Search arXiv⌕ Search

arXiv subjects

Rui Sun

Publications and source records attributed to Rui Sun.

At least 19 recordsLinked to original sources
arXiv2026

RubricAtlas: Mining Type-Level Rubrics from Expert Reports to Evaluate Deep Research Agents

Deep research agents conduct multi-step Web searches and synthesise retrieved information into analytical reports. Evaluating these outputs requires checking which analyses they deliver, beyond the presence of relevant facts. Query-specific binary rubrics make individual requirements explicit, but task-level checklists may leave the overall analytical structure underexamined. We introduce RubricAtlas, which derives type-level analytical requirements from expert reports on different objects within the same research type and organises detailed checks around recurring analytical questions. The requirements give related tasks a shared standard for what to analyse, without prescribing their answers. Our framework links binary checkpoints to report passages and evaluates completeness separately from argument validity. Built from 3,079 Chinese financial research reports, RubricAtlas covers 99 research types. Reference-report selection experiments show that individual reports cover only about half of their type's analytical questions, and that selecting requirements from different reports changes evaluation scores and agent rankings. Across ten agents, two judge pipelines achieve checkpoint agreement of $κ$ = 0.71 and agent-ranking agreement of Kendall $τ$ = 0.87. Controlled deletions reduce scores on the affected analyses, while padding produces no detectable increase. These findings support type-level rubrics for comparing the analytical completeness of Web deep research outputs, with individual decisions traceable to report evidence.

cs.AI↗
arXiv2026

AutoVend: Benchmarking LLM Agents on Long-Horizon Autonomous Vending Operations

Large language model (LLM) agents are extending agent-mediated commerce on the Web from shopping for consumers to operating businesses for sellers. In online-to-offline (O2O) commerce, decisions are issued through Web-based platforms but realized through costly physical execution. Evaluating such agents requires going beyond individual transactions to ask whether decisions made online turn into sustained economic returns. We introduce AutoVend, a benchmark in which one LLM agent manages 20 vending machines across five site types through a Web-based operating interface for 180 simulated days. Its data-grounded simulator combines a shared warehouse, procurement lead times, paid replenishment, inventory expiration, and recurring operating costs. We evaluate five LLM agents against a rule-based inventory-control baseline and diagnose their site coordination, execution, and performance over time. Four of five evaluated LLM agents fall short of this baseline, and returns per dollar of inference cost rank models differently from returns alone. Operational analyses show that gains in individual metrics do not consistently yield stronger economic growth, as inventory and execution costs erode these gains. Trajectory analysis further reveals difficulties in grounding feasible actions, managing inventory commitments, and converting feedback into effective policy revisions.

cs.AI↗
arXiv2026

LiveSearch: Questions as Executable Verifier Code for a Live Benchmark of Professional-Source Financial Search Agents

In finance, the answer that matters is usually the current one: prices, rates, and disclosures hold only as of a given day, so retrieving up-to-date facts is a core capability of a financial agent. Yet most financial benchmarks overlook timeliness. They fix each answer at annotation time, so the gold goes stale as markets move and, once leaked into training data, rewards recall rather than search and analysis. We introduce LiveSearch, a live benchmark that stores each question as executable verifier code: a typed program of calls to a professional financial data service, followed by a declarative extractor. The question is frozen, while its gold answer is determined by executing the program against professional financial data at evaluation time, and a daily job re-executes it with no human or language model in the loop. A memorized answer is thus likely outdated, and a model must reach the current one through tools and analysis. Construction is equally automatic: a language model only verbalizes programs enumerated from real tool responses and never chooses an answer. Under general web search, six contemporary tool-calling models reach 70.9% to 76.7% answer accuracy. Over three consecutive days, 70% of served questions change their gold value, and even under tolerance-aware matching, a literally reused earlier snapshot fails on a majority of questions within a single day. Per-question tracking uncovers dynamic shifts in correctness masked by aggregate accuracy.

cs.AI↗
arXiv2026

The Capacity of Private Information Retrieval from Coded Storage Systems with Colluding, Robust and Byzantine Servers

We consider the problem of private information retrieval from MDS-coded databases with $T$ colluding servers, $U$ robust servers, and $B$ Byzantine servers, referred to as U-B-MDS-TPIR. In the U-B-MDS-TPIR setting, $M$ files are stored across $N$ servers, and in this paper each file is independently encoded using an $(N,K)$ GRS code. We characterize the capacity among linear, full support-rank U-B-MDS-TPIR schemes as $C=\frac{N-U-2B}{N-U}(1+\frac{K+T-1}{N-U-2B}+\cdots+(\frac{K+T-1}{N-U-2B})^{M-1})^{-1}$ when $N-U-2B\ge KT$. When $K=1$ or $T=1$, the same expression gives the capacity without the linearity and full support-rank restrictions. We further extend the construction to the retrieval of $P$ out of $M$ files. For $P \ge \frac{M}{2}$, the resulting multi-file scheme achieves capacity when $K=1$ or $T=1$.

cs.IT↗
arXiv2026

Same Answer, Different Reasons: Trace-Level Divergence in Multilingual Large Language Models

Multilingual large language models are increasingly deployed in settings where visible reasoning traces support verification, education, and safety auditing, yet standard evaluations typically measure only final-answer agreement or holistic response similarity. As a result, they establish outcome consistency without determining whether users across languages receive comparable supporting evidence. We introduce a trace-level divergence framework that compares visible reasoning traces as structured sequences, decomposing agreement into structural similarity, order consistency, and holistic similarity, with pair-specific calibration to enable comparison across languages. Across three reasoning benchmarks, seven languages, and six frontier LLMs, the natural baseline---embedding each whole response and comparing the two by cosine---systematically overstates agreement: $52.1\%$--$67.3\%$ of examples in the top quartile of holistic similarity are not in the top quartile of structural similarity. This dissociation is stable across sentence encoders and step segmentations, and blind bilingual annotation studies in two English-anchored language pairs show that human judgments align with structural similarity rather than with the holistic baseline. A same-language control shows that the level of divergence is language-specific: two English traces for paraphrases of the same question agree more closely than an English trace and its counterparts in other languages. Our results show that answer-level and holistic evaluations can substantially overestimate the consistency of multilingual explanatory artifacts, motivating trace-level evaluation for reliable multilingual deployment.

cs.AI↗
arXiv2026

OpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory

Multimodal large language models (MLLMs) have made rapid progress on visual reasoning, yet many tasks require evidence that is difficult to obtain through internal reasoning alone: our controlled study on 22 benchmarks shows that simply increasing the inference budget of the base model yields only modest gains. Such tasks demand grounded perception, external knowledge, and numerical measurements obtained through heterogeneous tools, whose outputs must be made interpretable to the model and retained across reasoning steps. We introduce OpenVLHarness, a multimodal reasoning harness organized around a capability layer and persistent multimodal memory. The capability layer combines task-level tool interfaces with evidence rendering, providing consistent access to heterogeneous backends and converting their outputs into model-readable observations. Persistent multimodal memory retains the corresponding images, structured data, and web references as addressable artifacts throughout each query. A coding interface executes generated programs over stored data, enabling computations that combine results from multiple tools. Together, these mechanisms support an iterative process of acquiring, inspecting, and composing multimodal evidence. Evaluated on 22 benchmarks with eight open and proprietary backbones, OpenVLHarness improves average scores by 9.7-12.6 points over matched no-tool baselines, including +11.2 with Qwen3-VL-8B, +9.7 with Kimi K3, and +9.9 with GPT-6 Astra. With GPT-6 Luna at medium reasoning effort, OpenVLHarness outperforms the base model, Codex, and Visual Sketchpad at maximum effort by 7.7-8.5 points while incurring at most 37% of their model-token cost.

cs.CL↗
arXiv2026

Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents

Large Language Model (LLM) agents increasingly rely on external memory systems to accumulate experience across tasks. Yet nearly all existing approaches, from graph-structured memories to reflective insight stores, access memory through fixed, hand-designed heuristics. We argue that this static view of memory is a core bottleneck for agentic learning because optimal memory behavior is fundamentally context-dependent. The early stages of the tasks, benefit from minimal retrieval because memory is sparse; recurring goal types benefit from plan reuse rather than generic nearest-neighbor lookup; stuck agents benefit from re-retrieval with alternative queries; and across long task streams, the memory store itself must be consolidated and pruned to remain useful. We present Memory as a Controlled Process (MemCon), a framework that models memory operations as a Markov Decision Process and learns an online policy that adaptively decides when, what, and how much to retrieve, when to inject a distilled plan, and when to consolidate or forget. MemCon is backend-agnostic: it wraps any existing memory implementation, learns from task-by-task binary feedback with no pretraining and no additional LLM calls, and uses a lightweight tabular contextual bandit with UCB exploration that converges within tens of tasks. Across 6 benchmarks, 3 agent frameworks, and 3 LLM backbones, MemCon consistently outperforms multiple memory baselines by up to 15.2 points in task success while reducing token consumption by 5--20%.

cs.CL↗
arXiv2026

The ${\cal N}=1$ supersymmetric Pati-Salam models with extra $SU(2)_{L_2/R_2}$ gauge symmetry from intersecting D6-branes

By introducing an extra stack of D6-branes to standard ${\cal N}=1$ supersymmetric Pati-Salam models, we extend the landscape of its complete search. In this construction, the $d$-stack of D6-branes is introduced besides the standard $a,~b,~c$-stacks. More intersections from the extra stacks of D6-branes appear, and thus Higgs/Higgs-like particles arise from more origins. Among these models, we find eight new classes of ${\cal N}=1$ supersymmetric Pati-Salam models with gauge symmetries $SU(4)_C\times SU(2)_L\times SU(2)_{R_1}\times SU(2)_{R_2}$ and $SU(4)_C\times SU(2)_{L_1}\times SU(2)_{R}\times SU(2)_{L_2}$, where $d$-stack of D6-branes carries the gauge symmetries $SU(2)_{R_2}$ and $SU(2)_{L_2}$, respectively. The $SU(2)_{L_1/R_1} \times SU(2)_{L_2/R_2}$ can be broken down to the diagonal $SU(2)_{L/R}$ gauge symmetry via bifundamental Higgs fields. In such a way, we for the first time successfully constructed three-family supersymmetric Pati-Salam models from non-rigid D6-branes with extra $d$-stacks of D6-branes as visible sectors. Interestingly, by introducing extra stack of D6-branes to the standard supersymmetric Pati-Salam models, the number of filler brane reduces in general, and eventually the models without any $USp(N)$ gauge symmetry present. This reduces the exotic particles from filler brane intersection yet provides more vector-like particles from ${\cal N}=2$ subsector that are useful in renormalization group equation evolution as an advantage. Moreover, interesting degeneracy behavior with the same gauge coupling ratio exists in certain class of models.

hep-th↗
arXiv2026

SoK: Decentralized Agent Economic Infrastructure

Decentralized agent economies increasingly build a single task from protocols that were designed and secured separately. This creates a simple problem: a workflow can look correct at each step and still produce the wrong outcome. For example, a correct escrow may release payment on an authorized approval that provides little evidence that the delivered work actually satisfied the task. We systematize this problem across the full lifecycle of an agent task. Our study organizes security and economic requirements into 17 property families over six stages, with receipt soundness and completeness assessed separately. We examine 12 systems and standards, five reusable mechanism families, and four classical baselines. We introduce guarantee closure, a task-relative criterion for determining whether guarantees established at one stage remain available and constrain the later decisions that depend on them. We apply the criterion to controlled and native workflows, covering 840 matched executions and an exhaustive 11,648-case check over a finite objective-task domain. Our results expose recurring failures between verification and settlement, where conforming work can remain unaccepted or valid evidence can be ignored. Public records and model judgments further distinguish recorded approval from evidence of task conformance, while economic analysis identifies the report, penalty, and shared-error assumptions behind these guarantees. These findings show where end-to-end guarantees fail and what must be repaired to preserve them across the workflow.

cs.CR↗
arXiv2026

From Search to Research: Exploring Search Scaling in Autonomous Quantitative Factor Mining

Inference scaling has been shown to improve large language model (LLM) performance, and this principle naturally extends to autonomous LLM agents through increased search budgets, which we refer to as *search scaling*. Although prior work has characterized the mechanisms, scaling behavior, and performance limits of LLM inference scaling, much less is known about these questions in autonomous research. Therefore, we investigate how search scaling affects research performance and what mechanisms drive these gains using 50 quantitative factor-mining tasks grounded in financial research reports. Each task requires an agent to carry out an end-to-end research loop, from interpreting a hypothesis and implementing it in code to evaluating and iteratively refining the resulting factor. Across nine models, we examine how model capability, search depth, and search organization shape factor quality by tracing performance across varying budgets, transferring intermediate research states between models, and comparing different search strategies. We find that (1) initial performance is more strongly associated with model capability, while deeper search can narrow cross-model gaps; (2) model grafting shows that the early research state materially shapes final performance; and (3) parallel search outperforms sequential search under the same iteration budget, consistent with benefits from broader coverage of the search space. Further trajectory analysis shows that higher-performing models more effectively diagnose failures, revise search directions, and preserve the intended economic hypothesis when selecting candidates. These findings suggest that future progress in autonomous research will require stronger models together with adaptive policies for deploying test-time computation throughout the research process.

cs.AI↗
arXiv2026

One Analyst Is Not Ground Truth: Grading Agent-Built Financial Models Against Observed Professional Practice

Language model agents can now construct complete financial models, but it remains unclear whether they have learned the financial judgment that gives those models meaning. Existing benchmarks often anchor correctness to an expert-authored solution (e.g., numerical targets or detailed rubrics) which introduces a implicit assumption: \emph{one expert solution can serve as ground truth.} This is appropriate when finance provides a unique answer, but not when judgment is required. Evidence from professional practice challenges this assumption: When financial models built by different analysts for the same company are graded against one another, the median score is only 0.33 under standard tolerances, \emph{revealing that single-reference grading confounds professional disagreement with error.} We therefore introduce GAUGE, a benchmark that decomposes financial-model evaluation into deterministic checks, rubric judgments and numerical rules. Built from 1,001 professional valuation spanning 922 companies and all 25 GICS industry groups, GAUGE checks mechanical properties deterministically and evaluates judgment-bearing quantities against ranges in professional practice. We then validate GAUGE as a measurement instrument by testing expertise ordering, held-out professional values, and robustness to LM judgements. Our findings show that current LM agents are far better at constructing financial models than at deriving the company-specific assumptions that drive valuation, a gap that even persists after fine-tuning.

cs.LG↗
arXiv2026

Artificial intelligences and human scientists exhibit complementary strengths in theory building

We investigate the effectiveness of artificial intelligences (AI)-specifically large language models (LLMs)-relative to human scientists at high-level cognitive tasks in social science such as theory formulation, predictions of novel empirical results, and theory revision in response to new evidence. The research domain was academic discourse regarding gender and race inequality. Our findings, comparing 25 LLMs with 13 senior researchers and 60 doctoral scholars, reveal that the AIs outperformed most humans individually on most of the present tasks, while human theories were more diverse and exhibited greater gains in predictive accuracy from aggregation. AI-generated theories were more extensively elaborated, involving additional theoretical paths and latent variables, and were rated as higher quality than human theories by independent raters blinded to source. However, this theoretical complexity was in part ornamental, in that it was not associated with more accurate predictions about empirical patterns in data; in contrast, human scientists achieved greater predictive efficiency with simpler theories. The AIs were significantly more likely than human scientists to revise their theories to incorporate new evidence; human scientists updated their beliefs in a selective way that is sensitive to prior prediction errors. We speculate that the superior processing capacity of artificial intelligences makes them especially well-suited to tasks requiring grappling with complexity, but that the greater diversity of human ideas is essential to wise crowds and collective creativity.

cs.AI↗
arXiv2026

The Global Minimum Tax, Investment Incentives and Asymmetric Tax Competition

This paper investigates the OECD's global minimum tax (GMT) in a formal model of tax competition between asymmetric countries. We consider both profit shifting and real responses of multinational enterprises, and highlight the role of the substance-based income exclusion (SBIE) in investment incentives and tax rate setting. The GMT reduces the true tax rate differential and benefits the large country, while the revenue effect is generally ambiguous for the small country. In the short run where tax rates are fixed, the GMT reduces the small country's revenue if profit shifting costs are low and increases it otherwise. In the long run where countries adjust tax rates, the GMT reshapes the tax game and the competition pattern. We reveal that the minimum rate binds the small country only if it is low. With the rise of the GMT rate, countries will set tax rates below the minimum to boost capital investments and collect top-up taxes. Simulations show that a moderate GMT rate can raise both countries' revenues and the large country's welfare in the long run. However, it may reduce the small country's welfare if the welfare weight of private income is high.

econ.GN↗
arXiv2026

Modeling Clinical Workflow for SYNTAX Scoring from Coronary Angiography Videos

The SYNTAX score is a clinically established tool for assessing anatomical lesion complexity in coronary artery disease and guiding subsequent treatment. However, automated SYNTAX scoring is commonly formulated as a direct regression problem from coronary angiography videos to patient-level scores. In this work, we reformulate SYNTAX scoring as a vessel segment identity-preserving anatomical reasoning problem and propose a hierarchical modeling framework that explicitly aligns learning with the clinical workflow. Our approach maintains vessel segment identity across frames and views, estimates stenosis severity at the segment level, and aggregates evidence hierarchically according to coronary anatomy. Simultaneously, to address the scarcity of domain-specific data, we integrate and complete multiple public coronary angiography datasets, constructing a large-scale resource featuring completed vessel segmentation and derived structural annotations. Experiments demonstrate that vessel segment-level stenosis embedding enhances explanatory power and reduces prediction variability compared to baseline models, with the R^2 score improving by 0.201 and dev STD decreasing by 18.4%. These results highlight the necessity of structure-aligned modeling for reliable and stable automated SYNTAX scoring from multi-view coronary angiography videos. The GitHub link is https://github.com/VersaceSu7/SYNTAX_score_777.

cs.CV↗
arXiv2026

GraphSkillEvo: Evolutionary Optimization of Graph-Structured Agent Skills

Skills can improve the performance of Large Language Model (LLM) agents by providing task-specific procedural guidance, while skill optimization further improves their effectiveness through iterative refinement. However, existing skill optimization methods typically represent skills as unstructured natural-language instructions, creating two key challenges: 1) Unstructured skills often lack explicit workflow-level guidance and contain substantial redundancy, making them difficult for LLMs to execute; 2) the vast search space of unconstrained natural-language skills makes skill optimization ineffective. To address these challenges, we propose representing skills as graph-structured natural-language artifacts. In graph-structured skills, each node represents an execution step together with its operational guidance, while directed edges encode context-dependent transitions between steps. Compared to unstructured skills, graph-structured skills can provide clear workflow-level guidance. Moreover, the proposed graph-structured skill can also facilitate skill optimization. Building on this structured representation, we introduce GraphSkillEvo, a population-based evolutionary optimization framework with mutation and crossover operators for graph-structured skills. By maintaining multiple candidate skills and combining effective components, GraphSkillEvo enables broader and more comprehensive exploration of the structured skill space than purely LLM-based iterative self-refinement. Extensive experiments across five agent benchmarks demonstrate that GraphSkillEvo consistently outperforms the strong skill optimization baseline SkillOpt, improving average accuracy by 4.01% on GPT-5.4-nano and 1.76% on GPT-5.4. Our code is available at https://github.com/ruisun7/GraphSkillEvo.

cs.LG↗
arXiv2026

TRACE: Training Reasoning Agents for Causal Exploration with Synthesized Rewards

Reinforcement learning with verifiable rewards (RLVR) has advanced language-model reasoning in domains such as mathematics and code, where objective answers are inexpensive to check. Diagnostic reasoning over complex data lacks this advantage: establishing the true cause of an anomaly often requires costly expert investigation and may remain ambiguous after the fact. We ask whether this asymmetry of verification can instead be engineered. We sample an intervention, inject it into a controlled simulator, and generate the observations it would produce. The hidden intervention provides an oracle label and objective reward, while the agent must still investigate noisy, confounded, and distributed evidence. We instantiate this approach in TRACE, a digital-advertising diagnostic environment with 12 root causes and fine-grained segment attribution. Agents investigate each episode using Python and SQL and must identify both the root cause and, when applicable, the affected segment assignment. On a held-out 235-episode test set, the strongest prompted baseline, Claude Opus 5, reaches 0.686 FullAttr@1. Supervised fine-tuning raises Qwen3.5-35B-A3B from 0.159 to 0.637, and subsequent RL with synthesized rewards reaches 0.757, outperforming all evaluated prompted baselines, including frontier closed-source models and a prompted Qwen3.5-122B-A10B model. The resulting policy also uses substantially fewer tool calls than the prompted 35B base. These results provide evidence that access to a scalable, objective training signal can be a more important constraint than model scale alone. More broadly, simulation-based verification can make otherwise ambiguous diagnostic reasoning tasks amenable to scalable reinforcement learning.

cs.AI↗
arXiv2026

Alpha-R1: Alpha Screening with LLM Reasoning via Reinforcement Learning

Signal decay and regime shifts pose recurring challenges for data-driven investment strategies in non-stationary markets, where conventional time-series and machine learning approaches often struggle to generalize beyond historical correlations. While large language models (LLMs) offer strong capabilities for processing unstructured information, their potential to support quantitative factor screening through explicit economic reasoning remains underexplored. Existing factor-based methods typically reduce alphas to numerical time series, overlooking the semantic rationale that determines when a factor is economically relevant. We present Alpha-R1, an RL-aligned LLM framework for context-aware alpha screening. Its core mechanism, semantic gating, evaluates each candidate factor's semantic profile against a dynamically constructed market state description, selecting a sparse subset of factors whose economic rationale aligns with current market conditions. The selection model is trained via group relative policy optimization (GRPO), using realized portfolio returns as the primary reward signal. Under a 12-month out-of-sample evaluation, Alpha-R1 achieves annualized returns of 47.87% on S&P 500 and 40.57% on CSI 300 with Sharpe ratios of 1.62 and 2.23. These results, obtained under a bounded candidate-pool evaluation protocol, provide evidence for second-stage semantic factor reranking in non-stationary markets. The full implementation and resources are available at https://github.com/FinStep-AI/Alpha-R1.

q-fin.TR↗
arXiv2026

APEx: Distillation of Agent Procedural Experience for Adaptive Deep Research Question Answering

Deep research agents augment large language models with external tools to answer complex, long-horizon questions through multi-turn reasoning. Learning from prior experience is crucial for continual improvement, yet existing methods either retrieve verbose task-specific traces that burden decision-making, or distill procedural skills that remain decoupled from downstream policy adaptation. We propose APEx, a hierarchical experience utilization framework that organizes interaction history into instance-level trajectory memories and category-level procedural skills, and couples them through a closed-loop architecture of Executor, Distiller, and Planner. The three modules are optimized via a three-stage alternating GRPO training paradigm, enabling reward-guided skill distillation rather than fixed-prompt generation. At test time, distilled skills serve as procedural priors for online Planner adaptation through skill-guided test-time reinforcement learning, allowing ground-truth-free self-improvement with skill-alignment regularization to prevent policy drift. Experiments on 7 benchmarks demonstrate that APEx achieves state-of-the-art performance, surpassing GPT-5.4 by 14.7 points and the strongest memory-augmented baseline by 3.0 points.

cs.AI↗