arXiv subjects

Víctor Gutiérrez-Basulto

Publications and source records attributed to Víctor Gutiérrez-Basulto.

At least 19 recordsLinked to original sources

arXiv2026

AUDITFLOW: Executable Symbolic Environments for Structured Financial Reporting Verification

Structured financial reporting verification requires connecting reported facts to accounting constraints, reconstructing expected values, and executing numerical checks. We study this problem through XBRL numerical consistency verification and introduce \textsc{AuditFlow}, a graph-grounded multi-agent framework built around an executable symbolic environment. The environment links a reusable US-GAAP taxonomy graph with filing-specific XBRL evidence and provides typed tools for retrieval, traversal, computation, and rule evaluation. Two specialized auditors investigate constraints and filing evidence, while a senior auditor coordinates disagreement-triggered refinement. Deterministic tools produce verification evidence; evidential aggregation combines structured reports, including agent-declared overrides, into final decisions. On a newly collected 1{,}000-instance evaluation set of engine-recorded DQC violations, \textsc{AuditFlow} achieves $87.70\%$ joint audit accuracy with Claude Sonnet 4.6, compared with $68.60\%$ for Single Agent and $86.70\%$ for Checker-only. Component and coverage analyses identify executable verification as the main source of accuracy, while additional gains from agent judgment depend on the backbone and checker coverage.

cs.AI↗

arXiv2026

Recognition Is Not Grounding: Benchmarking Multilingual Financial Evidence Extraction

Comparing financial disclosures across languages requires identifying shared accounting concepts and recovering all numerical facts that express them. Existing benchmarks evaluate aspects of these capabilities, but provide limited insight into whether both succeed on the same multilingual evidence. In particular, it remains unclear whether language models can recognize these concepts across languages and exhaustively recover their corresponding numerical evidence from the same contexts. We introduce MULTIFINGROUND, a benchmark designed to investigate this recognition-grounding gap in multilingual financial reports. Constructed from FY2025 disclosures of 90 issuers in English, French, and Greek, MULTIFINGROUND aligns independently authored reports through a shared accounting taxonomy and provides traceable fact-level annotations. The benchmark evaluates financial fact identification, concept recognition, exhaustive fact selection, and concept discrimination without candidate labels. Across 15 evaluated models, eight API and open-weight reference models achieve 97.3-99.4\% concept recognition accuracy, yet miss at least one target fact in 26.6-41.9\% of correctly recognized groups. Further analyses reveal substantial variation in concept discrimination and limited gains from the evaluated finance-specialized models. Our findings demonstrate that strong concept recognition can coexist with incomplete evidence recovery, highlighting the importance of evaluating semantic recognition and evidence completeness together.

cs.AI↗

arXiv2026

Factorized Hypothesis Search for Evidence-to-Taxonomy Retrieval

Large-taxonomy retrieval often assumes that the input already expresses the target concept. In many settings, however, the input is indirect evidence, such as a table cell whose meaning depends on its row, column, datatype, and context. We call this mismatch the retrieval readiness gap. Our analysis shows that the current index retrieves the target reliably when its semantics are explicit, while raw evidence often leaves it deep in the ranking. We propose Factorized Hypothesis Search (FHS), which maintains multiple partial interpretations over named semantic dimensions. These hypotheses support structured query rendering, multi-hypothesis retrieval, and dimension-level candidate verification. On both financial taxonomy tagging and CodiEsp clinical coding tasks, FHS achieves the best Recall@1, MRR, and final accuracy among the non-oracle methods. Replacing the factorized hypothesis path with a free-text ensemble causes the largest drop in head-ranking performance, while sequential refinement provides no additional gain over FHS's strong parallel first round.

cs.CL↗

arXiv2026

Herculean: An Agentic Benchmark for Financial Intelligence

As AI agents improve, the central question is no longer whether they can solve isolated well-defined financial tasks, but whether they can reliably carry out financial professional work. Existing financial benchmarks offer only a partial view of this ability, as they primarily evaluate static competencies such as question answering, retrieval, summarization, and classification. We introduce Herculean, the first skilled benchmark for agentic financial intelligence spanning four representative workflows, including Trading, Hedging, Market Insights, and Auditing. Each workflow is instantiated as a standardized MCP-based skill environment with its own tools, interaction dynamics, constraints, and success criteria, enabling consistent end-to-end assessment of heterogeneous agent systems. Across frontier agents, we find agents perform relatively well on Trading and Market Insights, but struggle substantially on Hedging and Auditing, where long-horizon coordination, state consistency, and structured verification are critical. Overall, our results point to a key gap in current agents in turning financial reasoning into dependable workflow execution in high-stakes financial workflows.

cs.AI↗

arXiv2026

FinAuditing: A Financial Taxonomy-Structured Multi-Document Benchmark for Evaluating LLMs

Going beyond simple text processing, financial auditing requires detecting semantic, structural, and numerical inconsistencies across large-scale disclosures. As financial reports are filed in XBRL, a structured XML format governed by accounting standards, auditing becomes a structured information extraction and reasoning problem involving concept alignment, taxonomy-defined relations, and cross-document consistency. Although large language models (LLMs) show promise on isolated financial tasks, their capability in professional-grade auditing remains unclear. We introduce FinAuditing, a taxonomy-aligned, structure-aware benchmark built from real XBRL filings. It contains 1,102 annotated instances averaging over 33k tokens and defines three tasks: Financial Semantic Matching (FinSM), Financial Relationship Extraction (FinRE), and Financial Mathematical Reasoning (FinMR). Evaluations of 13 state-of-the-art LLMs reveal substantial gaps in concept retrieval, taxonomy-aware relation modeling, and consistent cross-document reasoning. These findings highlight the need for realistic, structure-aware benchmarks. We release the evaluation code at https://github.com/The-FinAI/FinAuditing and the dataset at https://huggingface.co/collections/TheFinAI/finauditing. The task currently serves as the official benchmark of an ongoing public evaluation contest at https://open-finance-lab.github.io/SecureFinAI_Contest_2026/.

cs.CL↗

arXiv2026

Plan Then Retrieve: Reinforcement Learning-Guided Complex Reasoning over Knowledge Graphs

Knowledge Graph Question Answering aims to answer natural language questions by reasoning over structured knowledge graphs. While large language models have advanced KGQA through their strong reasoning capabilities, existing methods continue to struggle to fully exploit both the rich knowledge encoded in KGs and the reasoning capabilities of LLMs, particularly in complex scenarios. They often assume complete KG coverage and lack mechanisms to judge when external information is needed, and their reasoning remains locally myopic, failing to maintain coherent multi-step planning, leading to reasoning failures even when relevant knowledge exists. We propose Graph-RFT, a novel two-stage reinforcement fine-tuning KGQA framework with a 'plan-KGsearch-and-Websearch-during-think' paradigm, that enables LLMs to perform autonomous planning and adaptive retrieval scheduling across KG and web sources under incomplete knowledge conditions. Graph-RFT introduces a chain-of-thought fine-tuning method with a customized plan-retrieval dataset activates structured reasoning and resolves the GRPO cold-start problem. It then introduces a novel plan-retrieval guided reinforcement learning process integrates explicit planning and retrieval actions with a multi-reward design, enabling coverage-aware retrieval scheduling. It employs a Cartesian-inspired planning module to decompose complex questions into ordered subquestions, and logical expression to guide tool invocation for globally consistent multi-step reasoning. This reasoning retrieval process is optimized with a multi-reward combining outcome and retrieval specific signals, enabling the model to learn when and how to combine KG and web retrieval effectively.

cs.AI↗

arXiv2025

Consistency-Aware Editing for Entity-level Unlearning in Language Models

Large language models (LLMs) risk retaining sensitive, copyrighted, or harmful information from their training data. Entity-level unlearning addresses this issue by removing all knowledge of a specific entity while preserving the model's overall capabilities. Existing approaches typically rely on full-model fine-tuning or prompt-based interventions, which can be computationally expensive or brittle when handling paraphrased queries. Recently, model editing has emerged as an efficient alternative for updating knowledge in LLMs, offering a promising direction for unlearning. However, existing editing techniques are typically designed for instance-level updates, modifying responses to specific attributes of an entity rather than eliminating all knowledge associated with the entity. In this paper, we investigate how editing techniques can be adapted for effective and efficient entity-level unlearning. To this end, we introduce a novel consistency-aware editing (CAE) framework. CAE aggregates a diverse set of prompts related to a target entity, including its attributes, relations, and adversarial paraphrases. It then jointly learns a low-rank update guided by a consistency regularizer that aligns the editing directions across prompts. This promotes robust and comprehensive forgetting while minimizing interference with unrelated knowledge. We further examine where different entities are stored within the model and how many diverse prompts are needed for successful unlearning. We evaluate CAE on two challenging benchmarks, RWKU and ToFU, and demonstrate that it (i) provides insights into how entity-level knowledge is internally represented and deleted in LLMs, (ii) significantly improves forgetting accuracy and robustness over traditional unlearning and editing baselines, and (iii) enables scalable entity removal using only tens of carefully selected prompts.

cs.CL↗

arXiv2025

Uncovering and Mitigating Transient Blindness in Multimodal Model Editing

Multimodal Model Editing (MMED) aims to correct erroneous knowledge in multimodal models. Existing evaluation methods, adapted from textual model editing, overstate success by relying on low-similarity or random inputs, obscure overfitting. We propose a comprehensive locality evaluation framework, covering three key dimensions: random-image locality, no-image locality, and consistent-image locality, operationalized through seven distinct data types, enabling a detailed and structured analysis of multimodal edits. We introduce De-VQA, a dynamic evaluation for visual question answering, uncovering a phenomenon we term transient blindness, overfitting to edit-similar text while ignoring visuals. Token analysis shows edits disproportionately affect textual tokens. We propose locality-aware adversarial losses to balance cross-modal representations. Empirical results demonstrate that our approach consistently outperforms existing baselines, reducing transient blindness and improving locality by 17% on average.

cs.LG↗

arXiv2025

Advances in Logic-Based Entity Resolution: Enhancing ASPEN with Local Merges and Optimality Criteria

In this paper, we present ASPEN+, which extends an existing ASP-based system, ASPEN,for collective entity resolution with two important functionalities: support for local merges and new optimality criteria for preferred solutions. Indeed, ASPEN only supports so-called global merges of entity-referring constants (e.g. author ids), in which all occurrences of matched constants are treated as equivalent and merged accordingly. However, it has been argued that when resolving data values, local merges are often more appropriate, as e.g. some instances of 'J. Lee' may refer to 'Joy Lee', while others should be matched with 'Jake Lee'. In addition to allowing such local merges, ASPEN+ offers new optimality criteria for selecting solutions, such as minimizing rule violations or maximising the number of rules supporting a merge. Our main contributions are thus (1) the formalisation and computational analysis of various notions of optimal solution, and (2) an extensive experimental evaluation on real-world datasets, demonstrating the effect of local merges and the new optimality criteria on both accuracy and runtime.

cs.DB↗

arXiv2025

HyperMono: A Monotonicity-aware Approach to Hyper-Relational Knowledge Representation

In a hyper-relational knowledge graph (HKG), each fact is composed of a main triple associated with attribute-value qualifiers, which express additional factual knowledge. The hyper-relational knowledge graph completion (HKGC) task aims at inferring plausible missing links in a HKG. Most existing approaches to HKGC focus on enhancing the communication between qualifier pairs and main triples, while overlooking two important properties that emerge from the monotonicity of the hyper-relational graphs representation regime. Stage Reasoning allows for a two-step reasoning process, facilitating the integration of coarse-grained inference results derived solely from main triples and fine-grained inference results obtained from hyper-relational facts with qualifiers. In the initial stage, coarse-grained results provide an upper bound for correct predictions, which are subsequently refined in the fine-grained step. More generally, Qualifier Monotonicity implies that by attaching more qualifier pairs to a main triple, we may only narrow down the answer set, but never enlarge it. This paper proposes the HyperMono model for hyper-relational knowledge graph completion, which realizes stage reasoning and qualifier monotonicity. To implement qualifier monotonicity HyperMono resorts to cone embeddings. Experiments on three real-world datasets with three different scenario conditions demonstrate the strong performance of HyperMono when compared to the SoTA.

cs.AI↗

arXiv2025

Multi-level Mixture of Experts for Multimodal Entity Linking

Multimodal Entity Linking (MEL) aims to link ambiguous mentions within multimodal contexts to associated entities in a multimodal knowledge base. Existing approaches to MEL introduce multimodal interaction and fusion mechanisms to bridge the modality gap and enable multi-grained semantic matching. However, they do not address two important problems: (i) mention ambiguity, i.e., the lack of semantic content caused by the brevity and omission of key information in the mention's textual context; (ii) dynamic selection of modal content, i.e., to dynamically distinguish the importance of different parts of modal information. To mitigate these issues, we propose a Multi-level Mixture of Experts (MMoE) model for MEL. MMoE has four components: (i) the description-aware mention enhancement module leverages large language models to identify the WikiData descriptions that best match a mention, considering the mention's textual context; (ii) the multimodal feature extraction module adopts multimodal feature encoders to obtain textual and visual embeddings for both mentions and entities; (iii)-(iv) the intra-level mixture of experts and inter-level mixture of experts modules apply a switch mixture of experts mechanism to dynamically and adaptively select features from relevant regions of information. Extensive experiments demonstrate the outstanding performance of MMoE compared to the state-of-the-art. MMoE's code is available at: https://github.com/zhiweihu1103/MEL-MMoE.

cs.CV↗

arXiv2025

BUCA: A Binary Classification Approach to Unsupervised Commonsense Question Answering

Unsupervised commonsense reasoning (UCR) is becoming increasingly popular as the construction of commonsense reasoning datasets is expensive, and they are inevitably limited in their scope. A popular approach to UCR is to fine-tune language models with external knowledge (e.g., knowledge graphs), but this usually requires a large number of training examples. In this paper, we propose to transform the downstream multiple choice question answering task into a simpler binary classification task by ranking all candidate answers according to their reasonableness. To this end, for training the model, we convert the knowledge graph triples into reasonable and unreasonable texts. Extensive experimental results show the effectiveness of our approach on various multiple choice question answering benchmarks. Furthermore, compared with existing UCR approaches using KGs, ours is less data hungry. Our code is available at https://github.com/probe2/BUCA.

cs.CL↗

arXiv2025

N2C2: Nearest Neighbor Enhanced Confidence Calibration for Cross-Lingual In-Context Learning

Recent advancements of in-context learning (ICL) show language models can significantly improve their performance when demonstrations are provided. However, little attention has been paid to model calibration and prediction confidence of ICL in cross-lingual scenarios. To bridge this gap, we conduct a thorough analysis of ICL for cross-lingual sentiment classification. Our findings suggest that ICL performs poorly in cross-lingual scenarios, exhibiting low accuracy and presenting high calibration errors. In response, we propose a novel approach, N2C2, which employs a -nearest neighbors augmented classifier for prediction confidence calibration. N2C2 narrows the prediction gap by leveraging a datastore of cached few-shot instances. Specifically, N2C2 integrates the predictions from the datastore and incorporates confidence-aware distribution, semantically consistent retrieval representation, and adaptive neighbor combination modules to effectively utilize the limited number of supporting instances. Evaluation on two multilingual sentiment classification datasets demonstrates that N2C2 outperforms traditional ICL. It surpasses fine tuning, prompt tuning and recent state-of-the-art methods in terms of accuracy and calibration errors.

cs.CL↗

arXiv2025

MiCEval: Unveiling Multimodal Chain of Thought's Quality via Image Description and Reasoning Steps

Multimodal Chain of Thought (MCoT) is a popular prompting strategy for improving the performance of multimodal large language models (MLLMs) across a range of complex reasoning tasks. Despite its popularity, there is a notable absence of automated methods for evaluating the quality of reasoning steps in MCoT. To address this gap, we propose Multimodal Chain-of-Thought Evaluation (MiCEval), a framework designed to assess the correctness of reasoning chains by evaluating the quality of both the description and each reasoning step. The evaluation of the description component focuses on the accuracy of the image descriptions, while the reasoning step evaluates the quality of each step as it is conditionally generated based on the preceding steps. MiCEval is built upon a fine-grained dataset with annotations that rate each step according to correctness, relevance, and informativeness. Extensive experiments on four state-of-the-art MLLMs show that step-wise evaluations using MiCEval align more closely with human judgments compared to existing methods based on cosine similarity or fine-tuning approaches. MiCEval datasets and code can be found in https://github.com/alenai97/MiCEval.

cs.CL↗

arXiv2025

BLEnD: A Benchmark for LLMs on Everyday Knowledge in Diverse Cultures and Languages

Large language models (LLMs) often lack culture-specific knowledge of daily life, especially across diverse regions and non-English languages. Existing benchmarks for evaluating LLMs' cultural sensitivities are limited to a single language or collected from online sources such as Wikipedia, which do not reflect the mundane everyday lifestyles of diverse regions. That is, information about the food people eat for their birthday celebrations, spices they typically use, musical instruments youngsters play, or the sports they practice in school is common cultural knowledge but uncommon in easily collected online sources, especially for underrepresented cultures. To address this issue, we introduce BLEnD, a hand-crafted benchmark designed to evaluate LLMs' everyday knowledge across diverse cultures and languages. BLEnD comprises 52.6k question-answer pairs from 16 countries/regions, in 13 different languages, including low-resource ones such as Amharic, Assamese, Azerbaijani, Hausa, and Sundanese. We construct the benchmark to include two formats of questions: short-answer and multiple-choice. We show that LLMs perform better for cultures that are highly represented online, with a maximum 57.34% difference in GPT-4, the best-performing model, in the short-answer format. For cultures represented by mid-to-high-resource languages, LLMs perform better in their local languages, but for cultures represented by low-resource languages, LLMs perform better in English than the local languages. We make our dataset publicly available at: https://github.com/nlee0212/BLEnD.

cs.CL↗

arXiv2024

Multi-level Matching Network for Multimodal Entity Linking

Multimodal entity linking (MEL) aims to link ambiguous mentions within multimodal contexts to corresponding entities in a multimodal knowledge base. Most existing approaches to MEL are based on representation learning or vision-and-language pre-training mechanisms for exploring the complementary effect among multiple modalities. However, these methods suffer from two limitations. On the one hand, they overlook the possibility of considering negative samples from the same modality. On the other hand, they lack mechanisms to capture bidirectional cross-modal interaction. To address these issues, we propose a Multi-level Matching network for Multimodal Entity Linking (M3EL). Specifically, M3EL is composed of three different modules: (i) a Multimodal Feature Extraction module, which extracts modality-specific representations with a multimodal encoder and introduces an intra-modal contrastive learning sub-module to obtain better discriminative embeddings based on uni-modal differences; (ii) an Intra-modal Matching Network module, which contains two levels of matching granularity: Coarse-grained Global-to-Global and Fine-grained Global-to-Local, to achieve local and global level intra-modal interaction; (iii) a Cross-modal Matching Network module, which applies bidirectional strategies, Textual-to-Visual and Visual-to-Textual matching, to implement bidirectional cross-modal interaction. Extensive experiments conducted on WikiMEL, RichpediaMEL, and WikiDiverse datasets demonstrate the outstanding performance of M3EL when compared to the state-of-the-art baselines.

cs.CV↗

arXiv2024

ASPEN: ASP-Based System for Collective Entity Resolution

In this paper, we present ASPEN, an answer set programming (ASP) implementation of a recently proposed declarative framework for collective entity resolution (ER). While an ASP encoding had been previously suggested, several practical issues had been neglected, most notably, the question of how to efficiently compute the (externally defined) similarity facts that are used in rule bodies. This leads us to propose new variants of the encodings (including Datalog approximations) and show how to employ different functionalities of ASP solvers to compute (maximal) solutions, and (approximations of) the sets of possible and certain merges. A comprehensive experimental evaluation of ASPEN on real-world datasets shows that the approach is promising, achieving high accuracy in real-life ER scenarios. Our experiments also yield useful insights into the relative merits of different types of (approximate) ER solutions, the impact of recursion, and factors influencing performance.

cs.DB↗

arXiv2024

Instances and Labels: Hierarchy-aware Joint Supervised Contrastive Learning for Hierarchical Multi-Label Text Classification

Hierarchical multi-label text classification (HMTC) aims at utilizing a label hierarchy in multi-label classification. Recent approaches to HMTC deal with the problem of imposing an over-constrained premise on the output space by using contrastive learning on generated samples in a semi-supervised manner to bring text and label embeddings closer. However, the generation of samples tends to introduce noise as it ignores the correlation between similar samples in the same batch. One solution to this issue is supervised contrastive learning, but it remains an underexplored topic in HMTC due to its complex structured labels. To overcome this challenge, we propose $\textbf{HJCL}$, a $\textbf{H}$ierarchy-aware $\textbf{J}$oint Supervised $\textbf{C}$ontrastive $\textbf{L}$earning method that bridges the gap between supervised contrastive learning and HMTC. Specifically, we employ both instance-wise and label-wise contrastive learning techniques and carefully construct batches to fulfill the contrastive learning objective. Extensive experiments on four multi-path HMTC datasets demonstrate that HJCL achieves promising results and the effectiveness of Contrastive Learning on HMTC.

cs.CL↗