Searcharxiv⌕ Search

arXiv subjects

Kani Chen

Publications and source records attributed to Kani Chen.

At least 19 recordsLinked to original sources
arXiv2026

Repeated-Game Security for Restaking-Based Verifiable Inference

Restaking-based protocols enable verifiable LLM inference without the high proving cost of zkML or the hardware trust assumptions of TEEs. Their security is commonly justified by a one-round slashing condition: a rational provider should not cheat when the expected penalty exceeds the cost saving from dishonest inference. This paper shows that this condition can overstate security when inference is supplied repeatedly under the same stake. We model verifiable inference as a discounted repeated game and identify a repeated-game gap caused by proportional slashing: detected deviations reduce future penalty exposure, while cost savings are earned again across queries. We derive the gap in closed form, show that it persists under minimum-stake ejection, and extend it to memoryless bounded-slashing protocols covering deployed designs. We propose a deployable mechanism combining history-dependent challenges, reputation-weighted slashing, and stake vesting. The mechanism restores infinite-horizon subgame-perfect incentive compatibility against stationary mixed-strategy deviations above an explicit discount-factor threshold without per-query cryptographic verification. Evaluation across nine open-weight model pairs from 0.5B to 14B parameters shows that the audit signal has the required concave detectability response. A Stackelberg audit-budget analysis shows that improved signal responsiveness reduces the baseline audit rate by 2.6x at discount factor 0.95. Calibrated to deployed parameters, surveyed protocols pass one-round incentive compatibility but admit repeated-game deviations for discount factors between 0.92 and 0.98, with deviation-profit fractions of 1.5%--8%. Our mechanism reduces deviation profits by 31%--54% while maintaining low-latency economic verification.

cs.GT↗
arXiv2026

SocialFiVis: A Visual Analytics Sandbox for LLM-Grounded Multi-Agent Simulation in Social Finance

The emergence of social finance (SocialFi) transforms online communities into complex socio-economic systems. Within these spaces, collective decisions shape a "digital commons" characterized by social capital (e.g., community trust) and financial health (e.g., market liquidity). Governing such hybrid ecosystems is challenging because real-world interventions are costly and irreversible. While counterfactual simulation is essential for exploring alternative governance strategies, existing approaches fail to capture the non-linear interplay between governance rules, individual behaviors, and emergent economic outcomes. To systematically unpack this complexity, we operationalize the Institutional Analysis and Development (IAD) framework as our theoretical foundation, synthesizing prior literature with insights from formative expert interviews. Built on this framework, we present SocialFiVis, an IAD-embedded visual analytics sandbox. It introduces a robust model to quantify the dual-track digital commons, coupled with a two-phase simulation engine. This engine combines LLM-derived personas with a mechanism-guided Perception-Reasoning-Action (PRA) runtime to simulate heterogeneous, context-aware agents empirically grounded in the retained messaging cohort. A hierarchical multi-view interface with interpretable reasoning pathways enables community operators to explore counterfactual policies and trace system-level outcomes back to individual behavioral rationales. We evaluate SocialFiVis through two case studies, a user study, and follow-up interviews. Results demonstrate that SocialFiVis supports fine-grained behavioral attribution and helps explain emergent phenomena such as the structural decoupling of social capital and the resilience of messaging members under localized governance shocks.

cs.HC↗
arXiv2026

CORTIVA: Candidate-Score Fusion of Complementary Visual Teachers for EEG- and MEG-to-Image Retrieval

Decoding visual experience from non-invasive brain activity is central to neuroscience and brain-computer interfaces. Functional magnetic resonance imaging (fMRI) offers fine spatial detail, but its slow hemodynamics and burdensome acquisition limit temporally resolved decoding. Electroencephalography (EEG) and magnetoencephalography (MEG) provide millisecond resolution, making image retrieval compelling: identify the viewed image from one neural response and a fixed candidate bank. Contrastive alignment to pretrained visual representations enables zero-shot retrieval from EEG and MEG, but most systems collapse heterogeneous visual supervision into a single embedding before ranking. This early consolidation imposes one similarity geometry on every candidate order and removes encoder-specific disagreements from the final ranking. We propose CORTIVA, a candidate-score fusion framework that preserves this complementary evidence. Three decoding routes are aligned to heterogeneous visual targets, score the same indexed candidates independently, and combine only their temperature-scaled score vectors before ranking. On the 200-way THINGS-EEG2 benchmark, CORTIVA reaches 73.5% Top-1 and 95.3% Top-5 across ten participants, exceeding the strongest reported baseline by 10.3 and 5.4 percentage points. With a modality-specific neural encoder, the same fusion principle reaches 42.4% Top-1 on THINGS-MEG. Matched route-removal retraining and four weight controls demonstrate that CORTIVA's gain arises from integrating complementary route scores and persists with uniform weighting, without requiring a specialized weighting rule. Independent DINOv2 analyses further reproduce the local error neighborhoods and posterior neural-visual correspondence. These results establish candidate-score fusion as a simple and testable alternative to embedding-level consolidation for neural image retrieval.

cs.CV↗
arXiv2026

CRAFT: Counterfactual Credit Assignment from Free Sibling Rollouts for Self-Distilled Agentic Reinforcement Learning

Self-distilled agentic reinforcement learning augments trajectory-level reward with a token-level distillation loss, using as its teacher the same policy conditioned on privileged context. The prevailing recipe gates this loss by a single scalar, the teacher-student log-probability gap. This signal is doubly limited: it is retrospective, scoring only the realised rollout and never the counterfactual ones, and it is sign-blind, never signalling when a teacher-preferred action would have harmed the trajectory. We introduce CRAFT, a three-pillar credit-assignment scheme that addresses both limitations. Pillar 1, Counterfactual Token Importance, reuses the G-1 sibling rollouts that GRPO already samples and importance-weights them by the log-probability gap to form a self-normalised estimate of the group-level counterfactual change in advantage from up-weighting teacher-preferred actions at each step; this yields a signed per-token credit at near-zero extra compute. Pillar 2 is an asymmetric controller that raises the distillation weight as it lowers the reference-KL weight along an exponential moving average of gate activity, and conversely. Pillar 3 polarises the KL penalty token by token, switching between a mode-seeking and a mode-covering update according to the sign of the credit. Each pillar has an independent switch that, when disabled, renders the loss and gradient byte-identical to the baseline in IEEE-754 arithmetic, so any measured gain is attributable to algorithmic change rather than implementation drift. We prove the estimator's consistency and a variance bound, give structural and bit-exact reproducibility guarantees, and evaluate CRAFT across three agentic environments, four model scales, and five end-to-end methods, plus two tabulated prior-work baselines. Among these is Adaptive-CRINGE, a comparator sharing Pillar 2 with CRAFT, isolating the counterfactual contribution.

cs.LG↗
arXiv2026

Depth-Entropy Guided Sampling for Training-Free LLM Reasoning

Reinforcement learning (RL) has become the dominant paradigm for improving the reasoning capabilities of large language models, but it requires expensive training, curated data, and reward signals. Recent work shows that sampling from sharpened base-model distributions at test time recovers much of the RL gain, yet existing methods rely solely on output-layer likelihoods and ignore the transformer's internal forward-pass dynamics. We introduce Depth-Entropy Guided Sampling (DEGS), a training-free, test-time method that exploits layer-wise entropy collapse as an intrinsic quality signal. We observe that stronger reasoners -- including RL-posttrained variants -- exhibit a distinctive "late collapse": logit-lens decoded entropy stays elevated until deeper layers before converging. We define a per-sequence collapse depth $D(\mathbf{x})$ and a joint objective $\pi(\mathbf{x}) \propto p(\mathbf{x})^\alpha \exp(\beta D(\mathbf{x}))$ that combines sequence likelihood with this depth-entropy structure, instantiated inside an MCMC power-sampling framework (DEGS-MCMC). Across three open-weight models and four reasoning benchmarks, this near-chance per-candidate signal compounds over the sampling trajectory into state-of-the-art training-free accuracy, with gains largest out of domain and on the harder splits -- exactly where likelihood alone falls short -- at single-digit-percent wall-clock overhead. DEGS narrowly trails an in-house GRPO reference on the math splits GRPO was trained for, yet surpasses it out of domain on GPQA for all three models, without any training, reward model, or labeled data.

cs.LG↗
arXiv2026

Fine-Tuning Integrity for Modern Neural Networks: Structured Drift Proofs via Norm, Rank, and Sparsity Certificates

Fine-tuning is the dominant paradigm for adapting large machine learning models, yet current deployment pipelines provide no way to verify how a released model was updated. In particular, a model provider or auditor cannot check whether a fine-tuned model adheres to a claimed update procedure without access to its parameters. We introduce \emph{fine-tuning integrity} (FTI), a cryptographic objective for verifying that a deployed model differs from a trusted base model only within a declared class of admissible updates. We construct \emph{succinct model difference proofs} (SMDPs), zero-knowledge protocols that certify structured parameter drift without revealing model weights. Our framework supports three fundamental update classes: norm-bounded, low-rank, and sparse drift, covering common fine-tuning methods such as regularized training, LoRA, and prefix tuning. In all cases, proof size and verification cost depend on the structure of the update rather than the number of parameters. We prove soundness, zero-knowledge, and succinctness for each construction, and establish a matching $\Omega(n)$ lower bound showing that structural assumptions are necessary for succinct verification. A prototype evaluation on synthetic benchmarks and GPT-2 fine-tuning demonstrates that proofs remain compact and verification is efficient at realistic scales.

cs.CR↗
arXiv2026

Economic Security of VDF-Based Randomness Beacons: Models, Thresholds, and Design Guidelines

Randomness beacons based on Verifiable Delay Functions (VDFs) are increasingly proposed for blockchains and distributed systems, promising publicly verifiable delay and bias resistance. Existing analyses, however, treat adversaries purely as cryptographic entities and overlook that real attackers are economically motivated. A VDF may be sequentially secure, yet still vulnerable if a rational adversary can profit by purchasing faster hardware and exploiting reward spikes such as MEV opportunities. We develop a formal framework for economic security of VDF-based randomness beacons. Modeling the attacker as a rational agent facing hardware speedup, operating costs, and stochastic rewards, we cast the attack decision as an optimal-stopping problem and prove that optimal behavior has a monotone threshold structure. This yields tight necessary and sufficient conditions relating delay parameters to adversarial cost and reward distributions. We extend the analysis to grinding, selective abort, and multi-adversary competition, demonstrating how each amplifies effective rewards and increases required delays. Using realistic cloud costs, hardware benchmarks, and MEV data, we show that many proposed VDF delays, on the order of a few seconds, are economically insecure under plausible conditions. We conclude with deployable guidelines and introduce Economically Secure Delay Parameters (ESDPs) to support principled parameter selection in practical systems.

cs.CR↗
arXiv2026

RegGuard: Legitimacy and Fairness Enforcement for Optimistic Rollups

Optimistic rollups provide scalable smart-contract execution but remain unsuitable for regulated financial applications due to three structural gaps: semantic legitimacy, cross-layer state consistency, and ordering fairness. We introduce RegGuard, a unified framework that enhances optimistic rollups with comprehensive legitimacy guarantees. RegGuard integrates three coordinated mechanisms: a decidable semantic validator powered by the RegSpec rule language for encoding regulatory constraints; a cross-layer state pre-synchronization validator that detects inconsistent L1-L2 dependencies with probabilistic reliability bounds; and a cryptographically verifiable fair-ordering service that ensures transaction sequencing fairness with negligible violation probability. We implement a 15,000-line prototype integrated into an Optimism-based rollup and evaluate it under adversarial conditions. RegGuard reduces settlement failures by over 90%, prevents detectable ordering manipulation, and maintains 85% of baseline throughput.

cs.CR↗
arXiv2026

AIvilization v0: Toward Large-Scale Artificial Social Simulation with a Unified Agent Architecture and Adaptive Agent Profiles

AIvilization v0 is a publicly deployed large-scale artificial society that couples a resource-constrained sandbox with a unified LLM-agent architecture, aiming to sustain long-horizon autonomy while remaining executable under a rapidly changing environment. To mitigate the tension between goal stability and reactive correctness, keeping long-horizon objectives on course while each action remains valid in a fast-changing shared world, we introduce (i) a hierarchical branch-thinking planner that decomposes life goals into parallel objective branches and uses simulation-guided validation plus tiered re-planning to ensure feasibility; (ii) an adaptive agent profile with dual-process memory that separates short-term execution traces from long-term semantic consolidation, enabling persistent yet evolving identity; and (iii) a human-in-the-loop steering interface that injects long-horizon objectives and short commands at appropriate abstraction levels, with effects propagated through memory instead of brittle prompt overrides. The environment integrates physiological survival costs, non-substitutable multi-tier production, an AMM-based price mechanism, and a gated education-occupation system. In a large-scale public deployment with tens of thousands of agents, high-frequency transactions from the platform's mature phase reveal stable markets that reproduce key stylized facts of real economies and structured wealth stratification driven by education and access constraints. At the agent level, portraits evolve coherently over long horizons, and human steering is associated with measurably larger short-horizon profile updates. Controlled ablation experiments complement the deployment evidence, showing that our agent architecture is robust in multi-objective, long-horizon settings.

cs.MA↗
arXiv2026

PsyAgent: Constructing Human-like Agents Based on Psychological Modeling and Contextual Interaction

Human-like agents must express stable dispositions while adapting to roles, relationships, and norms. We present PsyAgent, a schema-first framework that operationalizes the trait-context interface by coupling a Big Five trait prior with explicit social-structural conditioning. PsyAgent comprises (i) Individual Structure (IS), a machine-usable trait-grounded profile, and (ii) Multi-Scenario Contexting (MSC), a curated library of role-relationship-norm frames spanning eight everyday arenas. At inference, fixed structured prompts couple the active MSC frame with the IS profile, encouraging behavior that is stable yet context-sensitive. To demonstrate learnability beyond prompt engineering, we use IS and MSC to synthesize supervision and fine-tune compact backbones with PEFT (SFT and optional DPO). Under a controlled psychometric-style evaluation protocol in percentile space, PsyAgent improves trait-faithfulness and long-horizon stability, and is competitive with several larger general-purpose instruction-tuned baselines under matched decoding and scoring controls. We further triangulate the automatic protocol with external benchmarks and a small blinded human study. Overall, PsyAgent provides a precise and data-efficient approach to personality-grounded, norm-aware agents.

cs.AI↗
arXiv2025

TweezeEdit: Consistent and Efficient Image Editing with Path Regularization

Large-scale pre-trained diffusion models empower users to edit images through text guidance. However, existing methods often over-align with target prompts while inadequately preserving source image semantics. Such approaches generate target images explicitly or implicitly from the inversion noise of the source images, termed the inversion anchors. We identify this strategy as suboptimal for semantic preservation and inefficient due to elongated editing paths. We propose TweezeEdit, a tuning- and inversion-free framework for consistent and efficient image editing. Our method addresses these limitations by regularizing the entire denoising path rather than relying solely on the inversion anchors, ensuring source semantic retention and shortening editing paths. Guided by gradient-driven regularization, we efficiently inject target prompt semantics along a direct path using a consistency model. Extensive experiments demonstrate TweezeEdit's superior performance in semantic preservation and target alignment, outperforming existing methods. Remarkably, it requires only 12 steps (1.6 seconds per edit), underscoring its potential for real-time applications.

cs.CV↗
arXiv2025

SwitchLingua: The First Large-Scale Multilingual and Multi-Ethnic Code-Switching Dataset

Code-switching (CS) is the alternating use of two or more languages within a conversation or utterance, often influenced by social context and speaker identity. This linguistic phenomenon poses challenges for Automatic Speech Recognition (ASR) systems, which are typically designed for a single language and struggle to handle multilingual inputs. The growing global demand for multilingual applications, including Code-Switching ASR (CSASR), Text-to-Speech (CSTTS), and Cross-Lingual Information Retrieval (CLIR), highlights the inadequacy of existing monolingual datasets. Although some code-switching datasets exist, most are limited to bilingual mixing within homogeneous ethnic groups, leaving a critical need for a large-scale, diverse benchmark akin to ImageNet in computer vision. To bridge this gap, we introduce \textbf{LinguaMaster}, a multi-agent collaboration framework specifically designed for efficient and scalable multilingual data synthesis. Leveraging this framework, we curate \textbf{SwitchLingua}, the first large-scale multilingual and multi-ethnic code-switching dataset, including: (1) 420K CS textual samples across 12 languages, and (2) over 80 hours of audio recordings from 174 speakers representing 18 countries/regions and 63 racial/ethnic backgrounds, based on the textual data. This dataset captures rich linguistic and cultural diversity, offering a foundational resource for advancing multilingual and multicultural research. Furthermore, to address the issue that existing ASR evaluation metrics lack sensitivity to code-switching scenarios, we propose the \textbf{Semantic-Aware Error Rate (SAER)}, a novel evaluation metric that incorporates semantic information, providing a more accurate and context-aware assessment of system performance.

cs.CL↗
arXiv2024

Chain of Attack: On the Robustness of Vision-Language Models Against Transfer-Based Adversarial Attacks

Pre-trained vision-language models (VLMs) have showcased remarkable performance in image and natural language understanding, such as image captioning and response generation. As the practical applications of vision-language models become increasingly widespread, their potential safety and robustness issues raise concerns that adversaries may evade the system and cause these models to generate toxic content through malicious attacks. Therefore, evaluating the robustness of open-source VLMs against adversarial attacks has garnered growing attention, with transfer-based attacks as a representative black-box attacking strategy. However, most existing transfer-based attacks neglect the importance of the semantic correlations between vision and text modalities, leading to sub-optimal adversarial example generation and attack performance. To address this issue, we present Chain of Attack (CoA), which iteratively enhances the generation of adversarial examples based on the multi-modal semantic update using a series of intermediate attacking steps, achieving superior adversarial transferability and efficiency. A unified attack success rate computing method is further proposed for automatic evasion evaluation. Extensive experiments conducted under the most realistic and high-stakes scenario, demonstrate that our attacking strategy can effectively mislead models to generate targeted responses using only black-box attacks without any knowledge of the victim models. The comprehensive robustness evaluation in our paper provides insight into the vulnerabilities of VLMs and offers a reference for the safety considerations of future model developments.

cs.CV↗
arXiv2024

NFTracer: Tracing NFT Impact Dynamics in Transaction-flow Substitutive Systems with Visual Analytics

Impact dynamics are crucial for estimating the growth patterns of NFT projects by tracking the diffusion and decay of their relative appeal among stakeholders. Machine learning methods for impact dynamics analysis are incomprehensible and rigid in terms of their interpretability and transparency, whilst stakeholders require interactive tools for informed decision-making. Nevertheless, developing such a tool is challenging due to the substantial, heterogeneous NFT transaction data and the requirements for flexible, customized interactions. To this end, we integrate intuitive visualizations to unveil the impact dynamics of NFT projects. We first conduct a formative study and summarize analysis criteria, including substitution mechanisms, impact attributes, and design requirements from stakeholders. Next, we propose the Minimal Substitution Model to simulate substitutive systems of NFT projects that can be feasibly represented as node-link graphs. Particularly, we utilize attribute-aware techniques to embed the project status and stakeholder behaviors in the layout design. Accordingly, we develop a multi-view visual analytics system, namely NFTracer, allowing interactive analysis of impact dynamics in NFT transactions. We demonstrate the informativeness, effectiveness, and usability of NFTracer by performing two case studies with domain experts and one user study with stakeholders. The studies suggest that NFT projects featuring a higher degree of similarity are more likely to substitute each other. The impact of NFT projects within substitutive systems is contingent upon the degree of stakeholders' influx and projects' freshness.

cs.CE↗
arXiv2024

Efficient semi-supervised inference for logistic regression under case-control studies

Semi-supervised learning has received increasingly attention in statistics and machine learning. In semi-supervised learning settings, a labeled data set with both outcomes and covariates and an unlabeled data set with covariates only are collected. We consider an inference problem in semi-supervised settings where the outcome in the labeled data is binary and the labeled data is collected by case-control sampling. Case-control sampling is an effective sampling scheme for alleviating imbalance structure in binary data. Under the logistic model assumption, case-control data can still provide consistent estimator for the slope parameter of the regression model. However, the intercept parameter is not identifiable. Consequently, the marginal case proportion cannot be estimated from case-control data. We find out that with the availability of the unlabeled data, the intercept parameter can be identified in semi-supervised learning setting. We construct the likelihood function of the observed labeled and unlabeled data and obtain the maximum likelihood estimator via an iterative algorithm. The proposed estimator is shown to be consistent, asymptotically normal, and semiparametrically efficient. Extensive simulation studies are conducted to show the finite sample performance of the proposed method. The results imply that the unlabeled data not only helps to identify the intercept but also improves the estimation efficiency of the slope parameter. Meanwhile, the marginal case proportion can be estimated accurately by the proposed method.

stat.ML↗
arXiv2023

Developing a Multilingual Dataset and Evaluation Metrics for Code-Switching: A Focus on Hong Kong's Polylingual Dynamics

The existing audio datasets are predominantly tailored towards single languages, overlooking the complex linguistic behaviors of multilingual communities that engage in code-switching. This practice, where individuals frequently mix two or more languages in their daily interactions, is particularly prevalent in multilingual regions such as Hong Kong, China. To bridge this gap, we have developed a 34.8-hour dataset of Mixed Cantonese and English (MCE) audio using our Multi-Agent Data Generation Framework (MADGF). We fine-tuned the open-source multilingual Automatic Speech Recognition (ASR) model, Whisper, with the MCE dataset, leading to impressive zero-shot performance. The traditional metrics overlook important factors such as latency in real-world applications and code-switching scenarios. We have introduced a novel evaluation metric called Fidelity to the Original Audio, Accuracy, and Latency (FAL). This metric aims to overcome the limitations of traditional metrics used to assess ASR systems.

cs.SD↗
arXiv2023

Test-Time Compensated Representation Learning for Extreme Traffic Forecasting

Traffic forecasting is a challenging task due to the complex spatio-temporal correlations among traffic series. In this paper, we identify an underexplored problem in multivariate traffic series prediction: extreme events. Road congestion and rush hours can result in low correlation in vehicle speeds at various intersections during adjacent time periods. Existing methods generally predict future series based on recent observations and entirely discard training data during the testing phase, rendering them unreliable for forecasting highly nonlinear multivariate time series. To tackle this issue, we propose a test-time compensated representation learning framework comprising a spatio-temporal decomposed data bank and a multi-head spatial transformer model (CompFormer). The former component explicitly separates all training data along the temporal dimension according to periodicity characteristics, while the latter component establishes a connection between recent observations and historical series in the data bank through a spatial attention matrix. This enables the CompFormer to transfer robust features to overcome anomalous events while using fewer computational resources. Our modules can be flexibly integrated with existing forecasting methods through end-to-end training, and we demonstrate their effectiveness on the METR-LA and PEMS-BAY benchmarks. Extensive experimental results show that our method is particularly important in extreme events, and can achieve significant improvements over six strong baselines, with an overall improvement of up to 28.2%.

cs.LG↗
arXiv2022

Asymptotic Statistical Analysis of $f$-divergence GAN

Generative Adversarial Networks (GANs) have achieved great success in data generation. However, its statistical properties are not fully understood. In this paper, we consider the statistical behavior of the general $f$-divergence formulation of GAN, which includes the Kullback--Leibler divergence that is closely related to the maximum likelihood principle. We show that for parametric generative models that are correctly specified, all $f$-divergence GANs with the same discriminator classes are asymptotically equivalent under suitable regularity conditions. Moreover, with an appropriately chosen local discriminator, they become equivalent to the maximum likelihood estimate asymptotically. For generative models that are misspecified, GANs with different $f$-divergences {converge to different estimators}, and thus cannot be directly compared. However, it is shown that for some commonly used $f$-divergences, the original $f$-GAN is not optimal in that one can achieve a smaller asymptotic variance when the discriminator training in the original $f$-GAN formulation is replaced by logistic regression. The resulting estimation method is referred to as Adversarial Gradient Estimation (AGE). Empirical studies are provided to support the theory and to demonstrate the advantage of AGE over the original $f$-GANs under model misspecification.

math.ST↗