arXiv subjects

Kun Wang

Publications and source records attributed to Kun Wang.

At least 37 records · Page 2Linked to original sources

arXiv2026

Optimal No-Regret Learning for Repeated Prophet Inequality

We study repeated prophet inequalities under prefix feedback. In each of $T$ rounds, a learner encounters fresh values drawn independently from $n$ boxes with unknown $[0,1]$-supported distributions in a fixed order and must irrevocably accept one, observing only the prefix up to its stopping box. Regret is measured against the optimal stopping policy that knows the distributions. We give an efficient algorithm achieving $\widetilde O(\sqrt{T})$ expected regret, matching the lower bound up to logarithmic factors. Our algorithm explores directly through near-optimal policies, combining empirical backward induction with box-specific reach bonuses. A relative-drop aggregation rule then exploits the nesting structure of observed prefixes to preserve exploration, thereby removing the polynomial dependence on the box number $n$. This resolves an open question posed by Liu et al. (2025).

cs.LG↗

arXiv2026

RAGCell: Retrieval-Augmented Generation as Supervision for Versatile Single-cell Analysis

Single-cell foundation models (scFMs) are transforming computational biology by enabling generalizable, task-agnostic representations for versatile single-cell analysis. Despite their progress in facilitating rapid deployment for downstream tasks, off-the-shelf scFMs still have some overlooked concerns: (I) (Pretraining Cost.) Pretrain-based scFMs necessitate pretraining on a vast volume of cells, rendering it draining resources in applications. (II) (Heterogeneous Gap.) Large Language Models (LLM)-based scFMs ignore the tremendous heterogeneous gap between LLM textual and raw cellular spaces, leading to insufficient capability when facing downstream tasks. To this end, we introduce RAGCell, a versatile single-cell analysis framework that achieves a double-win in both cost-effectiveness and high performance. The success of RAGCell lies in two key aspects: Leveraging LLMs to construct cell-level and feature-level knowledge databases, which serve as supervision signals for training the cell model and significantly reduce the training cost ($>$pretrain-based scFMs). Aligning cell representations with text embeddings from the bi-level knowledge databases, enabling knowledge transfer from textual spaces to cellular spaces and effectively mitigating the heterogeneous gap ($>$LLM-based scFMs). Through extensive experiments on six downstream single-cell analysis tasks, we demonstrate that RAGCell achieves outstanding performance compared to state-of-the-art scFMs while operating at less than $\sim$1/10 the cost of pretrain-based scFMs.

q-bio.GN↗

arXiv2026

Towards Evolving Context Parameterization for Large Language Models

Context parameterization enables large language models (LLMs) to internalize contexts into reusable model parameters, avoiding repeated processing across subsequent queries. However, existing methods typically assume static contexts and lack explicit mechanisms for distinguishing validity states under continual updates. To study this real-world scenario, we formalized the Memory Updating with Sequential Evolution (MUSE) task and constructed MUSE-bench to evaluate update incorporation and unaffected-information preservation. The resulting challenge requires preserving the global state while adjusting the contribution of memory evidence. Motivated by this, we proposed PLUME, a training-free method that constructs a global update representation, activates memory evidence to form a local parameter view, and adaptively integrates their predictions during decoding. Comprehensive evaluation on MUSE-bench demonstrated PLUME's effectiveness in sequential evolution settings, yielding relative improvements of 29.9% in average ROUGE-L Recall and 54.9% in LLM-as-a-Judge. Our codes are available at: https://github.com/xiaobingshi-LLM/PLUME.

cs.CL↗

arXiv2026

SaFeR-Steer: Evolving Multi-Turn MLLMs via Synthetic Bootstrapping and Feedback Dynamics

MLLMs are increasingly deployed in multi-turn settings, where attackers can escalate unsafe intent through the evolving visual-text history and exploit long-context safety decay. Yet safety alignment is still dominated by single-turn data and fixed-template dialogues, leaving a mismatch between training and deployment. To bridge this gap, we propose SaFeR-Steer, a progressive multi-turn alignment framework that combines staged synthetic bootstrapping with tutor-in-the-loop GRPO to train a single student under adaptive, on-policy attacks. We also introduce Trajectory-Consistent Summative Reward (TCSR), which aggregates the historical minimum and average of turn rewards so that any low-quality turn affects the trajectory-level return. I. Dataset. We release STEER, a multi-turn multimodal safety dataset with STEER-SFT (12,934), STEER-RL (2,000), and STEER-Bench (3,227) dialogues spanning 1-10 turns. II. Experiment. Starting from Qwen2.5-VL-3B/7B, SaFeR-Steer substantially improves Safety/Helpfulness on both single-turn (48.30/45.86 $\rightarrow$ 81.84/70.77 for 3B; 56.21/60.32 $\rightarrow$ 87.89/77.40 for 7B) and multi-turn benchmarks (12.55/27.13 $\rightarrow$ 55.58/70.27 for 3B; 24.66/46.48 $\rightarrow$ 64.89/72.35 for 7B), shifting failures to later turns and yielding robustness beyond scaling alone. Code is available at https://github.com/Ed-Bg/SaFeR-Steer-full

cs.LG↗

arXiv2026

OSIRIS-REx Returned a Pristine Sample of Asteroid Bennu: Takeaways from the Mission's Contamination Control and Knowledge Program

NASA's OSIRIS-REx mission had the objective of delivering a pristine sample from asteroid (101955) Bennu to Earth for scientific analysis--where "pristine" signifies the absence of foreign materials that could affect sample measurements. OSIRIS-REx returned 121.6 g of regolith in September 2023; this study documents the systematic investigation of suspected contaminants encountered during the mission's sample analysis phase. Most suspected contaminants did not originate from the spacecraft or sample curation. Some were introduced during laboratory analyses, reinforcing the importance of procedural blanks and a strategic approach to sharing samples across laboratories with different analytical targets. Several suspected contaminants, such as phosphate and sodium fluoride particles, were ultimately identified as indigenous to Bennu, highlighting the critical role of contamination knowledge in preventing the dismissal of valuable scientific data. We find that the returned sample meets the definition of pristine, except for an isolated 1.24% (by mass) that escaped the sample container and was thereby contaminated with spacecraft particulates. These findings demonstrate the effectiveness of systematically applied contamination science and engineering practices and provide lessons and approaches to help maximize the scientific integrity of future planetary sample return missions.

astro-ph.EP↗

arXiv2026

In-Place Instruction Following in Diffusion Language Models

Diffusion Large Language Models (dLLMs) generate text via bidirectional iterative denoising, naturally supporting user-specified constraints anchored at arbitrary output positions, a paradigm known as In-place Prompting (IPP). We formalize this as the In-place Instruction Following (IIF) task and construct IIF-Bench, a hierarchical benchmark spanning literal, style, and discourse-function constraints, paired with a rubric-based local-global evaluation protocol. An inference-time attention-bias probe suggests that vanilla dLLMs often under-prioritize constraint spans during denoising. We then propose GRAFT, an IPP-oriented post-training framework combining constraint-aware SFT and preference optimization. On four representative dLLMs, GRAFT raises the average IIF score from 57.75 to 73.10 (+15.35 points), with absolute gains of 15.91 and 15.57 points on literal and discourse-function constraints, while preserving general generation ability.

cs.CL↗

arXiv2026

Train at Moving Edge: Online-Verified Prompt Selection for Efficient RL Training of Large Reasoning Model

Reinforcement learning (RL) has become essential for post-training large language models (LLMs) in reasoning tasks. While scaling rollouts can stabilize training and enhance performance, the computational overhead is a critical issue. In algorithms like GRPO, multiple rollouts per prompt incur prohibitive costs, as a large portion of prompts provide negligible gradients and are thus of low utility. To address this problem, we investigate how to select high-utility prompts before the rollout phase. Our experimental analysis reveals that sample utility is non-uniform and evolving: the strongest learning signals concentrate at the ``learning edge", the intersection of intermediate difficulty and high uncertainty, which shifts as training proceeds. Motivated by this, we propose HIVE (History-Informed and online-VErified prompt selection), a dual-stage framework for data-efficient RL. HIVE utilizes historical reward trajectories for coarse selection and employs prompt entropy as a real-time proxy to prune instances with stale utility. By evaluating HIVE across multiple math reasoning benchmarks and models, we show that HIVE yields significant rollout efficiency without compromising performance.

cs.LG↗

arXiv2026

MAOL: Morphology-Aware Ordinal Learning for Fine-Grained Industrial Defect Severity Grading

Fine-grained defect severity grading is essential for industrial inspection, yet remains challenging due to the ordinal nature of severity labels, the strong dependence on morphology-related cues, and the train-test discrepancy between clean annotated instances and noisy predicted instances in two-stage pipelines. We propose MAOL, a Morphology-Aware Ordinal Learning framework for fine-grained industrial defect severity grading. MAOL formulates severity grading as an instance-level ordinal learning task, incorporates explicit morphological features to enhance representation learning, introduces class-conditional adaptive ordinal thresholds to model defect-specific grading boundaries, and employs prediction-aware training via localization perturbation to improve robustness to imperfect predicted instances. Extensive experiments under both clean-ROI and predicted-instance settings demonstrate that MAOL consistently outperforms rule-based methods, nominal classification models, and existing ordinal baselines, especially in the predicted-instance setting. The proposed approach ranked third in the IDA 2026 Challenge on Fine-Grained Severity Grading for High-Precision Manufacturing.

cs.CV↗

arXiv2026

Five-Structures Preserving Algorithm for charge dynamics model

This paper develops a family of fast, structure-preserving numerical algorithms for the nonlinear Maxwell-Ampere Nernst-Planck equations. For the first-order scheme, the Slotboom transformation rewrites the Nernst-Planck equation to enable positivity preservation. The backward Euler method and centered finite differences discretize the transformed system. Two correction strategies are introduced: one enforces Gauss's law via a displacement correction, and the other preserves Faraday's law through potential reconstruction. The fully discrete scheme exactly satisfies mass conservation, concentration positivity, energy dissipation, Gauss's law, and Faraday's law, with established error estimates. The second-order scheme adopts BDF2 time discretization while retaining the same structure-preserving strategies, exactly conserving mass, Gauss's law, and Faraday's law. Numerical experiments validate both schemes using analytical solutions, confirming convergence orders and positivity preservation. Simulations of ion transport with fixed charges demonstrate exact preservation of Gauss's and Faraday's laws over long-time evolution, reproducing electrostatic attraction, ion accumulation, and electric field screening. The results fully support the theoretical analysis and the schemes' stability and superior performance.

math.NA↗

arXiv2026

GhostTac: Manipulating Tactile Sensors without Physical Contact

Tactile sensors are integral components of modern robotic systems, enabling robots to perceive and interact with the physical environment through tactile feedback. Despite their importance, the physical-layer security of tactile sensors has received little attention in prior work. In this paper, we present GhostTac, to the best of our knowledge, the first contactless attack that manipulates tactile sensing via electromagnetic interference (EMI). We identify that EMI exploits the nonlinear rectification and limited bandwidth amplification effects, allowing carefully crafted EMI signals to be converted into a persistent DC offset that bypasses on-board filtering and induces stable measurement deviations. Building on this mechanism, GhostTac enables fine-grained and controllable manipulation of sensor outputs by reshaping the spatial distribution and manipulating the magnitude at the targeted location. Such interference can induce unintended and harmful robot behaviors, such as causing a domestic robot to exert excessive force, resulting in physical damage or human injury. We evaluate GhostTac on 10 sensor modules and 2 dexterous hands, covering 15 tactile sensors of different types, and demonstrate consistent attack effectiveness across all tested devices. We further present three case studies on tactile grasping, slip detection, and material classification to illustrate practical impacts in real robotic tasks. We envision that our findings shed light on a new physical attack vector against tactile sensing in robotic systems.

cs.CR↗

arXiv2026

Graph Engineering in the Era of LLM Agents: From Individual Intelligence to System Intelligence

LLMs have evolved from language generators to autonomous agents capable of complex, long-horizon tasks. This evolution has produced paradigms including Prompt Engineering to elicit model capabilities, Context Engineering to manage information access, Harness Engineering to organize external tools and resources, and Loop Engineering to support continual reflection and self-improvement. Yet as tasks grow more complex, individual intelligence faces a fundamental limit: many tasks require heterogeneous expertise, interdependent subtasks, parallel execution, independent verification, and persistent state, exceeding any single agent's organizational capacity. Augmenting one agent's capabilities or context cannot resolve this architectural mismatch; intelligence must instead be distributed across specialized agents and organized at the system level. We call this System Intelligence: an agent system's ability to organize and coordinate multiple intelligent components into a coherent, adaptive whole pursuing a shared objective. Achieving it requires more than adding agents; it demands explicit structures to organize work, coordinate heterogeneous agents, and maintain evolving execution states. We introduce Graph Engineering, an emerging paradigm for next-generation agent systems. Unlike prior paradigms that mainly optimize individual interactions or agent-level behavior, Graph Engineering constructs explicit, dynamic, evolving graph structures representing tasks, agents, and system states. These abstractions provide a unified foundation for organizing complex objectives, orchestrating heterogeneous agents, modeling system dynamics, and enabling scalable agent evolution. We systematically review the principles, methodologies, and applications of Graph Engineering for LLM agents. Related papers, open-source data, and projects are collected at https://github.com/DEEP-JLU/Awesome-Graph-Engineering.

cs.IR↗

arXiv2026

Distance Is Not Enough: Forget-Retain Alignment Gap Predicts LLM Relearning Robustness

Machine unlearning aims to make a model forget specific data, yet unlearned LLMs often fail to stay unlearned: brief fine-tuning can revive removed knowledge. Existing robustness predictors rely on global weight-space displacement, but distance alone can be misleading when random or destructive updates collapse performance. We argue that relearning robustness depends on update structure: robust unlearning should affect forget-critical weights while sparing retain-critical ones. We introduce the Forget-Retain Alignment Gap (FRAG), a training-free predictor that scores an update's forget-retain alignment without running a relearning attack, and separates selective from dense updates more reliably than global distance. Building on the forget-critical, retain-sparing principle, Forget-Retain Pruning (FRP) improves relearning robustness. Our results suggest that weight selectivity better explains robustness than distance alone. Code is available at https://github.com/Yi1-Chen/FRAG.

cs.AI↗

arXiv2026

SkillNet: Create, Evaluate, and Connect AI Skills

Current AI agents can flexibly invoke tools and execute complex tasks, yet their long-term advancement is hindered by the lack of systematic accumulation and transfer of skills. Without a unified mechanism for skill consolidation, agents frequently ``reinvent the wheel'', rediscovering solutions in isolated contexts without leveraging prior strategies. To address this challenge, we introduce SkillNet, an open infrastructure for creating, evaluating, and organizing AI skills at scale. SkillNet structures skills within a unified ontology that supports creating skills from heterogeneous sources, establishing rich relational connections, and performing multi-dimensional evaluation across Safety, Completeness, Executability, Maintainability, and Cost-awareness. Our infrastructure integrates a repository of over 600,000 skills, an interactive platform, and a versatile Python toolkit. Experiments on ALFWorld, WebShop, and ScienceWorld show 40% higher average rewards and 30% fewer execution steps across multiple backbone models. Furthermore, SkillNet-Gym benchmarks skill retrieval, utilization, and composition, while SkillNet-Fabric enables task-specific skill routing through lightweight Wikis. By formalizing skills as evolving, composable assets, SkillNet provides a robust foundation for agents to move from transient experience to durable mastery.

cs.AI↗

arXiv2026

Mint-Agent: Introducing Finance-Native Agentic Foundation Models

Financial agents must do more than recall domain knowledge: they must be both reliable, executing precise operations over grounded evidence, and executive, sustaining long-horizon research whose conclusions remain auditable. We present Mint-Agent, a family of finance-native agentic models designed around these two scales of financial intelligence. Mint-Agent is built upon three pillars: data, harness, and algorithm. Our data engine constructs clean, specialized tasks for atomic financial capabilities and long-horizon agentic execution from real-world financial sources. MintHarness enables stable interaction with open-ended environments and maintains auditable evidence trails across extended research trajectories. Our training recipe combines SFT, critical-step OPD, and RLVR to develop separate financial reasoning and agentic execution experts, which are then unified through model merging and multi-teacher on-policy distillation into compact, general-purpose financial agents. This pipeline yields two flagship models, Mint-Cu (9B) and Mint-Ag (27B). Across professional financial benchmarks, our models demonstrate two defining strengths: (1) Reliability: Mint-Ag achieves 98.33% on RFC-Bench, surpassing GPT-5.6-Sol and Claude-Opus-4.8 by 3.66 and 3.00 points; and (2) Executability: Mint-Cu reaches 69.86% on FinSearchComp T2, outperforming Agents-A1-35B and Nex-N2-mini by 22.83 and 12.78 points, while Mint-Ag achieves 76.00% and 60.49% on FinanceAgentBench v1.1 and v2, respectively. These results establish a path toward trustworthy financial intelligence in which domain expertise, long-horizon execution, and auditable evidence are jointly engineered as a unified foundation for frontier agentic models.

cs.CL↗

arXiv2026

TINA+: Probing Residual Visual Knowledge in Unlearned Diffusion Models via Diffusion-Consistent Text-Free Inversion

Although text-to-image diffusion models exhibit remarkable generative power, concept erasure techniques are essential for preventing harmful content. Existing adversarial probes evaluate these methods by testing whether erased concepts can still be recovered. However, existing erasure and probe methods remain largely text-centric, focusing on whether the text-to-image mapping is severed while overlooking whether the corresponding visual knowledge remains. To investigate this question from a visual perspective, we leverage diffusion inversion to probe whether a generative trajectory can reconstruct visual instances of an erased concept. Under a null-text condition, standard inversion avoids the textual pathway but amplifies approximation errors, hindering faithful trajectory recovery. To address this challenge, we introduce TINA+, a diffusion-consistent Text-free INversion Attack equipped with optimization-based inversion. We also find that unconstrained diffusion inversion may discover spurious trajectories, even allowing a randomly initialized diffusion model to reconstruct the target concept. Such trajectories may falsely indicate residual visual knowledge. TINA+ therefore introduces Diffusion-Consistent Trajectory Regularization to suppress this failure mode. By penalizing trajectories that fall far below the expected marginal energy evolution of diffusion, TINA+ suppresses spurious inversion paths while preserving its ability to recover erased concepts. Experiments across twelve erasure methods, four concept-erasure tasks, and different model architectures demonstrate that TINA+ reliably probes residual visual knowledge through diffusion-consistent visual trajectories. These results provide stronger evidence that current methods often obscure concepts by severing text-image links rather than eliminating the underlying visual knowledge.

cs.CV↗

arXiv2026

Assessing Parameter Redundancy in Transformers for Jet Tagging

Transformer-based jet taggers, such as the Particle Transformer (ParT) and the More-Interaction Particle Transformer (MIParT), achieve excellent discrimination by exploiting correlations among jet constituents, but often require more trainable parameters than earlier deep-learning taggers. In this paper, we investigate whether comparable discriminating power can be achieved with substantially fewer parameters. We introduce an hourglass structure that replaces the feed-forward networks (FFNs) in the attention blocks while leaving the particle-interaction attention unchanged. We also introduce a lightweight particle-embedding layer to replace the original dense embedding network. Applying both modifications to ParT and MIParT yields the hourglass (HG) variants ParT-HG and MIParT-HG, respectively. We evaluate both models on benchmark datasets for top tagging and quark-gluon discrimination. Both variants retain comparable tagging performance, including background rejection at fixed signal efficiencies, while using only approximately 48% and 39.7% of the parameters of their respective baselines. On the larger JetClass dataset, accuracy and AUC decrease by less than 1%, and background rejection also decreases for several signal classes. Overall, our approach provides an alternative way to reduce the parameter count of Transformer jet taggers while largely retaining their tagging performance.

hep-ph↗

arXiv2026

Improving Deep Learning for Accelerated MRI With Data Filtering

Deep neural networks achieve state-of-the-art results for accelerated MRI reconstruction. Most research on deep learning based imaging focuses on improving neural network architectures trained and evaluated on fixed and homogeneous training and evaluation data. In this work, we investigate data curation strategies for improving MRI reconstruction. We assemble a large dataset of raw k-space data from 18 public sources consisting of 1.1M images and construct a diverse evaluation set comprising 48 test sets, capturing variations in anatomy, contrast, number of coils, and other key factors. We propose and study different data filtering strategies to enhance performance of current state-of-the-art neural networks for accelerated MRI reconstruction. Our experiments show that filtering the training data leads to consistent, albeit modest, performance gains. These performance gains are robust across different training set sizes and accelerations, and we find that filtering is particularly beneficial when the proportion of in-distribution data in the unfiltered training set is low.

eess.IV↗

arXiv2026

From Inaudible Inputs to Model Failures: Low-Frequency Safety Risks in LALMs

Large audio-language models (LALMs) have demonstrated strong capabilities in understanding diverse audio inputs. This diversity includes low-frequency signals that are inaudible to humans but can still enter the model and influence its generation. However, the practical impact of such low-frequency inputs on LALMs remains largely unexplored. In this paper, we propose Intermittent Low-Frequency Lockout (ILL), an inaudible red teaming method that evaluates this risk using a universal waveform template in a black box setting. ILL uses Sentence Attention Scale Estimation to determine active intervals and Frequency Confusion Transfer to construct a low-frequency waveform with continuous phase from corpus spectral variation. To mitigate this risk, we propose Distributional Requery Guard (DRG) to detect low-frequency distribution shifts and conditionally request a second recording for semantic recovery. Across six LALMs and multiple audio understanding tasks, ILL reduces accuracy by up to 67 percentage points while receiving a mean human audibility rating of 1.33, close to 1.17 for clean audio; DRG raises mean attacked accuracy from 28.5\% to 46.1\% after clean reacquisition. These findings identify a previously overlooked safety risk for LALMs and provide a foundation for future research on robust audio understanding.

cs.SD↗