Searcharxiv⌕ Search

arXiv subjects

Yixin Yang

Publications and source records attributed to Yixin Yang.

At least 19 recordsLinked to original sources

Flash evaporation Riemann Problem: Thermodynamic Relaxation and Oscillation-Free Real Fluid Finite Volume Method

The Flash evaporation Riemann problem (FeRP) formulated the wave structures of flash evaporation in the vapor-liquid homogeneous equilibrium limit. However, physical flash evaporation involves finite-rate interphase heat and mass transfer, inducing the wave structure to deviate from the equilibrium state. To solve such a non-equilibrium Riemann problem, this paper first derives the governing equation of mechanical equilibrium evolution in a relaxation system, based on which an oscillation-free Homogeneous Relaxation Real Fluid (HRRF) finite volume method is developed. In addition to the Euler equations with mass relaxation, the HRRF method evolves advection equations for the internal energy--pressure affine parameters and the vapor volume fraction, in order to eliminate numerical pressure oscillations inherent in conservative methods while preserving phase-transition-induced pressure changes, and consistently degenerate into a completely pressure-equilibrium method in supercritical flows. The HRRF method is then employed to comprehensively evaluate how thermodynamic relaxation affects the FeRP solution. Results indicate that the non-classical expansion shock of the FeRP loses its traveling-wave characteristics with increasing relaxation time, continuously degenerating into a self-similar rarefaction wave. Compared with the equilibrium limit, an increased mass transfer relaxation time smooths the expansion branch of the RP solution, lowers the intermediate pressure and velocity, and induces a pseudo-splitting of the rarefaction wave as well as a pseudo-advance of the phase transition onset. Meanwhile, the heat transfer relaxation time primarily dictates the temperature difference in the vapor-liquid mixture, thereby modulating the liquid superheat and in turn influencing the mass transfer rate.

physics.flu-dyn↗

From Signals to Evidence: Interpretable Underwater Acoustic Target Sensing with Large Language Models - Part II: Passive Detection

Underwater acoustic passive detection is essential to maritime surveillance and intelligent ocean sensing. Existing detection systems typically collapse multichannel array observations into a binary decision or a scalar confidence score, offering limited insight into the physical evidence that supports or contradicts a target declaration. This limitation is acute at low signal-to-noise ratio (SNR), where isolated spectral peaks and transient beamforming responses are inherently ambiguous and physically verifiable targets are characterized by an ensemble of coupled reinforcing cues. Building on the unified from-signals-to-evidence framework introduced in the companion Part I, this paper instantiates the framework for passive target detection. A physics-aware multichannel acoustic encoder encodes the array observations into complementary acoustic evidence tokens through a direction-guided architecture that combines minimum variance distortionless response (MVDR) processing with time-frequency and fine-frequency representations. These tokens are combined with environmental priors and fed into a pretrained large language model (LLM) that evaluates target-present and noise-only hypotheses and generates a structured detection report detailing target presence, bearing, spectral-line locations, received SNR, range, and per-field confidence scores. A physics-parameterized data engine extended to realistic oceanic channels provides traceable supervision at both discrete-label and structured-description levels. Experiments under white-noise and mixed ocean-background conditions demonstrate reliable detection, accurate bearing estimation, and physically interpretable structured reports, validating the feasibility of evidence-grounded underwater acoustic detection.

cs.SD↗

Stress-Testing Structure-Aware Calibration of Malware Graph Neural Networks under Type Shift

Post-hoc malware calibrators can condition confidence on graph structure, but their structural inputs may leave the support represented by validation data under malware-type shift. We study this risk on MalNet-Tiny by holding out each of four malware types across three seeds, freezing a graph isomorphism network, and fitting post-hoc mappings only on known-type data. On known-type tests, generic and combined structural calibrators have negative log likelihood (NLL) of 0.4341 and 0.4628. Their mean held-out-type gap widens to 0.1369, with a four-type 95% interval of [-0.0125, 0.2864] and improvement in one of four types. Eight label-independent nuisance variables produce a smaller mean NLL increase of 0.0415, but the paired combined-versus-nuisance interval includes zero. We then use calibration data to define a community-support guard: predictions outside the 95th percentile of calibration-standardized community displacement revert to the generic calibrator. The guard reduces combined NLL by 0.1133 and high-confidence errors from 43.25 to 33.25 per 400-sample cell. It also outperforms matched-rate random fallback on the original folds, while generic and uncalibrated probabilities retain lower held-out-type NLL. This exploratory recovery localizes harmful combined predictions; the stress test exposes heterogeneous calibration failures without establishing a general structural-calibration penalty or a deployment gain over simpler alternatives.

cs.CR↗

Token-Budgeted Escalation for Financial Document Question Answering: Cost Is Predictable, Benefit Is the Bottleneck

Retrieval-augmented generation systems can route difficult queries to deeper context, but batch deployments must allocate a shared token budget across calls whose costs vary by query. We formulate selective escalation as finite-batch allocation for financial document question answering. Each of 150 FinanceBench questions first receives a top-1 retrieval answer. Predictors estimate the adjudication-quality gain and token cost of an optional top-5 call, and the allocator prioritizes calls by predicted gain per token. At the nominal 10% budget, gain-per-token allocation improves adjudication quality over gain-only ranking by 0.034 (95% document-bootstrap confidence interval [0.001, 0.072]) while using 46.6% fewer total tokens than one-pass top-5 retrieval. Additional-call cost is accurately predictable (R-squared 0.93), whereas beneficial escalation remains difficult to rank (area under the receiver operating characteristic curve 0.60). These results show that heterogeneous cost is actionable under tight constraints, while progress across the full budget frontier depends on stronger query-specific benefit estimates.

cs.IR↗

From Signals to Evidence: Interpretable Underwater Acoustic Target Sensing with Large Language Models - Part I: Framework and Recognition

Underwater acoustic target detection and recognition are essential to marine surveillance and intelligent ocean sensing. Despite substantial advances in deep learning, existing systems typically produce binary detection decisions, categorical labels, or scalar confidence scores, offering limited insight into the physical evidence underlying their outputs. Large language models (LLMs) offer an opportunity to move underwater acoustic sensing toward interpretable, evidence-grounded understanding; however, generic audio encoders struggle to capture the distinctive physical characteristics of underwater acoustics, and physically annotated data are scarce. In this paper, the first of a two-part series, we propose a from-signals-to-evidence framework that integrates physics-aware acoustic encoding with a pretrained LLM, organized around a unified feature-evidence-decision chain with a shared reasoning core. An open input layer accommodates heterogeneous information-raw waveforms, time-frequency representations, environmental and geographic data, and textual physical priors-unifying detection and recognition within a common formulation. In Part I, we instantiate this framework for target recognition through an Underwater Acoustic Physics-aware Transformer (UPT), which tokenizes complementary acoustic representations and learns task-disentangled physical representations through four physics-oriented learning tasks. We develop a physics-parameterized data engine that provides traceable supervision at two levels of granularity and introduce an attribute-level evaluation protocol to assess the physical fidelity of generated descriptions. Experiments on simulated recordings of four vessels demonstrate accurate vessel recognition and reliable estimation of shaft frequency, blade-passing frequency, and machinery-related attributes.

cs.SD↗

From Candidate Tokens to Final Outcomes: Decoding Relative Long-Horizon Candidate Value from Reasoning Model Representations

Large reasoning models generate reasoning trajectories token by token. At a given prefix, multiple next-token candidates may be plausible, yet shape subsequent reasoning and differ in their chances of leading to the correct answer. Whether these differences are reflected in internal representations remains unclear. To study this question, we define relative long-horizon candidate value to quantify within-prefix differences and construct the Reasoning Candidate-Token Value (RCTV) dataset, pairing mathematical reasoning prefixes with plausible next-token candidates and their empirical long-horizon candidate values. Using RCTV, we examine hidden states, multilayer perceptron outputs, and attention outputs at candidate positions with sparse linear probes. Experimental results provide evidence that relative long-horizon candidate value is linearly decodable from these representations at multiple layers. Candidate-value probes can also outperform probability- and embedding-based baselines. Furthermore, a controlled online intervention experiment shows that using the decoded signals to guide token selection during generation can improve the chances of eventual success. Our findings provide evidence that internal representations encode information about relative long-horizon candidate value, offering insights into how local candidate-level representations relate to long-horizon reasoning outcomes.

cs.CL↗

MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement

Reinforcement learning (RL) is the central training paradigm for advancing large foundation models towards self-improvement. This report introduces the MiMo-V2.6 series, an omni-modal family that pushes the frontier of model intelligence by scaling RL compute. Prior to RL, we conduct mid-training on a broad multimodal corpus to provide ample exploration space, and build a solid infrastructure on the pretrained hybrid-SWA architecture to support subsequent scale-up. We scale RL compute along three dimensions: (1) larger batches and higher throughput, with an asynchronous training that consumes 1,568 samples and 2.7-3.7B tokens per step at context lengths of up to 1M; (2) more diverse and complex environments, spanning code, general, visual, and cyber domains under a mixture of agent harnesses; and (3) more grader compute, via groupwise agentic grading that yields more accurate reward signals for long-horizon tasks and steers the model towards shorter, more token-efficient solutions. To keep training stable at scale, we freeze the MoE router and establish a multi-layer defense against reward hacking. We further build infrastructure for mixed-task agentic RL, including a unified trajectory representation, high-concurrency multi-framework rollout, decoupled control and data planes, and training-inference consistency. We open-source the training dynamics, RL environments, and RL framework to facilitate reproduction and further research on scaled RL and model self-improvement.

cs.CL↗

Retrieval-Augmented Interpretable Learning: Towards Task-Specific Zero-Shot Models in Healthcare

We introduce Retrieval-Augmented Interpretable Learning (RAIL), a probabilistic meta-learning framework for zero-shot generation of task-specific interpretable models that synthesizes coefficient-space structure from natural-language task descriptions and a memory of previously learned task-specific predictors. RAIL retrieves related source tasks, transfers structure through coefficient space, and generates a new predictor in the original diagnostic-feature space, enabling zero-shot and few-shot clinical procedure prediction with feature-level explanations. Its probabilistic formulation provides uncertainty over retrieval, model coefficients, and predictions, supporting uncertainty-aware deployment: uncertain predictions or unstable explanations can be flagged for additional clinical review rather than treated as automatic decisions. This makes RAIL particularly suited for healthcare settings, where prediction tasks are highly long-tailed, new clinical targets arise frequently, and models must remain inspectable, uncertainty-aware, and compatible with human oversight. Across long-tailed clinical procedure prediction tasks, RAIL improves low-data model generation, benefits from clinically informed task representations, and yields retrieval, uncertainty, and coefficient-level diagnostics that make model behavior more transparent. These results suggest a path toward scalable clinical prediction systems that can adapt to new tasks while preserving interpretability and reliability.

cs.LG↗

Asymptotic Rigidity and Boundary Structure of Yang's Numerical Invariants for the Bidisk Submodules $[z^k-w^\ell]$

Let \[ M_{k,\ell}=[z^k-w^\ell]\subset H^2(\mathbb D^2), \qquad k,\ell\in\mathbb N,\quad k\ne\ell . \] We study the large-index asymptotics of Yang's numerical invariants and the boundary structure of their generating function. Starting from the exact staircase formula obtained in our preceding work, we prove that \[ Σ_j(M_{k,\ell}) = \frac{C_{k,\ell}}{j} + O(j^{-2}), \] where \[ C_{k,\ell} = \int_0^\infty \frac{x^2} {(x+1/k)^2(x+1/\ell)^2}\,dx . \] The first strict descent together with $C_{k,\ell}$ recovers the unordered pair $\{k,\ell\}$, yielding an asymptotic rigidity principle. At the next order we obtain a periodic correction \[ Σ_j(M_{k,\ell}) = \frac{C_{k,\ell}}{j} + \frac{Ψ_{k,\ell}(j)}{j^2} + O(j^{-3}), \] whose least period is \(\operatorname{lcm}(k,\ell)\), and we determine the leading amplitudes of the strict drops. For Yang's generating function \[ \mathcal P_{k,\ell}(t) = \sum_{j=0}^{\infty}Σ_j(M_{k,\ell})t^j, \] we prove that it has radius of convergence one and a logarithmic singularity at $t=1$; hence it is never a polynomial for $k\ne\ell$, giving a negative answer to Yang's polynomiality question within this quasi-homogeneous family. The periodic higher-order corrections generate root-of-unity polylogarithmic boundary modes. The leading logarithmic coefficient together with the second-order boundary support determines the unordered pair $\{k,\ell\}$, while successive renormalized boundary limits recover the Fourier coefficients of every finite-order periodic asymptotic term.

math.FA↗

DiCoR: Decoupled Referent Disambiguation and Contour Recalibration for Efficient Referring Remote Sensing Image Segmentation

Referring remote sensing image segmentation (RRSIS) aims to delineate targets specified by natural language expressions in remote sensing imagery. Existing methods mainly follow joint fusion segmentation (JFS) or decoupled prompt segmentation (DPS). JFS is efficient but often suffers from limited accuracy because referent localization and mask delineation are optimized under a unified objective, whereas DPS separates localization from mask generation using spatial prompts and foundation segmenters at the cost of higher memory consumption and inference latency. To bridge this gap, we propose DiCoR, a decoupled referent disambiguation and contour recalibration framework built on an efficient JFS pipeline. DiCoR addresses two key challenges: distinguishing the correct referent from ambiguous candidates and refining coarse masks after localization. A disambiguation-aware localization guidance strategy ranks salient candidate regions with adaptive linguistic cues and injects the resulting localization prior into fused features. A lightweight contour recalibration module further predicts residual corrections to coarse logits under localized contour supervision, improving mask quality with limited computational overhead. Experiments on RefSegRS, RRSIS-D, and RISBench show that DiCoR achieves the best segmentation accuracy across all three benchmarks. On RefSegRS, it improves mIoU and gIoU by 4.90% and 2.16% over a competitive JFS method while running 2.4-6.5 times faster than the evaluated DPS method, demonstrating a favorable accuracy-efficiency trade-off. Code is available at https://github.com/zyGao1126/DiCoR.

cs.CV↗

How to Recover Oscillation-Free Pressure in Real Fluids: The RFQC Method and Its Liquid-Upwind Anomaly

From the perspective of continuum thermodynamics, we revisit the pressure oscillation problem in finite-volume methods for multiphase real fluids and clarify the physical counterpart of the Real Fluid Quasi-Conservative (RFQC) method. The pressure oscillation in conservative finite-volume methods originates from their implicit thermodynamic equilibrium assumption, whereas recovering an oscillation-free pressure requires additional physical information. The RFQC method achieves this by evolving the affine parameters xi and E0 of the isentropic internal-energy-pressure relation along pathlines, while the thermodynamic re-projection converts the deviation from the isentropic trajectory into an internal-energy error, thereby ensuring the thermodynamic consistency and numerical stability of the method. We then investigate the applicability limit of the RFQC method and identify a Liquid-upwind Anomaly (LUA) in extreme phase-change cases. For a Riemann problem involving liquid-vapor phase change, a numerical anomaly may occur if a liquid-upwind translational velocity is initially superimposed. Theoretical analysis reveals that this anomaly is initiated by the jump in the affine slope xi during phase change, which delays pressure rise in the downstream vapor cell. Concurrently, the re-projection removes the positive pressure increment, repeatedly generating large internal-energy errors and trapping the vapor cell in a cycle of delayed pressure recovery. The analysis indicates that the LUA is a start-up anomaly, which can be resolved by introducing a regularization strategy at the initial discontinuity. With the proposed regularization strategy, the RFQC method is equipped with enhanced accuracy and robustness for extreme thermodynamic flows, such as sonic phase-change jets.

math-ph↗

Numerical Invariants and Parameter Recovery for Quasi-Homogeneous Submodules $[z^k-w^\ell]$

We study Yang's higher numerical invariants for the quasi-homogeneous submodules \[ M_{k,\ell}=[z^k-w^\ell]\subset H^2(\mathbb D^2), \qquad k,\ell\in\mathbb N,\quad k\neq\ell. \] The Hilbert--Schmidt property and the low-order invariants $Σ_0$ and $Σ_1$ for this family follow from earlier results on $M_{θ,φ}$-type submodules. Our aim is to determine the complete higher-order sequence. Using the weighted homogeneous decomposition associated with $z^k-w^\ell$, we construct explicit orthonormal bases for the two defect spaces and determine the exact support of the shifted defect inner products. For $j\geq1$, set \[ A_j=\left\lceil\frac{j}{\ell}\right\rceil, \qquad B_j=\left\lceil\frac{j}{k}\right\rceil. \] We obtain \[ Σ_0(M_{k,\ell})=\frac{π^2}{6}, \qquad Σ_j(M_{k,\ell})=F(A_j,B_j), \] where $F$ is an explicitly evaluated symmetric function on $\mathbb N^2$. The sequence is nonincreasing, tends to zero, and satisfies $Σ_j>Σ_{j+1} \Longleftrightarrow k\mid j\ \text{or}\ \ell\mid j$. Hence its strict descent set is $k\mathbb N\cup\ell\mathbb N$. This descent set determines the additive semigroup $\langle k,\ell\rangle$, while the complete numerical invariant sequence determines the unordered pair $\{k,\ell\}$. The order of the two exponents is not recoverable because of coordinate-swap symmetry.

math.FA↗

LeaseGuard: Incumbent-Preserving Admission Control for Privileged LLM Agents

Privileged language-model agents can satisfy a new system task by displacing a healthy incumbent that depends on the same file, process, socket, lock, or capacity allocation. This failure arises because execution privilege determines whether an operation can run, not whether the requester may preempt the current resource owner. We present LeaseGuard, a deterministic admission layer that represents preemption authority through canonical resource leases, incumbent-health checks, effect-aware admission, coexistence limits, safe alternatives, and resource-scoped overrides before adapter execution. We evaluate it on a frozen benchmark of 60 newly authored conflict scenarios with matched controls across two local model families. Relative to a preservation prompt, LeaseGuard reduces unauthorized preemption from 73.3% to 0.0% and increases safe completion by 70.0 percentage points (scenario-clustered 95% CI [60.8, 79.2]). Requested-task success changes by -3.3 points (95% CI [-9.2, 2.5]). The fully evaluated v0.2 broker also rejects a forged incumbent task identity in a hash-linked stress audit. Expiry-only reclamation can still expose a healthy incumbent after missed renewal. The evidence supports incumbent-preserving admission when effects are completely mediated, task ownership is authenticated, and lease expiry reflects incumbent liveness.

cs.CR↗

Residual Community Prototypes Under-Reject Held-Out Malware Families in FCG-MFD

Open-set malware-family recognition must classify known families while rejecting families absent from training. We test whether Louvain-community summaries add rejection information beyond a graph neural network embedding and dimension-matched generic topology. The study uses a deduplicated, conflict-audited FCG-MFD corpus, five held-out families, and three optimization seeds. Community features are residualized against generic topology using known-family training data before nearest-prototype scoring. Residual community does not produce stable held-out-family rejection. Ranking effects reverse across families, the false-positive rate at 95 percent unknown recall worsens for every held-out family, and a validation-fitted threshold rejects only 4.48 percent of unknown samples. Accepted-known macro F1 improves in every family, but with five independent family units the exact two-sided sign-flip p-value is 0.0625, the smallest attainable value. The score remains associated with graph scale, while simple classifier uncertainty performs better on ranking, high-recall rejection, and OSCR. In this GIN/FCG-MFD setting, community-enriched prototypes change known-class geometry without creating a stable unknown margin. Graph open-set evaluations should pair structural features with matched topology controls, operational thresholds, and held-out-family analysis.

cs.CR↗

Partition-Matched Evaluation of Community Features under Distribution Shift in Android Malware Function-Call Graphs

Graph-based Android malware classifiers can lose accuracy under malware-type or family shifts. We test whether mesoscopic organization in function-call graphs provides shift-stable information beyond local degree profiles (LDP), global statistics, lightweight metadata, and size-matched random partitions. Using 15,000 MalNet-Tiny, Common, and Distinct graphs, six Leiden descriptors specified before evaluation, and five optimizer seeds, communities raise Tiny macro F1 from 78.7% to 81.3% but yield 29.4-point source-only Common degradation. One size-matched random partition yields 27.8-point degradation. Across five random partitions, mean degradation is 27.9 points; the 95% two-level bootstrap interval for random minus community degradation is [-4.2, 1.2] points. With metadata, the corresponding difference is -0.2 points with interval [-1.1, 0.6]. Removing modularity raises structure-only Common macro F1 from 51.9% to 53.6%. The tested signature adds IID signal but shows no repeatable shift-stability advantage, demonstrating why mesoscopic graph claims need partition-matched controls and repeated null draws.

cs.CR↗

Multimodal Prompt Learning with Irregular EHRs for Robust Monitoring of Critical Care Patients

Accurate assessment of patients in intensive care units (ICUs) is essential for timely clinical intervention and improved patient outcomes. Multimodal electronic health records (EHRs), including structured physiological time series and longitudinal clinical notes, provide complementary information for critical care prediction. However, in real-world clinical settings, individual modalities may be partially observed or entirely unavailable, resulting in substantial performance degradation for existing multimodal models. To address this challenge, we propose a multimodal prompt-learning framework for robust clinical prediction under diverse missing-modality scenarios. The proposed framework introduces four complementary types of prompts: generative prompts, missing-signal prompts, missing-type prompts, and temporal prompts. Generative prompts construct surrogate latent representations for unavailable modalities, while missing-signal prompts distinguish observed representations from generated ones. Missing-type prompts condition the model on different modality-availability configurations, whereas temporal prompts perform condition-specific aggregation over temporally encoded clinical sequences. Together, these prompts enable the model to capture missingness-aware intramodal dependencies and cross-modal interactions within a unified architecture. Extensive experiments demonstrate that our method outperforms existing approaches across evaluation metrics on two missingness settings. Ablation and robustness analyses further verify the complementary contributions of the four prompt types and the effectiveness of the proposed framework for clinical prediction from incomplete multimodal EHR data.

cs.AI↗

Graph Is the Verifier: Agentic Reinforcement Learning for Interprocedural Vulnerability Detection

Real-world vulnerabilities often span multiple functions, yet most learning-based detectors classify each function in isolation: on a sample of real CVEs, we find that 71.7% of vulnerable functions require evidence from outside the function to be classified correctly. Agentic reinforcement learning (RL) could close this gap by enabling a model to gather that evidence itself, but it lacks a reliable reward, since a reward defined on the final verdict alone can be obtained without performing any investigation. We propose VulAgentRL, an agentic RL framework for interprocedural vulnerability detection built on a Code Property Graph (CPG). The CPG serves two roles: at inference time the policy queries it for callers, callees, dataflow, and other queries, and at training time the same graph verifies the evidence the policy cites. Because every CPG node carries a persistent integer identifier, this verification is an exact comparison rather than a textual match, so the reward credits verdicts that are supported by evidence. We further initialize the policy by distilling teacher investigations, and show that this warm start is necessary, since RL cannot acquire tool-use behavior it never samples. Under a repository-level split that prevents leakage, VulAgentRL outperforms state-of-the-art baselines, including frontier models, on the strict pair-wise-correct metric while issuing fewer tool calls, and its advantage persists on an out-of-distribution corpus and under class imbalance.

cs.CR↗

Context-Adaptive Inference: A Unified Statistical and Foundation-Model View

Modern predictive systems are expected to adapt their behavior to the specific situation they are facing. A clinical model should not treat every patient the same; a retrieval-augmented model should change its answer when given different evidence; a mixture-of-experts model should route different inputs to different experts. We call this capability context-adaptive inference: before predicting, the system uses information about the current context to specialize its parameters or computation for that instance. This article provides a unified view of context-adaptive inference across three traditions that are usually treated separately: (i) explicit adaptation in statistics (e.g. varying-coefficient models, local regression, hierarchical sharing), (ii) rapid task-specific adaptation in meta-learning and transfer, and (iii) implicit adaptation in large foundation models via prompting, retrieval, and expert routing. We formalize these approaches under a common objective: to map context $c$ to adapted parameters $θ(c)$, then to predict via $f(x; θ(c))$. Under squared loss, linear prediction heads, and fixed features, we prove that explicit parameter adaptation and implicit routing are mathematically equivalent to kernel ridge regression on joint features of inputs and context. Building on this bridge, we propose practical design principles and evaluation metrics including adaptation-efficiency, routing stability, and context-specific robustness to guide when to specialize, how to constrain that specialization, and how to audit context-adaptive models in deployment. Finally, we identify open problems in identifiability, robustness under distribution shift, and efficient large-scale adaptation, outlining design principles for methods that are scalable, reliable, and transparent in real-world settings.

stat.ML↗