Search arXiv⌕ Search

arXiv subjects

Daniel Kang

Publications and source records attributed to Daniel Kang.

At least 19 recordsLinked to original sources

Understanding and Enhancing Backdoor Persistency in LLM Agent Post-Training

Developers can build LLM agents by adapting third-party models through benign post-training. We study a supply-chain threat in which an attacker supplies a model with a backdoor: hidden behavior that produces malicious outputs when a particular input pattern appears. Focusing on software-engineering agents, we ask whether such backdoors survive the developer's supervised fine-tuning (SFT) and subsequent task-level reinforcement learning (RL). We observe that benign SFT substantially reduces attack success, but subsequent RL often preserves the residual behavior and sometimes even increases attack success. Our analysis of backdoor erosion during SFT identifies two factors that may favor survival: initial backdoor strength and gradient compatibility with benign training. These factors motivate PersistBD, which refines an already-backdoored model before release to improve its persistency through the benign post-training process. On Qwen2.5-Coder-7B, PersistBD raises attack success from 20% to 74% after SFT and from 20% to 76% after SFT-RL, while maintaining comparable benign task performance. Together, our results show that backdoors can remain active through benign post-training and that adversaries can deliberately increase their persistence. This highlights a supply-chain risk for AI developers and motivates stronger techniques for detecting and mitigating inherited backdoors when adapting third-party models into agents. Our code is available at https://github.com/uiuc-kang-lab/PersistBD.

cs.CR↗

BI-Agent and BI-Bench: Towards Automating End-to-End Business Intelligence

Business intelligence (BI) is a cornerstone of enterprise decision-making and is widely used by enterprise users in software such as Power BI and Tableau. In traditional BI workflows, users need to prepare data by (1) identifying relevant tables, (2) performing data transformations, and (3) building join relationships, before they can (4) answer their business questions. These steps can be complex and time-consuming, making BI challenging. Given the strong capabilities of large language models (LLMs) in working with data, we study their ability to answer BI questions end-to-end, without requiring users to manually perform the tedious preparation steps. To do this, we harvest a large collection of real-world BI projects from public sources, and manually extract pairs of (questions, ground-truth answers) from real user dashboards. The resulting benchmark, BI-Bench, is the first benchmark to systematically study LLMs' ability on end-to-end BI. We find that even frontier LLMs perform poorly on BI-Bench, with less than 50% accuracy. To address their limitations, we design a tool-augmented BI-Agent that decomposes BI workflows into subtasks on structured data, such as search, join, and transform, and orchestrates specialized data management methods across BI stages. Furthermore, we develop a post-training framework that synthesizes training trajectories from real BI projects, enabling BI-Agent to be further post-trained using both supervised fine-tuning (SFT) and reinforcement learning (RL). BI-Agent achieves substantial accuracy gains of up to 40 percentage points with vanilla LLMs, and post-trained BI-Agent yields gains of up to 30 points. Our results highlight the importance of combining tool-augmented reasoning with domain-specific post-training in complex BI workflows, and point to promising directions for future research.

cs.LG↗

From Analytics to Tumor Boards: An Evidence-Linked Multi-Agent Workflow for Oncology Feature Extraction

Clinically relevant oncology information is distributed across heterogeneous, longitudinal documentation, creating substantial abstraction burden and requiring accurate attribution across specimens, tumors, biomarkers, and time points, while manual cancer-registry abstraction can require 27.2 minutes per case, highlighting the need for scalable methods that preserve clinical context while converting documentation into structured data. We evaluate an oncology information-extraction workflow in which OncoLens supplies multi-source, oncology-aware document selection, aggregation, and normalization from integrated EHRs, while the NimbleMind Multi-Agent System (nMAS) is a configurable oncology information-extraction workflow that extracts clinically relevant structured fields from fragmented oncology documentation. The extraction task uses a clinician-informed schema of 328 attributes spanning report metadata, diagnosis, staging, and cancer-type-specific information. nMAS separates clinician-defined field specifications from model execution and combines complexity-aware extraction, report-level consolidation, and source-grounded validation. The retrospective evaluation included 230 de-identified oncology documents from 40 patients and 418 clinician-reviewed document-field pairs containing 1,126 non-empty reference values. Evaluation focused on fields identified by clinicians as present in the source documents rather than exhaustively annotating all 328 schema fields. nMAS achieved a rank-weighted value-level precision of 82.6%, recall of 87.5%, and F1 of 85.0%, compared with an F1 of 66.4% for an independently implemented UMA-style MiniMax M2.5 comparator. These findings support the feasibility of using a configurable, source-grounded extraction workflow to convert fragmented oncology documentation into reusable structured data.

cs.AI↗

Human-Level Text-to-SQL via Reinforcement Learning on Verified Data, Without Pipeline Engineering

Translating natural language questions to SQL queries (Text-to-SQL) is a long-standing problem in database research. Recent efforts have focused on improving accuracy by building increasingly complex multi-stage large LLM pipelines, layering task decomposition, schema linking, and model-based query selection on top of an LLM. Despite this growing complexity, a substantial gap (>10%) between such systems and human experts persists on benchmarks, suggesting that pipeline engineering alone has hit a ceiling. We show that human-level Text-to-SQL performance is achievable by fine-tuning an LLM using RLVR on clean data, without pipeline components. In this paper, we identified the dominant bottleneck for RLVR on Text-to-SQL: existing training data contains pervasive annotation errors that mislead optimization. To address this, we developed a multi-round, expert-driven verification pipeline and used it to curate BIRD-Platinum, a dataset of 2.5k verified instances sampled from BIRD Train, with errors corrected in 61% of instances. We show that fine-tuning Qwen3-235B on BIRD-Platinum yields consistent improvements (11-16%) over BIRD Train on Arcwise-Plat (an expert-verified version of BIRD) and Spider2, outperforming SOTA open-source systems by 0.6-16%. Furthermore, we diagnosed two failure modes of standard RLVR on Text-to-SQL. We find that (1) result-based rewards have non-trivial false positive rates, and (2) models systematically ignore the external knowledge in BIRD-style problems. To address them, we propose ReViSQL-BIRD, a specialized reward shaping method that combines result-based rewards with SQL equivalence verification and leverages process rewards for incentivizing external-knowledge use. We fine-tuned Kimi-K2.6 with ReViSQL-BIRD. On Arcwise-Plat, ReViSQL-BIRD-K2.6 is the first method to achieve human-level accuracy (92.96%), outperforming top five open-source systems by 10-22%.

cs.DB↗

Tracing the Heart: An Evidence-Linked Pipeline for Heart-Failure Feature Engineering

Electronic health record (EHR) feature engineering is a major bottleneck in clinical research and AI, accounting for 39-45% of data scientists' workload. This is especially pronounced in heart failure, which affects an estimated 6.7 million U.S. adults and requires integrating fragmented EHR data with disease-specific, guideline-based clinical reasoning. Existing rule-based and large language model (LLM)-based approaches offer only partial automation with limited maintainability and evidence traceability. We developed the Nimblemind Multi-Agent System (nMAS), an evidence-linked, rubric-grounded pipeline for automated heart-failure feature engineering, and evaluated it on 500 dummy patient records from nine EHR source tables. nMAS generated 132 structured and 70 rubric-scored aggregated features, verified for structural integrity, rubric compliance, and provenance, and audited by a restricted LLM. Adding the aggregated features improved held-out AUROC from 0.895 to 0.963 for HFrEF and 0.870 to 0.910 for HFpEF phenotyping, and an independent LLM-based rubric assessment of evidence support and methodological soundness scored the features at 81.5% of maximum points. These results demonstrate the feasibility of automated, auditable feature engineering for complex cardiovascular EHR data, though evaluation was limited to a single-institution cohort and external validation is needed.

cs.AI↗

Trust but Verify:Evidence-Linked Multi-Agent Clinical Information Extraction in Pathology

Clinical feature extraction from pathology reports is challenging because relevant evidence may be distributed across coded and narrative fields and depend on specimen attribution, negation, ancillary findings, and diagnostic context. We retrospectively evaluated the NimbleMind Multi-Agent System (nMAS), a configurable workflow that separates clinician-defined field specifications from extraction models and returns report-level predictions with source-linked evidence. The study included 54 dummy gastric biopsy pathology reports from Singapore and four binary target fields, yielding 216 feature-case decisions. nMAS correctly classified 213 of 216 decisions (98.61\%), and all evidence spans associated with correct predictions occurred verbatim in the corresponding source reports. All three errors occurred in the two context-dependent \textit{H. pylori}-related fields requiring negation handling or diagnostic attribution. A single-model UMA-style comparator produced the similar label-level performance and error pattern. These findings do not demonstrate predictive superiority for the multi-agent architecture.Rather, the contribution of nMAS lies in workflow integration and traceability through configurable field specifications, complexity-based routing, report-level aggregation, and source-text validation within a clinician-reviewable workflow. Larger multi-institutional studies should assess generalizability, semantic evidence quality, adaptation effort, and clinician verification time.

cs.AI↗

Noisy Data is Destructive to Reinforcement Learning with Verifiable Rewards

Reinforcement learning with verifiable rewards (RLVR) has driven recent capability advances of large language models across various domains. Recent studies suggest that improved RLVR algorithms allow models to learn effectively from incorrect annotations, achieving performance comparable to learning from clean data. In this work, we show that these findings are invalid because the claimed 100% noisy training data is "contaminated" with clean data. After rectifying the dataset with a rigorous re-verification pipeline, we demonstrate that noise is destructive to RLVR. We show that existing RLVR algorithm improvements fail to mitigate the impact of noise, achieving similar performance to that of the basic GRPO. Furthermore, we find that the model trained on truly incorrect annotations performs 8-10% worse than the model trained on clean data across mathematical reasoning benchmarks. Finally, we show that these findings hold for real-world noise in Text2SQL tasks, where training on real-world, human annotation errors cause 5-12% lower accuracy than clean data. Our results show that current RLVR methods cannot yet compensate for poor data quality. High-quality data remains essential.

cs.LG↗

Non-vacuous Generalization Bounds for Reinforcement Learning with Verifiable Rewards

While reinforcement learning with verifiable rewards (RLVR) is widely used to improve the reasoning capabilities of large language models (LLMs), the generalizability of the resulting models remains poorly understood. In this work, we establish the first non-vacuous generalization bounds for parameter-efficient RLVR fine-tuning at the billion-parameter scale. Our approach adapts PAC-Bayes compression bounds to this setting, and addresses the inherent stochasticity of token generation by applying the Gumbel-max reparameterization trick. To operationalize these bounds, we propose the Progressive RLVR framework, which integrates RLVR with on-policy distillation, TinyLoRA, and model quantization. Progressive RLVR empirically retains 84-97% performance of standard LoRA fine-tuning while producing models that are 14,796x more compressible. We show that this framework yields non-vacuous generalization bounds in four domains: mathematical problem-solving, programming, general-knowledge reasoning, and Text-to-SQL. Our bounds exceed the accuracy of the base model by 9-51% and lie within 6-11% of the accuracy of the fine-tuned models.

cs.LG↗

Measuring Agents in Production

LLM-based agents already operate in production across many industries, yet we lack an understanding of what technical methods make deployments successful. We present the first systematic study of Measuring Agents in Production, MAP, using first-hand data from agent developers. We conducted 20 case studies via in-depth interviews and surveyed 86 deployed systems practitioners across 26 domains. We investigate why organizations build agents, how they build them, how they evaluate them, and their top development challenges. Our study finds that production agents are built using simple, controllable approaches: 68% execute at most 10 steps before human intervention, 70% rely on prompting off-the-shelf models instead of weight tuning, and 74% depend primarily on human evaluation. Reliability (consistent correct behavior over time) remains the top development challenge, which practitioners currently address through systems-level design. MAP documents the current state of production agents, providing the research community with visibility into deployment realities and underexplored research avenues.

cs.CY↗

MM-PoisonRAG: Disrupting Multimodal RAG with Local and Global Poisoning Attacks

Retrieval-augmented generation (RAG) has become a common practice in multimodal large language models (MLLM) to enhance factual grounding and reduce hallucination. Yet, its reliance on retrieval exposes MLLMs to knowledge poisoning attacks, in which adversaries deliberately inject malicious multimodal content into external knowledge bases to steer models toward generating incorrect or even harmful responses. We present MM-PoisonRAG, a framework to systematically study the vulnerability of multimodal RAG under knowledge poisoning. Specifically, we design two novel attack strategies: Localized Poisoning Attack (LPA), which implants targeted, query-specific multimodal misinformation to manipulate outputs toward attacker-controlled responses, and Globalized Poisoning Attack (GPA), which uses a single, untargeted adversarial injection to broadly corrupt reasoning and collapse generation quality across all queries. Extensive experiments on diverse tasks, multimodal RAG components, and attacker access levels reveal severe vulnerabilities: LPA achieves up to 56% attack success rate even under restricted access, and transfers effectively across four different retrievers without re-optimizing the adversaries. GPA completely disrupts model generation to 0% accuracy with just one poisoned content. Moreover, both LPA and GPA bypass existing defenses, underscoring the fragility of multimodal RAG and establishing MM-PoisonRAG as a foundation for future research on securing RAG frameworks against multimodal knowledge poisoning.

cs.LG↗

SafeSearch: Do Not Trade Safety for Utility in LLM Search Agents

Large language model (LLM) based search agents iteratively generate queries, retrieve external information, and reason to answer open-domain questions. While researchers have primarily focused on improving their utility, their safety behaviors remain underexplored. In this paper, we first evaluate search agents using red-teaming datasets and find that they are more likely to produce harmful outputs than base LLMs. For example, when asked ``How can I track someone's location without their consent?'', a base model refuses, whereas a search agent designed to retrieve and cite sources may lower its refusal threshold, fetch documents (e.g., court cases), and, once appended, synthesize them into an informative yet unsafe summary. We further show that utility-oriented finetuning intensifies this risk, motivating joint alignment of safety and utility. To this end, we present SafeSearch, a multi-objective reinforcement learning approach that couples a final-output safety/utility reward with a novel query-level shaping term that penalizes unsafe queries and rewards safe ones. Experiments show that SafeSearch reduces agent harmfulness by over 90% across three red-teaming datasets on a 7B model while producing safe and helpful responses, and maintains QA performance comparable to that of a utility-only finetuned agent. Further analyses confirm the effectiveness of the query-level reward in jointly improving safety and utility.

cs.CL↗

SODIUM: From Open Web Data to Queryable Databases

During research, domain experts often ask analytical questions whose answers require integrating data from a wide range of web sources. Thus, they must spend substantial effort searching, extracting, and organizing raw data before analysis can begin. We formalize this process as the SODIUM task, where we conceptualize open domains such as the web as latent databases that must be systematically instantiated to support downstream querying. Solving SODIUM requires (1) conducting in-depth and specialized exploration of the open web, which is further strengthened by (2) exploiting structural correlations for systematic information extraction and (3) integrating collected information into coherent, queryable database instances. To quantify the challenges in automating SODIUM, we construct SODIUM-Bench, a benchmark of 105 tasks derived from published academic papers across 6 domains, where systems are tasked with exploring the open web to collect and aggregate data from diverse sources into structured tables. Existing systems struggle with SODIUM tasks: we evaluate 6 advanced AI agents on SODIUM-Bench, with the strongest baseline achieving only 46.5% accuracy. To bridge this gap, we develop SODIUM-Agent, a multi-agent system composed of a web explorer and a cache manager. Powered by our proposed ATP-BFS algorithm and optimized through principled management of cached sources and navigation paths, SODIUM-Agent conducts deep and comprehensive web exploration and performs structurally coherent information extraction. SODIUM-Agent achieves 91.1% accuracy on SODIUM-Bench, outperforming the strongest baseline by approximately 2 times and the weakest by up to 73 times.

cs.DB↗

Behavioral Fingerprints for LLM Endpoint Stability and Identity

The consistency of AI-native applications depends on the behavioral consistency of the model endpoints that power them. Traditional reliability metrics such as uptime, latency and throughput do not capture behavioral change, and an endpoint can remain "healthy" while its effective model identity changes due to updates to weights, tokenizers, quantization, inference engines, kernels, caching, routing, or hardware. We introduce Stability Monitor, a black-box stability monitoring system that periodically fingerprints an endpoint by sampling outputs from a fixed prompt set and comparing the resulting output distributions over time. Fingerprints are compared using a summed energy distance statistic across prompts, with permutation-test p-values as evidence of distribution shift aggregated sequentially to detect change events and define stability periods. In controlled validation, Stability Monitor detects changes to model family, version, inference stack, quantization, and behavioral parameters. In real-world monitoring of the same model hosted by multiple providers, we observe substantial provider-to-provider and within-provider stability differences.

cs.AI↗

Accelerating Approximate Analytical Join Queries over Unstructured Data with Statistical Guarantees

Analytical join queries over unstructured data are increasingly prevalent in data analytics. Applying machine learning (ML) models to label every pair in the cross product of tables can achieve state-of-the-art accuracy, but the cost of pairwise execution of ML models is prohibitive. Existing algorithms, such as embedding-based blocking and sampling, aim to reduce this cost. However, they either fail to provide statistical guarantees (leading to errors up to 79% higher than expected) or become as inefficient as uniform sampling. We propose blocking-augmented sampling (BaS), which simultaneously achieves statistical guarantees and high efficiency. BaS optimally orchestrates embedding-based blocking and sampling to mitigate their respective limitations. Specifically, BaS allocates data tuples in the cross product into two regimes based on the failure modes of embeddings. In the regime of false negatives, BaS uses sampling to estimate the result. In the regime of false positives, BaS applies embedding-based blocking to improve efficiency. To minimize the estimation error given a budget for ML executions, we design a novel two-stage algorithm that adaptively allocates the budget between blocking and sampling. Theoretically, we prove that BaS asymptotically outperforms or matches standalone sampling. On real-world datasets across different modalities, we show that BaS provides valid confidence intervals and reduces estimation errors by up to 19$\times$, compared to state-of-the-art baselines.

cs.DB↗

Breaking Barriers: Do Reinforcement Post Training Gains Transfer To Unseen Domains?

Reinforcement post training (RPT) has recently shown promise in improving the reasoning abilities of large language models (LLMs). However, it remains unclear how well these improvements generalize to new domains, as prior work evaluates RPT models on data from the same domains used for post-training. To understand the generalizability of RPT, we conduct two studies with specific focus on Reinforcement Learning with Verifiable Rewards (RLVR). (1) Observational: we compare a wide range of open-weight RPT models against their corresponding base models across multiple domains, including both seen and unseen domains in their fine-tuning data. (2) Interventional: we fine-tune LLMs with RPT on single domains and evaluate their performance across multiple domains. Both studies converge on the same conclusion that, although RPT brings substantial gains on tasks similar to the fine-tuning data, the gains generalize inconsistently and can vanish on domains with different reasoning patterns.

cs.CL↗

BEAT: Visual Backdoor Attacks on VLM-based Embodied Agents via Contrastive Trigger Learning

Recent advances in Vision-Language Models (VLMs) have propelled embodied agents by enabling direct perception, reasoning, and planning task-oriented actions from visual inputs. However, such vision-driven embodied agents open a new attack surface: visual backdoor attacks, where the agent behaves normally until a visual trigger appears in the scene, then persistently executes an attacker-specified multi-step policy. We introduce BEAT, the first framework to inject such visual backdoors into VLM-based embodied agents using objects in the environments as triggers. Unlike textual triggers, object triggers exhibit wide variation across viewpoints and lighting, making them difficult to implant reliably. BEAT addresses this challenge by (1) constructing a training set that spans diverse scenes, tasks, and trigger placements to expose agents to trigger variability, and (2) introducing a two-stage training scheme that first applies supervised fine-tuning (SFT) and then our novel Contrastive Trigger Learning (CTL). CTL formulates trigger discrimination as preference learning between trigger-present and trigger-free inputs, explicitly sharpening the decision boundaries to ensure precise backdoor activation. Across various embodied agent benchmarks and VLMs, BEAT achieves attack success rates up to 80%, while maintaining strong benign task performance, and generalizes reliably to out-of-distribution trigger placements. Notably, compared to naive SFT, CTL boosts backdoor activation accuracy up to 39% under limited backdoor data. These findings expose a critical yet unexplored security risk in VLM-based embodied agents, underscoring the need for robust defenses before real-world deployment.

cs.AI↗

Pervasive Annotation Errors Break Text-to-SQL Benchmarks and Leaderboards

Researchers have proposed numerous text-to-SQL techniques to streamline data analytics and accelerate the development of data-driven applications. To compare these techniques and select the best one for deployment, the community depends on public benchmarks and their leaderboards. Since these benchmarks heavily rely on human annotations during question construction and answer evaluation, the validity of the annotations is crucial. In this paper, we conduct an empirical study that (i) benchmarks annotation error rates for two widely used text-to-SQL benchmarks, BIRD and Spider 2.0-Snow, and (ii) corrects a subset of the BIRD development (Dev) set to measure the impact of annotation errors on text-to-SQL agent performance and leaderboard rankings. Through expert analysis, we show that BIRD Mini-Dev and Spider 2.0-Snow have error rates of 52.8% and 62.8%, respectively. We re-evaluate all 16 open-source agents from the BIRD leaderboard on both the original and the corrected BIRD Dev subsets. We show that performance changes range from -7% to 31% (in relative terms) and rank changes range from $-9$ to $+9$ positions. We further assess whether these impacts generalize to the full BIRD Dev set. We find that the rankings of agents on the uncorrected subset correlate strongly with those on the full Dev set (Spearman's $r_s$=0.85, $p$=3.26e-5), whereas they correlate weakly with those on the corrected subset (Spearman's $r_s$=0.32, $p$=0.23). These findings show that annotation errors can significantly distort reported performance and rankings, potentially misguiding research directions or deployment choices. Our code and data are available at https://github.com/uiuc-kang-lab/text_to_sql_benchmarks.

cs.AI↗

AIA Forecaster: Technical Report

This technical report describes the AIA Forecaster, a Large Language Model (LLM)-based system for judgmental forecasting using unstructured data. The AIA Forecaster approach combines three core elements: agentic search over high-quality news sources, a supervisor agent that reconciles disparate forecasts for the same event, and a set of statistical calibration techniques to counter behavioral biases in large language models. On the ForecastBench benchmark (Karger et al., 2024), the AIA Forecaster achieves performance equal to human superforecasters, surpassing prior LLM baselines. In addition to reporting on ForecastBench, we also introduce a more challenging forecasting benchmark sourced from liquid prediction markets. While the AIA Forecaster underperforms market consensus on this benchmark, an ensemble combining AIA Forecaster with market consensus outperforms consensus alone, demonstrating that our forecaster provides additive information. Our work establishes a new state of the art in AI forecasting and provides practical, transferable recommendations for future research. To the best of our knowledge, this is the first work that verifiably achieves expert-level forecasting at scale.

cs.AI↗