Searcharxiv⌕ Search

arXiv subjects

Bolun Liu

Publications and source records attributed to Bolun Liu.

13 recordsLinked to original sources

Verbal tics in frontier language models: A critical review of current releases, research evidence, and public discussion

Repeated praise, canned reassurance, formulaic contrasts, and conspicuous vocabulary figure prominently in discussions of large language models. Context governs how these habits should be read. A conventional phrase can orient a reader, whereas a fluent answer can reinforce a false belief. This critical review examines linguistic habits and sycophancy across eight developer families. We verify public offerings from OpenAI, Anthropic, Google DeepMind, xAI, ByteDance, Moonshot AI, DeepSeek, and Xiaomi against official release and API documentation through 1 October 2026. We then synthesize empirical research on lexical overrepresentation, stylistic variation, social warmth, and agreement alongside benchmark methods and dated public discussions in English and Chinese. Published studies document recurring linguistic patterns and agreement that distorts human judgment, but comparable measurements for the newest releases remain sparse in the retrieved material. Current user accounts report both persistent complaints and clearer prose, and experiences vary by task and prompt design. We propose separate measures for recurrence, contextual appropriateness, and belief distortion paired with exact service records and language-specific annotation. This framework keeps claims about writing quality and conversational reliability testable as model services change.

cs.CL↗

Seeing Is No Longer Believing: Frontier Image Generation Models, Synthetic Visual Evidence, and Real-World Risk

Image generation systems can produce plausible photographs, readable documents, and consistent depictions of people and places. Presented as records of real events, these outputs can influence decisions in news, finance, identity verification, medicine, and law. This narrative review examines public model documentation, incident reports, empirical studies, governance materials, and illustrative English and Chinese practitioner commentary available through 1 October 2026. We separate vendor capability claims, documented incidents, experimental findings, and prospective harm pathways. Realism, text rendering, reference consistency, conversational editing, search grounding, and low generation cost shape the conditions under which synthetic images acquire evidentiary weight. Documented episodes show how fabricated visuals enter public and institutional workflows, although public reporting cannot measure misuse prevalence for current releases. Comparing provider safeguards, provenance credentials, watermarks, platform labels, and statutory duties shows that origin signals do not verify a depicted event or transaction. Our framework matches artifact types and decision stakes to specific controls. Consequential decisions require authenticated source records, independent channel corroboration, and proportionate review before action.

cs.CL↗

From Viral Exposure to Engagement: A Causal Mediation Framework for Digital Platforms

At Roblox, viral hits can change how exposed players engage with the viral game and with the rest of the platform. We develop a causal mediation framework to decompose the effect of viral-game playtime on subsequent player engagement into a direct effect and indirect effects through measured player interaction and discovery behaviors. The framework uses a causal DAG to state the assumed causal structure and a time-ordered design that measures covariates, exposure, mediators, and outcomes in non-overlapping time windows. Under stated identification assumptions and a model linear in the exposure and each mediator at fixed covariates, we estimate every path as a function of player covariates with cross-fitted R-learners, which use only prediction models and therefore scale to platform-size data. Averaging the fitted surfaces inherits their regularization bias, so we add a per-player bias correction before averaging (Chernozhukov et al., 2018). In simulation, the correction restored nominal coverage of 90% to 95%, against at most 39% for intervals built by averaging the surfaces. Counterintuitively, interval coverage worsened with more data on a difficult mediator-outcome path, bias shrank more slowly than the standard error, reducing corrected coverage from 93.6% to 57.3% as n increased from 2k to 200k. We applied the same design to two de-identified viral games with different lifecycle histories, comparing effects across outcome scopes, player subgroups, and a placebo window without a viral event. The results indicate that subsequent engagement is not explained by viral exposure alone, player interaction behaviors mediate 46% engagement within the viral game, whereas new game discovery mediates 52% engagement in other games. Finally, we compare the estimator with its alternatives and generalize the study design into a reusable framework for similar mediation studies across product domains.

stat.AP↗

Characterizing Underrepresentation when Generalizing Causal Survival Estimates

Randomized trial findings are routinely generalized to broader target populations to inform decisions in medicine and public policy. When the trial sample is misaligned with the target population, generalized effect estimates can yield suboptimal decisions. We characterize the target subgroups that are underrepresented in the trial cohort in time-to-event settings. Here, underrepresentation arises in two distinct ways: baseline underrepresentation, when a subgroup's covariate profile is rare in the trial relative to the target, and time-varying underrepresentation, when differential censoring erodes a subgroup over follow-up. Existing diagnostics either conflate these mechanisms or apply only to the baseline case. We show that the asymptotic variance of the efficient generalizability estimator admits a decomposition in which only two terms can diverge as representation weakens: a sampling-weight term capturing baseline misalignment and a censoring-weight term capturing differential loss to follow-up. Underrepresented subgroups are therefore those with imprecise generalized estimates, and the larger of the two terms gives the responsible channel. We use this decomposition in TRACE (Temporal Rashomon Analysis of CEnsored data), which combines a self-normalized variance objective with a tree-based search to localize underrepresented subgroups, attribute each to baseline mismatch or censoring, and find the follow-up intervals where censoring drives the variance inflation. We solve the joint problem with a parametric dynamic program nested inside a Rashomon ensemble of near-optimal trees. In synthetic benchmarks, TRACE recovers the planted underrepresented regions. In a case study generalizing the National Job Training Partnership Act (JTPA) Study to U.S. JTPA-eligible adults, restricting to reliably generalizable subgroups narrows the 95% confidence interval for the refined estimand by 60%.

stat.ME↗

An Inspectable LLM Council for Multi-Model Research Answer Aggregation

Long-form research questions often require answers that combine complementary details, reconcile conflicting estimates, and retain qualifications across several large language model outputs. We present Standard Council, a two-stage synthesis workflow in which an analyst organizes three independent candidate answers into a structured textual state and a separate writer composes the final response from the question and that state. We evaluate the workflow on 100 closed-book tasks from the Deep Research Accuracy, Completeness, and Objectivity benchmark, DRACO, retaining 600 final answers, 100 analyst states, and 720 whole-answer judgments. Under the primary rubric judge, Council scores 70.98, exceeding GPT-5.6 Sol by 6.18 points and Gemini 3.8 Flash by 8.34 points; its 0.47-point difference from Claude Opus 5 has a 95% paired interval of $[-0.69,1.61]$. In the four-arm fusion comparison, Council scores 1.87 points above direct fusion, with a 95% paired interval of $[0.33,3.48]$. It covers 51.1% of positive criteria met by exactly one candidate, compared with 43.7% for direct fusion. On an exploratory 68-task subset with matching returned model labels, the Council-minus-direct-fusion difference is $-1.69$ points, with a 95% interval of $[-2.80,-0.62]$. The full comparison characterizes workflows that differ in structure and generation budget; estimating the analyst stage's effect requires a controlled component study. Traced cases show how selected estimates, provenance qualifications, and selection decisions pass through the saved state into final answers.

cs.CL↗

From language-model stock rankings to testable economic rules: A computational audit

We test the stability, reproducibility and investment outcomes of language-model stock rankings. Four models and five numerical comparators share a portfolio engine over 72 monthly holding periods in the Shanghai Stock Exchange (SSE) 50, China Securities Index (CSI) 300 and CSI 500. Rankings use nine characteristics, and five repeated SSE 50 runs measure variation under identical inputs. Linear rules fitted to development-period model preferences are frozen before unseen-month, larger-pool and controlled-intervention tests. Their mean Spearman agreement with model rankings is 0.923-0.984 in the SSE 50 and 0.795-0.985 after transfer. Aggregate rank-change error falls relative to a zero-change prediction in twelve archived feature-group comparisons and eight matched single-feature comparisons, with Holm adjustments applied in separate nine-plus-three and six-plus-two families. Prediction of individual entries and exits remains weak (event Jaccard 0.000-0.125). Historical mean model compound annual growth rates range from 6.26% to 11.17%. At 10 basis points per side and six-month blocks, the twelve-comparison model-minus-rule return family and factor-controlled associations yield no adjusted finding. Three higher-cost, twelve-month-block comparisons favor a Terra rule within their twelve-test slices, with no adjusted finding across the full 144-test sensitivity grid. Two input-intervention batches totaling 5,184 responses supply paired intervention-return tests. The three-model batch has no bootstrap-adjusted finding at the primary block length; a Luna row-order effect appears under heteroskedasticity- and autocorrelation-consistent (HAC) adjustment within its three-test family but not in a pooled 24-test adjustment. Compact rules approximate aggregate rankings; return conclusions depend on comparison families, uncertainty methods and tie priorities.

cs.CE↗

Patch correctness in automated program repair: A five-dimensional evidence framework and cross-benchmark audit from classical repair to coding agents

Tests guide automated program repair and determine which candidate patches are reported as successful. Repository-level coding agents extend this feedback loop by generating tests and revising code after observing execution results. This technical report introduces a five-dimensional evidence profile and reporting protocol to audit how test observations support patch-correctness claims across classical repair systems and coding agents. We assemble an audit dataset of 39 benchmark and evaluation artifacts: 32 primary empirical studies, one institutional technical audit, and six prior taxonomic baselines. We code the 33 primary contributions along five dimensions: test provenance and exposure, behavioral scope, oracle basis and dependence, execution integrity, and label adjudication. The central distinction separates checks visible during repair search from evidence reserved for final assessment. Four worked comparative audits show how changed candidate populations, oracle rules, or execution budgets alter reported detection rates. Additional tests can expose missed faults while rejecting valid alternatives or reducing correct-repair yield under a fixed budget. Repository-level audits separate issue reproduction from candidate-patch discrimination and distinguish automated discrepancy detection from requirements-based adjudication. An evidence matrix links each coded decision to supporting passages and records shared benchmark ancestry. We specify a reporting protocol for repair benchmarks and maintenance pipelines that retains uncertain judgments and records validation cost alongside fault detection and valid-patch retention.

cs.SE↗

Clustering-based Transfer Learning for Dynamic Multimodal MultiObjective Evolutionary Algorithm

Dynamic multimodal multiobjective optimization presents the dual challenge of simultaneously tracking multiple equivalent pareto optimal sets and maintaining population diversity in time-varying environments. However, existing dynamic multiobjective evolutionary algorithms often neglect solution modality, whereas static multimodal multiobjective evolutionary algorithms lack adaptability to dynamic changes. To address above challenge, this paper makes two primary contributions. First, we introduce a new benchmark suite of dynamic multimodal multiobjective test functions constructed by fusing the properties of both dynamic and multimodal optimization to establish a rigorous evaluation platform. Second, we propose a novel algorithm centered on a Clustering-based Autoencoder prediction dynamic response mechanism, which utilizes an autoencoder model to process matched clusters to generate a highly diverse initial population. Furthermore, to balance the algorithm's convergence and diversity, we integrate an adaptive niching strategy into the static optimizer. Empirical analysis on 12 instances of dynamic multimodal multiobjective test functions reveals that, compared with several state-of-the-art dynamic multiobjective evolutionary algorithms and multimodal multiobjective evolutionary algorithms, our algorithm not only preserves population diversity more effectively in the decision space but also achieves superior convergence in the objective space.

cs.AI↗

Sensitivity Analysis when Generalizing Causal Effects from Multiple Studies to a Target Population: Motivation from the ECHO Program

Unobserved effect modifiers can induce bias when generalizing causal effect estimates to target populations. In this work, we extend a sensitivity analysis framework assessing the robustness of study results to unobserved effect modification that adapts to various generalizability scenarios, including multiple (conditionally) randomized trials, observational studies, or combinations thereof. This framework is interpretable and does not rely on distributional or functional assumptions about unknown parameters. We demonstrate how to leverage the multi-study setting to detect violation of the generalizability assumption through hypothesis testing, showing with simulations that the proposed test achieves high power under real-world sample sizes. Finally, we apply our sensitivity analysis framework to analyze the generalized effect estimate of secondhand smoke exposure on birth weight using cohort sites from the Environmental influences on Child Health Outcomes (ECHO) study.

stat.ME↗

Process-Supervised Reward Models for Verifying Clinical Note Generation: A Scalable Approach Guided by Domain Expertise

Process-supervised reward models (PRMs) excel at providing step-by-step verification for large language model (LLM) outputs in domains like mathematics and coding. However, their application to fields lacking ground-truth answers, such as clinical note generation, poses significant challenges. We introduce a novel framework for training PRMs to deliver step-level reward signals for LLM-generated clinical notes. By precisely defining meaningful "steps," injecting realistic "errors" informed by domain expertise, and leveraging LLMs to generate process supervision data at scale, we overcome previous limitations. Our PRM, built on LLaMA-3.1 8B, consistently outperforms proprietary reasoning and non-reasoning models, achieving state-of-the-art performance on two key evaluations: (1) distinguishing gold-standard from error-containing samples with 98.8% accuracy, and (2) selecting physician-preferred clinical notes with 56.2% accuracy. We investigate critical components for effective PRM training, including optimal loss functions and data selection strategies, and present a comprehensive physician reader study identifying predictors of downstream Best-of-N performance. Our study sheds light on unlocking the potential of PRMs for diverse generative tasks across domains.

cs.CL↗

Towards Adapting Open-Source Large Language Models for Expert-Level Clinical Note Generation

Proprietary Large Language Models (LLMs) such as GPT-4 and Gemini have demonstrated promising capabilities in clinical text summarization tasks. However, due to patient data privacy concerns and computational costs, many healthcare providers prefer using small, locally-hosted models over external generic LLMs. This study presents a comprehensive domain- and task-specific adaptation process for the open-source LLaMA-2 13 billion parameter model, enabling it to generate high-quality clinical notes from outpatient patient-doctor dialogues. Our process incorporates continued pretraining, supervised fine-tuning, and reinforcement learning from both AI and human feedback. We introduced a new approach, DistillDirect, for performing on-policy reinforcement learning with Gemini 1.0 Pro as the teacher model. Our resulting model, LLaMA-Clinic, can generate clinical notes comparable in quality to those authored by physicians. In a blinded physician reader study, the majority (92.8%) of individual evaluations rated the notes generated by LLaMA-Clinic as "acceptable" or higher across three criteria: real-world readiness, completeness, and accuracy. In the more challenging "Assessment and Plan" section, LLaMA-Clinic matched physician-authored notes in real-world readiness score. We highlight key considerations for future clinical note-generation tasks, emphasizing the importance of pre-defining a "best practice" note format, rather than relying on LLMs to determine this for clinical practice.

cs.CL↗

Bayesian Hyperbolic Multidimensional Scaling

Multidimensional scaling (MDS) is a widely used approach to representing high-dimensional, dependent data. MDS works by assigning each observation a location on a low-dimensional geometric manifold, with distance on the manifold representing similarity. We propose a Bayesian approach to multidimensional scaling when the low-dimensional manifold is hyperbolic. Using hyperbolic space facilitates representing tree-like structures common in many settings (e.g. text or genetic data with hierarchical structure). A Bayesian approach provides regularization that minimizes the impact of measurement error in the observed data and assesses uncertainty. We also propose a case-control likelihood approximation that allows for efficient sampling from the posterior distribution in larger data settings, reducing computational complexity from approximately $O(n^2)$ to $O(n)$. We evaluate the proposed method against state-of-the-art alternatives using simulations, canonical reference datasets, Indian village network data, and human gene expression data.

stat.ME↗

Spectral goodness-of-fit tests for complete and partial network data

Networks describe the, often complex, relationships between individual actors. In this work, we address the question of how to determine whether a parametric model, such as a stochastic block model or latent space model, fits a dataset well and will extrapolate to similar data. We use recent results in random matrix theory to derive a general goodness-of-fit test for dyadic data. We show that our method, when applied to a specific model of interest, provides an straightforward, computationally fast way of selecting parameters in a number of commonly used network models. For example, we show how to select the dimension of the latent space in latent space models. Unlike other network goodness-of-fit methods, our general approach does not require simulating from a candidate parametric model, which can be cumbersome with large graphs, and eliminates the need to choose a particular set of statistics on the graph for comparison. It also allows us to perform goodness-of-fit tests on partial network data, such as Aggregated Relational Data. We show with simulations that our method performs well in many situations of interest. We analyze several empirically relevant networks and show that our method leads to improved community detection algorithms. R code to implement our method is available on Github.

stat.ME↗