Searcharxiv⌕ Search

arXiv subjects

Kunpeng Liu

Publications and source records attributed to Kunpeng Liu.

At least 19 recordsLinked to original sources
arXiv2026

AegisFlow: Training-Free Non-Myopic Path-Safe Guided Flow Matching

Pretrained generative models, such as diffusion and flow matching models, have demonstrated exceptional capabilities in synthesizing high-quality data. However, steering their generative trajectories to satisfy strict constraints--such as identity preservation in image editing or obstacle avoidance in robotics--remains fundamentally challenging. While recent training-free guidance methods enforce path-safety, they rely heavily on min-norm controllers that only account for immediate constraint satisfaction. These myopic approaches often yield suboptimal trajectories that hover dangerously close to the boundary, severely restricting the model's ability to optimize toward its target. In this paper, we propose AegisFlow, a training-free, non-myopic path-safe guidance framework that recasts guided generation as a safety-critical optimal control problem with global scope. By integrating a time-varying barrier function directly into the global cost functional, AegisFlow naturally enforces constraints along the entire generative path. To compute the global optimal control, we derive a tractable, mathematically grounded approximation of the Hamilton-Jacobi-Bellman (HJB) equation. Comprehensive experiments demonstrate that AegisFlow outperforms state-of-the-art methods across text-guided image manipulation and robotic planning, establishing a significantly better trade-off between target alignment and constraint satisfaction. Code is available at https://github.com/LiuKunpeng03/AegisFlow.

cs.AI↗
arXiv2026

BoardSafety: AI-Ready Data from Manufacturer Documentation for Development-Board Safety

The development of large language models (LLMs) are capable to assist with hardware experiments. In this scenario, this raises a practical question: Can LLMs determine whether a proposed operation is safe for a specific development board? Documented operating limits and safety requirements are board-specific, even across products from the same manufacturer. This variation makes board-specific safety judgments by an LLM difficult to evaluate. To address this problem, we introduce the BoardSafety Benchmark. The benchmark is built from vendor datasheets and hardware guides. For data transformation, these documents are converted into a board-specific safety rule base and are then used to create labeled operating scenarios. Each data item pairs a board and a proposed operation to an applicable safety rule with ground-truth label, indicating the level of safety. The categorized level for each operation includes safe, unsafe, or insufficient-info. In addition, the benchmark covers diverse operations, board-specific cases, boundary conditions, and missing-information cases. We evaluate a series of models by varying their access to board-specific rules and the use of structured-rationale prompting. Our results demonstrate that providing the safety rules generally improves accuracy, whereas structured-rationale prompting produces gains that vary across models and rule-access conditions. Overall, the evaluated models cannot yet reliably determine whether an operation is safe, which implies that critical exploration is required for this field.

cs.AI↗
arXiv2026

Generative Embodied Multiple Behavior Control Systems for Human-like Agents

Building autonomous agents that reproduce human behavior in realistic 3D environments has been a longstanding objective in AI. Cognitive neuroscience commonly believes that human behaviors are controlled by multiple control systems including goal-directed and habitual. However, existing human-like agent frameworks primarily focus on goal-directed system and habitual system is largely overlooked. In this paper, we address this gap by proposing GEMS, a Generative Embodied Multiple Behavior Control System that jointly models goal-directed and habitual behaviors for generating believable human-like behavior. GEMS employs a cognitive module to generate natural language behaviors and a motion generation module to construct them in 3D environment. The cognitive module comprises a Habitual Controller that retrieves habitual actions from habit memory in response to environmental stimuli, a Goal-directed Controller that proposes goal-directed actions for current goal, and an Arbiter that governs their influence and selects the final action, supporting diverse and naturalistic behavior patterns generation. To construct these behaviors, the motion generation module adopts a Text-to-Motion model to generate versatile and expressive 3D human-scene interactions without relying on predefined actions. Extensive evaluations, human and ablation studies demonstrate that human-likeness performance is substantially improved by GEMS. The efficacy of GEMS indicates the benefits of leveraging habitual behavior and multiple behavior control system coordination for believable embodied human-like agents design.

cs.AI↗
arXiv2026

StaRPO: Stability-Augmented Reinforcement Policy Optimization

Reinforcement learning (RL) is effective in enhancing the accuracy of large language models in complex reasoning tasks. Existing RL policy optimization frameworks rely on final-answer correctness as feedback signals and rarely capture the internal logical structure of the reasoning process. Consequently, the models would generate fluent and semantically relevant responses but logically inconsistent, structurally erratic, or redundant. To this end, we propose StaRPO, a stability-augmented reinforcement learning framework that explicitly incorporates reasoning stability into the optimization objective. Our StaRPO decomposes stability into two computable lightweight metrics: the Autocorrelation Function (ACF) to evaluate local step-to-step coherence, and Path Efficiency (PE) to evaluate global goal-directedness of the reasoning trajectory. These stability rewards are combined with task rewards to provide complementary and process-aware feedback. We validate the effectiveness of using ACF and PE rewards by showing their correlation with logic errors on two backbone models. Experiments on four reasoning benchmarks show that StaRPO consistently outperforms compared baselines and can enhance both final-answer accuracy and logical stability.

cs.AI↗
arXiv2026

DualOPSD: Closing the Feedback Loop in On-Policy Self-Distillation

On-policy self-distillation (OPSD) combines student generated trajectories with dense supervision from a privileged teacher of the same model. Its pointwise clipping protects the student from locally incompatible teacher directions, but the fixed teacher receives no feedback about the supervision the student rejects, which limits effective learning from the teacher. We introduce DualOPSD, which alternates updates in a coupled teacher-student system. The student retains OPSD's clipped forward KL objective. The privileged teacher adapts to the updated student after each student step. A residual target based on the frozen initial model preserves the effect of privileged context during teacher adaptation. Across Qwen3-8B/4B/1.7B and three math benchmarks, DualOPSD improves avg@12 accuracy over OPSD by 9.83 percentage points. Further diagnostic analyses show lower bidirectional KL, less persistent mismatch, and retained sensitivity to privileged reference. Together, these results demonstrate that DualOPSD can be a more effective form of privileged self-distillation. Code is available at: https://github.com/on1262/DualOPSD.

cs.LG↗
arXiv2026

Explore with Memory, Learn without It: In-Context Exploration and Self-Distillation for LLM Reasoning

Exploration bounds learning efficiency in reinforcement learning with verifiable rewards (RLVR). Current methods explore mostly through independent sampling, without fully using the information gathered along the way, such as known wrong answers and environment feedback. We propose LEAF, which injects a dynamic, problem-specific memory into the context before every rollout. The memory contains the most recent response, its reward, and other environment feedback. It is updated after every verifier call. We decouple policy updates from rollouts and update the policy through self-distillation. On four benchmarks and two open models, LEAF accelerates exploration by about 4x compared with a random sampling baseline. On unseen test sets, it improves accuracy over the strongest baseline by 1.59 and 1.83 percentage points on the two models, while using less compute than any RL baseline. These results highlight the potential of LEAF in improving both exploration and learning efficiency. Code is available at: https://github.com/on1262/LEAF

cs.LG↗
arXiv2026

More Reasoning Is Not Always Better: Executing Clinical Guidelines with LLMs and Rules

Large language models (LLMs) have shown considerable promise in clinical decision support. However, existing approaches often rely on multi-step reasoning or multi-agent collaboration to execute clinical guidelines, introducing substantial computational overhead without reliably preventing decision errors. To address these limitations, we propose a framework for Medical Decision Execution (MedDEx) that integrates structured patient states with deterministic rule execution to support guideline-based treatment planning. MedDEx separates clinical record interpretation from guideline execution, leveraging LLMs to extract patient information and generate treatment plans while assigning guideline-defined decision logic to a rule executor. Experiments across multiple clinical guideline benchmarks demonstrate that MedDEx improves decision accuracy while reducing inference costs. Further analyses highlight the critical roles of structured patient representations and deterministic rule execution, suggesting that additional LLM reasoning does not necessarily translate into better guideline adherence.

cs.CL↗
arXiv2026

Which and When to Admit: Gradient Admission for Data-Centric Small Language Model Finetuning

LoRA fine-tuning adapts small language models (SLMs) to heterogeneous instruction data within a low-rank update subspace, making it vulnerable to three structural problems: conflicting gradients that cancel, static data selection that cannot track evolving learning dynamics, and subspace saturation that causes later updates to overwrite useful directions. We argue that effective adaptation therefore requires controlling which data-induced gradients enter the LoRA subspace and when. We propose GRADE (GRadient-Aligned Data-centric rEcipe), a data-centric framework combining two mechanisms: a state-aware selector that continually admits samples aligned with the evolving multi-task gradient field, and a self-calibrating step-level gate that rejects updates likely to cause destructive overwrite near saturation. Across three current-generation backbones and a heterogeneous seven-dataset instruction pool, GRADE outperforms strong data-selection and PEFT-stabilization baselines in accuracy and robustness. It is the only method to improve consistently over standard LoRA on every architecture, while producing more coherent gradient trajectories and less destructive overwrite. These results show that successful SLM adaptation depends not only on which data are selected, but also on which gradients are allowed to enter and persist in the constrained update subspace.

cs.LG↗
arXiv2026

Rethinking Data Augmentation under Covariate Shift: Invariant-Guided Diffusion and Prototype Reweighting

In many industrial applications, 1) tabular data is scarce and imbalanced and thus requires synthetic expansion; 2) input distributions drift between training and deployment (covariate shift); 3) validation sets often diverge from unseen test environments; or 4) standard generative models simply mimic outdated source distributions. This learning setting limits the stability of standard augmentation and adaptation pipelines. We generalize the task under such setting as the Augmented and Weighted Learning under Covariate Shift problem (AWL-CS). AWL-CS imposes two critical challenges on existing methods: 1) misleading generative guidance where models optimize for source similarity rather than downstream task relevance, and 2) structural instability of distributional density where reweighting mechanisms overfit to noisy validation signals. To tackle these challenges, we propose IGDPR (Invariant-Guided Diffusion with Prototype Reweighting), a unified framework that synergizes stable synthesis and structural adaptation: i) To achieve task-relevant generation, we steer the diffusion sampling process using invariant potentials to ensure synthetic samples align with stable decision boundaries rather than outdated correlations. ii) To ensure stable adaptation, we develop a prototype-based reweighting strategy that assesses sample reliability through structural clusters instead of isolated points, effectively filtering validation noise. Extensive experiments on real data demonstrate our method improves data quality by augmenting the most beneficial data for robust learning.

cs.LG↗
arXiv2026

ElecSafety: Diagnosing Large Language Model Safety Judgments for Microcontroller Boards

Large language models (LLMs) progressively support embedded hardware development, but a plausible recommendation may pose safety risks when acting on the physical hardware. Because electrical safety can be different between every microcontroller board, a practical question is raised: can LLMs have the capability to make the right decision on whether a proposed user operation is safe for the board? Although benchmarks for embedded development exist, they primarily target code generation and hardware design tasks. This highlights the lack of tested board-specific judgment regarding electrical safety. To address this gap, we introduce the ElecSafety benchmark to evaluate whether an LLM can respond with the correct label and derive that label from the proper manufacturer constraints and rules. We extract these constraints from vendor datasheets and hardware manuals. Each scenario is defined by a decisive condition and a gold label (safe, hazardous, or cannot determine) before translating the scenario into natural language. We categorize the scenarios into violation and compliant cases, board-swap, near-boundary, and missing-information. We then evaluated the scenario on six open-weight LLMs under various conditions of rule access and token limitations. The finding shows board rules can improve accuracy, whereas limiting the token has a smaller impact. Even the best configuration remains unreliable on near-boundary and missing-information scenarios, and a correct label often rests on the inaccurate constraints.

cs.AR↗
arXiv2026

Flat-Consensus Diffusion for Robust Data Reshaping under Noisy Evaluator

Data shape determines how features are structured, how patterns are separated, and how distributions cover the underlying domain. Poor data shape can make models learn noise rather than generalizable structure. This paper studies robust feature-centric data reshaping: generating feature transformations that remain useful, stable, and reproducible under noisy evaluation and imperfect data conditions. We view reshaping operation sequence search as reward-guided diffusion generation, and robust reshaping as searching for regions in the latent reward landscape rather than isolated high-reward transformations. The key challenge is dual instability: noisy evaluators distort local reward guidance, while stochastic generative trajectories can converge to inconsistent solutions. We propose FCDiff, a flat-consensus diffusion framework that addresses both failures through a micro-macro decomposition. The micro layer replaces point-estimate reward guidance with Gaussian-smoothed, Monte Carlo averaged gradients, steering generation toward locally flat reward regions. The macro layer aggregates independently guided trajectories with a weighted Frechet-mean barycenter, selecting consensus-supported basins and filtering stochastic outliers. Across an 8-dataset headline cohort under heavy-tailed evaluator noise, FCDiff attains the best aggregate rank on lower-tail reliability and robustness against both search-based AutoFE and robustness-oriented generative baselines, with statistically significant accuracy gains over every generative baseline. Our results show that robust data reshaping requires searching for flat, consensus-supported regions rather than sharp single-trajectory optima.

cs.AI↗
arXiv2026

City Editing: Hierarchical Agentic Execution for Dependency-Aware Urban Geospatial Modification

Urban renewal requires incremental modifications to existing geospatial plans, yet manually updating complex layouts under spatial constraints is labor-intensive and error-prone. To tackle this, we propose CEAE, a hierarchical agentic framework that formulates urban renewal as machine-executable GeoJSON editing from natural-language instructions. CEAE decomposes instructions into hierarchical geometric intents, executing edits from coarse to fine while preserving spatial consistency through a self-reflective execution-validation loop. Experimental results show that CEAE outperforms baselines in execution validity, robustness, and geometric accuracy.

cs.MA↗
arXiv2026

From Reading Code to Reading Spec: A Verified Layer for LLM-Driven Codebase Maintenance

The rapid growth of LLM-generated code increases software complexity and the maintenance burden on engineers. While LLMs offer a potential automated alternative, this structural complexity hinders their ability to manage codebases directly. We introduce the Provable Representation Of Original Functionality (PROOF), which manages codebases indirectly via structured specifications. To enable full-lifecycle codebase management strictly through these specifications, PROOF abstracts codebase topology into a hierarchical natural-language representation. To establish absolute trust, the system proves semantic equivalence by reconstructing source code exclusively from this specification. This verified foundation drives maintenance requests, executing code modifications while synchronously updating itself to prevent semantic drift. Experiments on real-world repositories confirm the effectiveness of these specifications.

cs.SE↗
arXiv2026

Efficient Geothermal Well-Control Optimization via Diffusion-Surrogate Reinforcement Learning

Real-time decision-making for enhanced geothermal systems (EGS) is challenging because long-term production periods involve high-dimensional control spaces and a large number of time-consuming high-fidelity hydrothermal simulations. Reinforcement learning provides a natural framework for state-dependent sequential control, but direct policy training with numerical simulators is computationally expensive. To address this issue, we propose a diffusion-surrogate guided reinforcement learning framework for long-horizon EGS well-control optimization. The reservoir temperature and pressure fields are used as system states, while injection rates are selected as control actions. A learned surrogate environment is constructed using conditional diffusion models to predict the evolution of reservoir temperature and pressure fields and a separate reward model to estimate the corresponding economic return. The surrogate environment is then integrated with Proximal Policy Optimization (PPO) for efficient policy training. Experiments on a fractured EGS benchmark show that the diffusion surrogate can accurately reproduce reservoir-state evolution over multiple control stages. The resulting surrogate-assisted PPO policy achieves competitive well-control performance compared with direct simulator-based PPO and existing optimization methods, while substantially reducing the dependence on expensive high-fidelity simulations. These results demonstrate the potential of diffusion-based surrogate environments for efficient reinforcement learning in geothermal well-control optimization.

cs.AI↗
arXiv2026

BrailleBench: Investigating Multi-Criteria Braille Comprehension in Large Language Models

Although Large language models (LLMs) mediate access to knowledge and computational assistance, their capabilities should benefit vulnerable groups in the same way. However, it is unclear whether existing AI systems are inclusive enough for blind and deafblind users to access the same functionality through Braille, whose indicators, contractions, and digital representations introduce distinct requirements for model comprehension. To this end, we introduce BrailleBench, a benchmark for evaluating LLMs in Braille comprehension from different Criteria. BrailleBench aligns 5,570 instances from five datasets, including mathematics, commonsense, and multi-hop question answering across English and Braille Grades 1 and 2. Different configurations are designed to understand whether the systems can comprehend Braille-authored content, express answers in Braille, and complete end-to-end Braille interaction. To ensure the quality and prevent evaluation bias, the benchmark is built through a deterministic, expert-reviewed pipeline via a self-created Braille Toolkit without using any data instances generated by LLMs. We evaluate six representative LLMs from various aspects. The results reveal a persistent gap between print-English capability and Braille accessibility. Braille understanding and expression are asymmetric, where Grade 2 is especially fragile on the input side compared to Grade 1, and fully Braille requests further reduce performance. The experimental observations provide valuable guidance for the development of future Braille AI systems. All related resources in BrailleBench are publicly available for future research.

cs.AI↗
arXiv2026

Visual Representation and History Modeling for Navigation World Models

Navigation World Models (NWMs) predict action-conditioned visual futures for planning. Two practical challenges are central to their design: selecting a suitable visual representation and efficiently modeling observation history for repeated candidate queries. Standard Global-Softmax attention provides flexible interactions but repeatedly processes the same history, leading to increasing computation and memory costs for long contexts and multi-query planning. We study both problems within a unified conditional flow-transformer framework. We first compare five frozen visual representations under the same dynamics model and evaluation. To reduce redundant history computation, we design Cached-Linear, a hybrid architecture that combines local and shifted-window attention for target mixing with linear attention for reusable history access. We further develop Balanced Gated Delta Network (GDN), which augments this design with frame-wise recurrent memory for temporal history modeling. Experiments on RECON, SACSoN, and SCAND show that representation choice depends on the prediction objective: PAE-L performs best for reconstruction, RAE-B for direct prediction, and V-JEPA for long-horizon rollout. Under shared-history workloads, Cached-Linear substantially reduces computation and memory compared with Global-Softmax, while Balanced GDN improves selected direct-prediction endpoints with efficient context reuse. Overall, we systematically study visual representation and history modeling for NWMs and develop hybrid reusable-history architectures for efficient long-context and multi-query prediction.

cs.CV↗
arXiv2026

DeltaFlow: Noise-Adaptive Bidirectional Gated Delta Networks for Embedded Language Flows

Embedded Language Flows (ELF) rely primarily on full non-causal attention for iterative denoising, repeatedly incurring quadratic sequence-mixing cost at each sampling step. Gated Delta Networks (GDNs) provide an efficient recurrent alternative, but their standard causal formulation cannot directly capture the bidirectional context required by ELF. We introduce DeltaFlow, a noise-adaptive bidirectional GDN backbone for continuous language denoising. We study two variants: DeltaFlow-A, which alternates scan directions across layers, and DeltaFlow-P, which performs parallel forward and backward scans within each layer. We further introduce noise-adaptive memory control and scheduled Temporal State Consistency (TSC) to stabilize hidden representations across nearby noise levels. On OpenWebText, using a 32-step stochastic differential equation sampler, DeltaFlow-P reduces generated perplexity from 24.218 for the full-attention ELF baseline to 21.228 while maintaining comparable unigram entropy, with 36B training-token exposure compared with 45B for the baseline. In a denoiser-only benchmark, DeltaFlow-P achieves a 2.72x throughput speedup over the full-attention baseline at a sequence length of 16k. These results show that DeltaFlow is a promising alternative to dense attention for efficient continuous language denoising.

cs.CL↗
arXiv2026

Mind your key: An Empirical Study of LLM API Credential Leakage in iOS Apps

The rapid integration of large language models (LLMs) into mobile applications has introduced a new class of credential security risk: leaked credentials that grant unauthorized access to LLM inference services, causing financial damage to developers. Prior work on credential leakage has focused primarily on Android apps; to date, no empirical study has systematically investigated LLM API key leakage in iOS applications. We present the first in-depth empirical study of API key leakage in LLM-integrated apps. We construct a high-quality dataset of 444 iOS applications, filtered from 1092 candidates through a standardized process, and develop LLMKeyLens, a dynamic analysis framework that detects LLM API key leakage via traffic interception, provider-specific key extraction, and active validity confirmation, requiring neither source code access nor binary decryption. Our analysis reveals that 282 applications expose exploitable LLM API credentials in network traffic, spanning at least ten providers. We identify three leakage patterns: JWT-based token leakage (48%), unauthenticated backend proxy access (33%), and plaintext API key transmission (19%). To assess remediation, we re-analyzed the same 282 vulnerable applications three months after responsible disclosure; only 28% had remediated the reported vulnerability, while 72% remained exploitable, with persistent issues stemming from unauthenticated backends and broken JWT implementations. Our findings show that LLM API key leakage is both prevalent and persistent in the iOS ecosystem, exposing a systemic gap between developer practice and secure integration principles, and suggest that secure LLM integration requires not only developer awareness but also explicit security guidance from providers and platform-level enforcement.

cs.SE↗