arXiv subjects

Wei Jin

Publications and source records attributed to Wei Jin.

At least 19 recordsLinked to original sources

arXiv2026

GC4NC: A Benchmark Framework for Graph Condensation on Node Classification with New Insights

Graph condensation (GC) is an emerging technique designed to learn a significantly smaller graph that retains the essential information of the original graph. This condensed graph has shown promise in accelerating graph neural networks while preserving performance comparable to those achieved with the original, larger graphs. Additionally, this technique facilitates downstream applications like neural architecture search and deepens our understanding of redundancies in large graphs. Despite the rapid development of GC methods, particularly for node classification, a unified evaluation framework is still lacking to systematically compare different GC methods or clarify key design choices for improving their effectiveness. To bridge these gaps, we introduce \textbf{GC4NC}, a comprehensive framework for evaluating diverse GC methods on node classification across multiple dimensions including performance, efficiency, privacy preservation, denoising ability, NAS effectiveness, and transferability. Our systematic evaluation offers novel insights into how condensed graphs behave and the critical design choices that drive their success. These findings pave the way for future advancements in GC methods, enhancing both performance and expanding their real-world applications. Our code is available at https://github.com/Emory-Melody/GraphSlim/tree/main/benchmark.

cs.LG↗

arXiv2026

Benchmarking Generalist and Specialist Time-Series Foundation Models for PPG-Based Health Assessment

Time-series Foundation Models (FMs) are pretrained on either heterogeneous corpora or specific signal domains. Using Photoplethysmography (PPG) as a representative modality, we compare generalist time-series FMs, PPG-specialist FMs, and traditional machine learning (ML) baselines for PPG-based health assessment. We conduct a two-stage benchmark. First, we evaluate multiple FM families across 23 downstream evaluations (e.g., atrial fibrillation detection, vital-sign estimation). Second, we conduct a focused four-FM comparison (MOMENT, Chronos-2, GPT-PPG, MOMENT-PPG) at multiple model sizes on an expanded 51-task suite, under both frozen-backbone and full-model tuning, to study tuning strategy, model scaling, and cross-dataset transfer. The multi-model benchmark shows that PPG-specialist FMs achieve the strongest aggregate performance, but their advantage is not uniform: generalist FMs remain competitive on several tasks, and traditional ML baselines outperform FMs on selected clinical regression tasks. The focused analysis shows that the advantage of PPG-specific pretraining appears mainly with frozen backbones and on targets that the waveform directly reflects, such as atrial fibrillation and vital signs. Full-model tuning narrows this advantage, model rankings change with the tuning strategy, and the largest model variant is the best choice in only about one-third of cases. Additionally, controlled experiments with two generalist models show that initializing PPG pretraining from generalist checkpoints provides an early advantage over training from scratch, even when the pretrained features offer little downstream benefit. Together, these findings suggest that PPG-specific pretraining is useful but task-dependent, and that FMs should be evaluated against simpler statistical, traditional ML, and clinical history-based baselines.

cs.LG↗

arXiv2026

Adaptive Auto-Harness: Sustained Self-Improvement for Agentic System Deployment on Open-Ended Task Streams

Auto-harness systems improve LLM agents by adapting prompts, skills, tools, and supporting infrastructure from execution feedback. Sustaining these gains on open-ended task streams requires handling growing histories, heterogeneous tasks, and shifting distributions. We introduce Adaptive Auto-Harness, a unified framework that separates persistent evolution workspaces from deployable harnesses and supports interchangeable evolution procedures. We compare configurations adapted from five prior approaches on recorded streams of prediction markets, security challenges, and event forecasts. With the same Sonnet solver, our full system reaches 80.9\% market all-task accuracy and 50.2\% security Pass@1, versus 22.2\% and 37.2\% without evolution; market accuracy also exceeds a market-following control's 72.0\%. Our analyses show that artifact accumulation does not guarantee sustained gains, and knowledge that transfers across tasks can still lose value as the task distribution changes. Continued evolution can therefore help preserve earlier gains through revision of accumulated knowledge. Code: https://github.com/A-EVO-Lab/a-evolve.

cs.LG↗

arXiv2026

Comparing classes of highly symmetric graphs: From $2$-arc-transitive to $2$-distance-transitive

A $2$-distance-transitive graph is a vertex-transitive graph whose vertex stabilizer is transitive on both the first- and second-step neighborhoods. This concept simultaneously generalizes both distance-transitive graphs and $2$-arc-transitive graphs. In this paper, we first determine the vertex-quasiprimitive types of $2$-distance-transitive graphs of odd order, partially answering a question posed by A. Devillers, M. Giudici, C. H. Li and C. E. Praeger in 2012. We then prove that a $2$-distance-transitive graph of valency $p+1$, where $p$ is a prime, is $2$-arc-transitive if and only if it has girth at least $4$. We also show that every locally-primitive $2$-distance-transitive graph of valency at most $8$ is $2$-arc-transitive, with the icosahedron as the unique exception. Finally, we prove that if $Γ$ is a $G$-locally-primitive, $(G,2)$-distance-transitive graph of valency at least $3$ and $G$ is soluble, then either $Γ\cong \K_{p,p}$ for some prime $p$, or the order of $Γ$ is not square-free.

math.CO↗

arXiv2026

Amply regular graphs with $μ$ close to half the valency and group divisible designs

In this paper, we classify connected amply regular graphs with diameter $d \geq 4$ and parameters $(v, k, λ, μ)$ satisfying $μ= \frac{k-1}{2}$, where $k\geq 5$ is odd. We prove that such a graph must be exactly one of the following: the $5$-cube, the graph $\K_2 \square Λ$, where $Λ$ is the unique bipartite $(0,2)$-graph on $14$ vertices, or the point--block incidence graph of a group divisible design with the dual property, namely a $GDDDP\left(2, k+1;\, k;\, 0, \frac{k-1}{2}\right)$. For the last family, we give equivalent characterizations in terms of bipartite $Q$-regular graphs and relation graphs of symmetric association schemes with five classes. Furthermore, we present constructions of such amply regular graphs, yielding infinite families of examples derived from Paley graphs, Peisert graphs, and Paley digraphs.

math.CO↗

arXiv2026

From Single-Actor Environments to Multi-Actor Event Worlds: Can LLMs Be Event World Models?

World models help agents assess the consequences of their actions before acting. Many real systems, however, involve multiple actors who make their own decisions, can evolve even when the agent takes no action, and are only partially observed through event logs. We formalize these systems as event worlds and ask whether large language models can predict future observations from the log and the agent's proposed actions. We build complementary benchmarks from human Diplomacy games and real Polymarket trades. Our hierarchical framework compares prediction with and without other actors' actions, then tests multi-step rollouts, predictions for actions never taken, and usefulness to agents. Fine-tuning and more data improve prediction when all actors' recorded actions are given. With only the agent's action, models struggle when other actors strongly affect the outcome, and larger models and longer histories help little. This gap also limits multi-step prediction and agent decisions, with models overestimating plan success. Yet imagined experience and initialization from world models can still improve policy learning. Anticipating other actors' actions remains a central challenge for event world modeling.

cs.AI↗

arXiv2026

Language Models as AI Research World Models

AI research agents automate the cycle of proposing, implementing, and evaluating experiments, opening a path toward recursive self-improvement. Yet their ability to propose experiments outpaces their capacity to execute them in real environments, making outcome prediction a key capability for sustained self-improvement under limited experimental budgets. We investigate language models as Research World Models (RWMs), which predict the outcomes of candidate interventions across research environments. Our evaluation draws on over 2,600 experimental records from nine research environments spanning pretraining, post-training, and inference, representing more than 171,000 H100 GPU-hours of experimentation. Research knowledge acquired from real experimental experience improves RWM predictions of unseen interventions within the same environment (Spearman +0.27), and can be reused across environments. For example, using only pretraining experience from OLMo3, Marin, and Nanochat, an RWM reduces selection regret in the Qwen3 environment by 78% compared with zero-experience setting. These benefits extend to multi-round Autoresearch under a fixed selection budget: RWMs with in-env and cross-env research knowledge increase the best gain achieved by 15.8% and 11.6%, respectively. Ablations across 13 language models used as RWMs show that adding research knowledge can improve intervention ranking more than changing models or increasing reasoning effort alone. These findings support language models as RWMs and motivate accumulating experimental data for future RWM training.

cs.CL↗

arXiv2026

Agent Policy-Value Audit: Separating Transition Composition from Event Selection in Financial LLM Agents

Financial LLM agents are often evaluated by comparing their end-to-end returns with those of a baseline and testing the paired difference against zero. This measures whether deploying the agent changes realized performance, but it does not isolate event-selection skill. An agent that frequently changes positions from flat to long can earn a positive paired return from an upward-drifting event pool even when it selects events at random. We propose the Agent Policy-Value Audit, which holds fixed the observed count of each ordered action-change type and randomly reassigns them across eligible events. The average payoff from these reassignments is the composition benchmark; the difference between observed deployment value and this benchmark is selection value. In semi-synthetic benchmarks based on real earnings-event returns, a zero-centered paired test falsely attributes passive exposure to selection skill in $11.6\%$ of no-skill replications, while the transition-matched audit reduces this rate to $5.3\%$. Applied retrospectively to 723 earnings events at 44 U.S. consumer-facing firms, the audit decomposes the agent's gross deployment value of $+15.2$ bps/event into a $+25.8$ composition benchmark and a $-10.6$ selection value. The agent does not detectably outperform matched random assignments. Financial-agent evaluations should report deployment value separately from event-selection value.

cs.AI↗

arXiv2026

On 2-distance-transitive circulant digraphs

Circulant digraphs form a prominent class of Cayley digraphs defined on finite cyclic groups. Building on the existing classification of $2$-arc-transitive circulant graphs, this paper presents a complete classification of $2$-distance-transitive circulant digraphs. Our main theorem establishes that every connected $2$-distance-transitive circulant digraph is isomorphic to one of the following: the undirected cycle $C_n$, the complete bipartite graph $\K_{\frac{n}{2},\frac{n}{2}}$, the complete multipartite graph $\K_{m[b]}$ with $m\geq 3,b\geq 2$, the graph $\K_{\frac{n}{2},\frac{n}{2}}-\frac{n}{2}\K_2$ for odd $\frac{n}{2}$, prime-order Paley graphs, the directed cycle $\overrightarrow{C}_n$, the oriented graph $G(p^m,r)$ satisfying Condition~\ref{p-power-normal-2dt-cond}, the oriented graph $ C_r(b,1)$ with $r\geq 3,b\geq 2$ and $rb=n$, the lexicographic product oriented graph $ G(p^m,r)[\overline{\K}_d]$ where $G(p^m,r)$ obeys Condition~\ref{p-power-normal-2dt-cond}.

math.CO↗

arXiv2026

EpiQAL: Benchmarking Large Language Models in Epidemiological Question Answering and Reasoning

Reliable epidemiological reasoning requires synthesizing study evidence to infer disease burden, transmission dynamics, and intervention effects at the population level. Existing medical question answering benchmarks primarily emphasize clinical knowledge or patient-level reasoning, yet few systematically evaluate evidence-grounded epidemiological inference. We present EpiQAL, to our knowledge the first diagnostic benchmark for epidemiological question answering over research literature, comprising three subsets built from open-access articles across diverse diseases. The three subsets progressively test factual recall, multi-step inference, and conclusion reconstruction under incomplete information, and are constructed through a quality-controlled pipeline combining taxonomy guidance, multi-model verification, and difficulty screening. Experiments on fifteen models spanning open-source and proprietary systems reveal that current LLMs show limited performance on epidemiological reasoning, with multi-step inference posing the greatest challenge. Model rankings shift across subsets, and scale alone does not predict success. Chain-of-Thought prompting benefits multi-step inference but yields mixed results elsewhere. EpiQAL provides fine-grained diagnostic signals for evidence-grounding, inferential reasoning, and conclusion reconstruction.

cs.CL↗

arXiv2026

Temporal Graph Prototype-conditioned Conformal Prediction for Fraud Detection

Conformal prediction (CP) provides distribution-free coverage guarantees and has emerged as a principled tool for uncertainty quantification. In edge-level fraud detection on temporal interaction graphs, where false positives and false negatives both carry substantial cost, such coverage guarantees are particularly appealing for risk-aware decision making. However, directly applying existing graph conformal predictors yields inefficient prediction sets due to two recurring properties of fraud data. Fraudulent interactions are often embedded in benign-dominated neighborhoods that dilute calibration signals, while extreme class imbalance leaves scarce labeled-fraud support in the calibration split and leads to overly conservative class-conditional thresholds. To address these issues, we propose ProtoCP, a conformal prediction framework for edge-level fraud detection on temporal graphs. ProtoCP improves calibration efficiency by focusing calibration on fraud-relevant subgraph context and producing more stable nonconformity scores under class imbalance and temporal drift. Specifically, it leverages learned prototypes to suppress benign-dominated noise in the calibration context and introduces a neighborhood-relative scoring mechanism with temporal score diffusion for stable class-conditional calibration. Experiments on four fraud benchmarks (YelpChi, S-FFSD, FTFD, and BankSim) show that ProtoCP achieves the target coverage with consistently smaller prediction sets than state-of-the-art baselines. Our codes are available at https://github.com/Picard1701ent/ProtoCP.git

cs.LG↗

arXiv2026

LVCG: Learning ECG Representations in the Latent Vectorcardiogram Space

Electrocardiography (ECG) is a cornerstone of cardiac assessment, making the learning of informative ECG representations fundamental to tasks ranging from disease diagnosis to clinical report generation. However, existing methods operate almost exclusively in the observable ECG signal space. In practice, the standard twelve-lead ECG represents multiple projections of the same underlying cardiac electrical activity from different spatial orientations. Therefore, representation learning in the ECG space inevitably introduces substantial redundancy, which may lead to spurious correlations and increased risk of overfitting. To address this and motivated by the Frank vectorcardiogram (VCG) model, we propose learning a unified latent representation of cardiac electrical activity directly in the VCG space. We introduce LVCG, the first general self-supervised representation learning framework designed to operate in this physically grounded latent space. By learning view-invariant latent VCG representations rather than lead-specific artifacts, VCG minimizes redundancy and improves generalization. LVCG generally outperforms ECG-space baselines across tasks, demonstrating enhanced robustness and generalization, especially in domain shift settings.

cs.LG↗

arXiv2026

The cyclic diagnosability of Cayley graphs generated by transposition triangle-free unicyclic graphs under the PMC and MM* models

Diagnosability is a critical parameter for evaluating the reliability and self-diagnostic capacity of multiprocessor systems. As an advanced extension of traditional diagnosability, cyclic diagnosability has been proposed to enable more comprehensive assessment of the diagnostic capabilities of interconnection networks. In this paper, we prove that under both the PMC and MM* models, the cyclic diagnosability of Cayley graphs generated by triangle-free unicyclic graphs is 5n-10 for n\ge11.

math.CO↗

arXiv2026

Classification of two-distance-transitive Cayley graphs of the semi-dihedral groups

The class of 2-distance-transitive graphs naturally generalizes distance-transitive graphs and plays a central role in algebraic graph theory. Classifying such graphs for a prescribed underlying group is a key open problem. A vertex-transitive graph $Γ$ is said to be $2$-distance-transitive if, for each $i\in \{1,2\}$, any two pairs of vertices with identical distance $i$ in $Γ$ can be mapped to each other via some automorphism of the graph. In this paper, we present a complete classification of all $2$-distance-transitive Cayley graphs of the semi-dihedral groups.

math.CO↗

arXiv2026

A classification of locally-quasiprimitive circulant digraphs

Circulant digraphs are Cayley digraphs over finite cyclic groups and constitute a fundamental class of objects in algebraic graph theory. Extending the classification of locally-primitive circulant graphs \cite{JZ-2026}, we completely determine all locally-quasiprimitive circulant digraphs. Our main theorem shows that a connected locally-quasiprimitive circulant digraph is isomorphic to one of the following: the complete graph $\K_n$, the complete bipartite graph $\K_{n/2,n/2}$, the graph $\K_{n/2,n/2}-\frac{n}{2}\K_2$ (with $n/2$ odd), the cycle $\C_n$, the directed cycle $\vec \C_n$, a normal circulant digraph of prime valency, the lexicographic product $\vec \C_m[\overline{\K_b}]$, or the tensor product $\vec \C_m\times \K_b$ with $\gcd(m,b)=1$.

math.CO↗

arXiv2026

Scalable Graph Condensation with Evolving Capabilities

The rapid growth of graph data creates significant scalability challenges as most graph algorithms scale quadratically with size. To mitigate these issues, Graph Condensation (GC) methods have been proposed to learn a small graph from a larger one, accelerating downstream tasks. However, existing approaches critically assume a static training set, which conflicts with the inherently dynamic and evolving nature of real-world graph data. This work introduces a novel framework for continual graph condensation, enabling efficient updates to the distilled graph that handle data streams without requiring costly retraining. This limitation leads to inefficiencies when condensing growing training sets. In this paper, we introduce GECC (\underline{G}raph \underline{E}volving \underline{C}lustering \underline{C}ondensation), a scalable graph condensation method designed to handle large-scale and evolving graph data. GECC employs a traceable and efficient approach by performing class-wise clustering on aggregated features. Furthermore, it can inherit previous condensation results as clustering centroids when the condensed graph expands, thereby attaining an evolving capability. This methodology is supported by robust theoretical foundations and demonstrates superior empirical performance. Comprehensive experiments including real world scenario show that GECC achieves better performance than most state-of-the-art graph condensation methods while delivering an around 1000$\times$ speedup on large datasets.

cs.LG↗

arXiv2026

EpiEvolve: Self-Evolving Agents for Streaming Pandemic Forecasting under Regime Shifts

Epidemic LLM forecasters are usually trained and evaluated as static supervised models, whereas operational pandemic forecasting is a streaming process in which labels arrive after predictions and disease regimes shift over time. We study this mismatch in weekly COVID-19 hospitalization trend forecasting across five variant regimes. We introduce EpiEvolve, a self-evolving agent that wraps an LLM forecaster trained on the warm-start period and keeps its weights fixed during streaming. EpiEvolve adapts by storing forecast outcomes in a hierarchical episodic memory, reflecting on delayed labels, retrieving cases relevant to the current regime, and distilling recurring errors into strategic rules. The resulting context lets the forecaster reuse its own past predictions and outcomes in later weeks while following a chronological protocol that prevents future leakage. On the streaming dataset, EpiEvolve reaches $0.629$ average accuracy, compared with $0.561$ for the static backbone and $0.325$ for the external CDC ensemble, and reduces recovery lag after regime shifts from $5$ to $2$ weeks. Ablations show that reflection, strategic memory, and regime-aware retrieval each contribute to the gains.

cs.AI↗

arXiv2026

Exposing Vulnerabilities in Explanation for Time Series Classifiers via Dual-Target Attacks

Interpretable time series deep learning systems are often assessed by checking temporal consistency on explanations, implicitly treating this as evidence of robustness. We show that this assumption can fail: Predictions and explanations can be adversarially decoupled, enabling targeted misclassification while the explanation remains plausible and consistent with a chosen reference rationale. We propose TSEF (Time Series Explanation Fooler), a dual-target attack that jointly manipulates the classifier and explainer outputs. In contrast to single-objective misclassification attacks that disrupt explanation and spread attribution mass broadly, TSEF achieves targeted prediction changes while keeping explanations consistent with the reference. Across multiple datasets and explainer backbones, our results consistently reveal that explanation stability is a misleading proxy for decision robustness and motivate coupling-aware robustness evaluations for trustworthy time series tasks.

cs.LG↗