoHint responses and 6,270 GeoVerify responses. Our evaluation reveals three main patterns. First, hint repetition varies by only 1.5 percentage points across referenceability tiers and by less than 3 points across distance bands, while the resulting localization error increases steadily with hint distance. Second, behavior depends strongly on no-context performance: at sites with low no-context accuracy, the median ratio between localization error and hint distance is approximately 1.00, whereas at higher-accuracy sites it ranges from 0.24 to 0.69. After correcting for bias introduced by the site grouping procedure, only one of the five models retains a negative accuracy estimate when given a nearby hint. Third, in GeoVerify, no model reaches d' = 1 for decoys immediately beyond the 150 m tolerance. Model rankings also change when sensitivity is separated from response bias, and 83.8% of false acceptances are reported with confidence of at least 0.8. We release the benchmark, construction pipeline, audit decisions, and scoring code.

cs.CV↗

arXiv2026

Comparables XAI: Faithful Example-based AI Explanations with Counterfactual Trace Adjustments

Explaining with examples is an intuitive way to justify AI decisions. However, it is challenging to understand how a decision value should change relative to the examples with many features differing by large amounts. We draw from real estate valuation that uses Comparables-examples with known values for comparison. Estimates are made more accurate by hypothetically adjusting the attributes of each Comparable and correspondingly changing the value based on factors. We propose Comparables XAI for relatable example-based explanations of AI with Trace adjustments that trace counterfactual changes from each Comparable to the Subject, one attribute at a time, monotonically along the AI feature space. In modelling and user studies, Trace-adjusted Comparables achieved the highest XAI faithfulness and precision, user accuracy, and narrowest uncertainty bounds compared to linear regression, linearly adjusted Comparables, or unadjusted Comparables. This work contributes a new analytical basis for using example-based explanations to improve user understanding of AI decisions.

cs.HC↗

arXiv2026

Spectral properties of the magnetic Robin Laplacian on curvilinear half-plane

We analyse magnetic Robin Laplacian in curvilinear half-plane with a smooth boundary. It is well known that the spectrum of the Robin Laplacian is unstable with respect to boundary deformations. This means that if the boundary is a straight line then the spectrum of the Robin Laplacian is purely essential. From the other hand, the perturbation of the boundary produces eigenvalues below the essential spectrum. In this paper, the Robin-Laplace operator with a compactly supported magnetic field is considered. We prove that the spectrum of the magnetic Robin Laplacian is stable under small and local deformations of the boundary.

math.SP↗

arXiv2026

MicroEvo: Knowledge-Guided LLM Sampling for Efficient Microarchitecture Design Space Exploration

Microarchitecture design space exploration suffers from expansive search spaces and expensive PPA evaluation, leaving only a small simulation budget for design decision-making. Existing methods perform blind search without considering microarchitectural dependencies and fail to learn from the iterative search effectively, leading to wasted evaluations and weak Pareto convergence. In this paper, we propose MicroEvo, a knowledge-guided framework that couples off-the-shelf LLMs with Monte Carlo Tree Search (MCTS) for multi-objective microarchitecture optimization. MicroEvo combines LLM-driven evolutionary operators, a Pareto-aware tree policy that balances Pareto contribution and diversity, an active knowledge accumulation mechanism that extracts and reuses optimization insights, and state-aware directives that adapt the search behavior online. Experiments show that MicroEvo improves Pareto-front quality by up to 36.2% over NSGA-II and achieves 10.6x higher search efficiency, and also demonstrates strong scalability to a complex industrial-scale core. The code repository is available at: https://github.com/GEAR-SEU/MicroEvo-ICCAD-26.

cs.AI↗

arXiv2026

A Survey on Self-Improving Test-Time Intelligence: Feedback-Driven Adapting, Learning, and Scaling at Inference

The ability of AI systems to improve their behavior during deployment is becoming increasingly important. As inference moves beyond the static execution of a fixed trained model, a growing body of work studies how models can refine their behavior on the fly by exploiting test-time information and additional computation. These developments have largely evolved along two directions: methods that modify the model's state using test-time signals, and methods that improve predictions through extra inference-time resources such as more sampling and tool use. However, these directions are often studied in separate communities with different terminology, making their connections harder to see. In this survey, we present feedback-driven Test-Time Intelligence (TTI) as a unified perspective for understanding such deployment-time improvement. We use this view to relate test-time adaptation, test-time learning, and test-time scaling, highlighting both their distinctions and their growing overlap in hybrid systems. This unified framework helps connect previously fragmented ideas and provides a clearer conceptual foundation for studying inference-time self-improvement. We review major methodological paradigms, representative applications, and open challenges across vision, language, multimodal learning, generative models, robotics, and healthcare. Our goal is to provide a coherent foundation and research roadmap for the study of self-improving AI systems at test time.

cs.LG↗

arXiv2026

Small Updates, Big Doubts: Does Parameter-Efficient Fine-tuning Enhance Hallucination Detection ?

Parameter-efficient fine-tuning (PEFT) methods are widely used to adapt large language models (LLMs) to downstream tasks and are often assumed to improve factual correctness. However, how the parameter-efficient fine-tuning methods affect hallucination behavior remains insufficiently understood, especially on QA datasets. In this work, we systematically investigate the impact of PEFT on hallucination detection through a comprehensive empirical study across three open-weight LLM backbones and three fact-seeking QA benchmarks. For each model, we evaluate performance using seven unsupervised hallucination detection methods spanning three complementary approaches: semantic consistency based detectors, confidence based detectors, and entropy based detectors. This multifaceted evaluation enables us to characterize how PEFT reshapes uncertainty across different detection paradigms. In conclusion, our experimental results show that PEFT consistently strengthens hallucination detection ability, substantially improving AUROC across a wide range of hallucination detectors. Besides, further analyses using linear probes and representation diagnostics indicate that PEFT methods primarily reshapes how uncertainty is encoded and surfaced, comparing with injecting new factual knowledge into the models.

cs.CL↗

arXiv2026

Mind-VLA: Instruction-Aware Spatial Representation Alignment for Vision-Language-Action Models

Recent Vision-Language-Action (VLA) methods improve generalization by aligning their representations with 3D scene geometry. However, these methods are fundamentally instruction-agnostic: the representations align the entire scene uniformly, neglecting the 3D geometry of the specific target object designated by the language instruction. This causes failures on fine-grained manipulation and target occlusion tasks, where success depends on accurate 3D understanding of the target object rather than the entire scene. To address this, we present Mind-VLA, an instruction-aware spatial representation alignment method for VLA models. Specifically, Mind-VLA first obtains the target object specified by the language instruction, then prepares its canonical target views and extracts the corresponding VAE and VGGT features. Finally, the latent representation of the VLA model is aligned with these features to enable instruction-aware 3D understanding. Mind-VLA reaches 94.4% on LIBERO and 4.47 on CALVIN with a compact 345M-parameter backbone. On real-robot tasks with target occlusion, Mind-VLA reaches 54% average success, outperforming the matched scene-VGGT control by 26 percentage points.

cs.RO↗

arXiv2026

Fast Weight Attention for Continual Learning

Recurrent fast-weight memories and selective state-space models compress an expanding context into a fixed-size recurrent state, making the state transition an online learning rule. We study this rule under read-after-write autoregressive semantics. For the prefix-prediction objective considered here, the local fast-memory example revealed at step $t$ is the prefix-aligned pair $(\mathbf{x}_t,\mathbf{y}_t)=(ϕ(\mathbf{k}_{t-1}),\mathbf{v}_t)$. The common same-step association $(ϕ(\mathbf{k}_t),\mathbf{v}_t)$ remains causal, but optimizes a different internal objective. We derive normalized first-order updates for squared-error regression and negative inner-product objectives. The regression family comprises Falcon-1 (a scalar NLMS update), Falcon-2 (its per-column extension), and Falcon-3 (a sliding-window mini-batch update); Falcon-1A/Falcon-2A/Falcon-3A are the corresponding inner-product variants. We provide recurrent, masked-parallel, and chunk-parallel forms, together with numerically stable positive-decay renormalization. Representative variants remain competitive in language modeling and improve length extrapolation on variable-digit addition. This framework separates temporal alignment, plasticity, forgetting, and bounded rehearsal in recurrent sequence models.

cs.LG↗

arXiv2026

Observation of Quantized Charge Accumulation in a Quantum Anomalous Hall System

The quantum anomalous Hall effect in magnetically doped topological insulators exhibits a quantized Hall conductance $σ_{xy} = e^2/h$ arising from the two-dimensional surface states. While conventional transport probes confirm this quantization, they remain insensitive to the field-induced surface charge accumulation as a direct manifestation of $σ_{xy}$. Here, we experimentally validate an out-of-plane capacitive method that directly detects this quantized charge accumulation in a quantum anomalous Hall system. Using Corbino and simple disk devices, we measure charge accumulation proportional to field variation $ΔB$, with dissipation characterized by longitudinal conductance $σ_{xx}$ and frequency $f$. A quantitative dissipation model extracts the intrinsic quantized charge density $η_0 = (e^2/h)ΔB$, which is confirmed through $f$- and $σ_{xx}$-dependent measurements. Under ultra-low dissipation ($σ_{xx} \approx 10^{-9}$ S), we directly resolve the fully quantized charge accumulation. This methodology establishes a direct charge-accumulation probe and provides a pathway toward detecting the topological magnetoelectric effect, a condensed matter manifestation of the four-dimensional quantum Hall effect.

cond-mat.mes-hall↗

arXiv2026

Topological Surface Charge Detection via Active Capacitive Compensation: A Pathway to the 4D Quantum Hall Effect

The topological magnetoelectric effect (TME) in three-dimensional topological insulators (TIs), described by $ΔP = \frac{e^2}{2h} N_{\rm Ch}^{(2)} ΔB$, serves as a condensed-matter realization of the four-dimensional quantum Hall effect (4D QHE). In dual-gate axion-insulator devices, the TME-induced polarization yields a current $I_{\rm TME} \propto (C_{\rm total}/C_{\rm S})\,Q_{\rm 4D\mathrm{-}QHE}$, where the signal is suppressed by the capacitance ratio $C_{\rm total}/C_{\rm S}$. Here we propose an active compensation scheme that introduces a tunable negative capacitance $C_{\rm comp} \approx -C_{\rm gate}$ into the gate line, effectively canceling the gate dielectric capacitance and driving $C_{\rm total}/C_{\rm S} \to 1$. We validate the method using a quantum anomalous Hall (QAH) device, which shares the same surface-state physics as the axion insulator but permits direct charge measurement via a single gate, recovering over $95\%$ of the quantized charge signal from an initially half-attenuated state. This compensation method provides a robust means of resolving minute TME signals, offering a promising pathway toward direct measurements of the 4D QHE.

cond-mat.mes-hall↗

arXiv2026

Optimizing Multi-Modality Trackers via Significance-Regularized Tuning

This paper tackles the critical challenge of optimizing multi-modality trackers by effectively adapting pre-trained models for RGB data. Existing fine-tuning paradigms oscillate between excessive flexibility and over-restriction, both leading to suboptimal plasticity-stability trade-offs. To mitigate this dilemma, we propose a novel significance-regularized fine-tuning framework, which delicately refines the learning process by incorporating intrinsic parameter significance. Through a comprehensive investigation of the transition from pre-trained to multi-modality contexts, we identify that parameters crucial to preserving foundational patterns and managing cross-domain shifts are the primary drivers of this issue. Specifically, we first probe the tangent space of pre-trained weights to measure and orient prior significance, dedicated to preserving generalization. Subsequently, we characterize transfer significance during the fine-tuning phase, emphasizing adaptability and stability. By incorporating these parameter significance terms as unified regularization, our method markedly enhances transferability across modalities. Extensive experiments showcase the superior performance of our method, surpassing current state-of-the-art techniques across various multi-modal tracking benchmarks. The source code and models are publicly available at https://github.com/zhiwen-xdu/SRTrack.

cs.CV↗

arXiv2026

CharTool: Tool-Integrated Visual Reasoning for Chart Understanding

Charts are ubiquitous in scientific and financial literature for presenting structured data. However, chart reasoning remains challenging for multimodal large language models (MLLMs) due to the lack of high-quality training data, as well as the need for fine-grained visual grounding and precise numerical computation. To address these challenges, we first propose DuoChart, a scalable dual-source data pipeline that combines synthesized charts with real-world charts to construct diverse, high-quality chart training data. We then introduce CharTool, which equips MLLMs with external tools, including image cropping for localized visual perception and code-based computation for accurate numerical reasoning. Through agentic reinforcement learning on DuoChart, CharTool learns tool-integrated reasoning grounded in chart content. Extensive experiments on six chart benchmarks show that our method consistently improves over strong MLLM baselines across model scales. Notably, CharTool-7B outperforms the base model by +8.0% on CharXiv (Reasoning) and +9.78% on ChartQAPro, while achieving competitive performance with substantially larger or proprietary models. Moreover, CharTool demonstrates positive generalization to out-of-domain visual math reasoning benchmarks.

cs.AI↗

arXiv2026

Bern2Edge: A Neurosymbolic Compiler for Edge Deployment via Bernstein Polynomial Networks

Deploying high-accuracy neural networks on resource-constrained edge devices remains challenging, as existing approaches treat training, compression, and hardware synthesis as separate stages, leaving a gap between software-trained models and efficient end-to-end deployment with limited support for interpretability. We propose Bern2Edge, an end-to-end framework that uses knowledge distillation to convert a pretrained teacher feed-forward network into hardware-efficient representations via Bernstein polynomial activations. This representation enables two deployment paths: (i) a high-fidelity LUT-based realization that preserves model fidelity under compression, and (ii) a symbolic rule-based representation derived from Bernstein activation geometry, enabling interpretable inference with explicit input-space constraints. The resulting BNNs achieve up to 2.12 percentage-point (pp) accuracy improvement over ReLU under identical compression constraints. At the system level, Bern2Edge achieves up to 99.8% latency reduction and 95.2% BRAM reduction relative to a W8A8 quantized teacher on an AMD Xilinx KV260 FPGA, while maintaining accuracy within 0.5 pp, and further deploys on a low-power Spartan-7 XC7S15 FPGA. The rule-based path reduces DSP usage by up to 89.0% at a cost of 1.5 pp in total accuracy.

cs.LG↗

arXiv2026

Toward Better Assessment of LLMs' Performance in Clinical Error Detection

Automated detection of errors in clinical documentation is a promising application of large language models (LLMs), yet decisions to deploy such models rest on benchmarks that evaluate each clinical note in isolation. Error-detection benchmarks are typically constructed by injecting errors into notes, such that each erroneous note has a natural counterpart. Aggregate discriminative metrics (e.g., balanced accuracy or F1) do not exploit this structure. We show that this omission is consequential. In particular, evaluating 15 diverse LLMs on 4 standardized clinical error-detection test sets across 3 languages, we find that 13 of 15 models fall below the level of random pairwise discrimination, even while achieving F1 scores that standard practice would read as moderate. We also observe that the underlying bias patterns differ across languages: the same model can default to "no error" on one language and over-flag errors on another. To diagnose where discrimination breaks down, we further introduce a procedure to score the evidence models cite in their outputs. We find that while models consistently locate error-relevant content, they fail to produce the corresponding correct verdict on the clean counterpart. Finally, we show that F1 and pairwise accuracy are driven in opposite directions by the same underlying bias, so that ranking models by F1 may systematically promote the weakest discriminators. For safety-critical clinical NLP applications, we advocate for supplementing aggregate metrics with paired evaluations in benchmark reporting. Code and analysis scripts are available at https://github.com/healthylaife/paired-clinical-eval.

cs.CL↗