arXiv subjects

Jiannan Xu

Publications and source records attributed to Jiannan Xu.

8 recordsLinked to original sources

arXiv2026

Co-Skill: A Collaborative Communication Framework for Skill Evolution

Agent evolution through skills becomes critical for LLM-based agents to iteratively improve task success rate. Hybrid evolution is a cost-efficient paradigm where a cloud LLM analyzes and generates skills while an edge SLM executes and internalizes them. However, existing hybrid methods, such as SkillRL, still suffer from low success rate and high token usage. We find this stems from blind communication: the cloud cannot perceive the edge's execution capability, while the edge does not understand the cloud's analysis needs. We thus propose the Collaborative Communication Framework (CCF) to achieve effective edge-cloud evolution. CCF is realized via three techniques: (1) a cloud-aware prefix-merged trajectory trie where the edge compresses trajectories by merging shared prefixes and pinpointing divergence points for efficient cloud analysis, (2) an edge-aware progressive skill tree where the cloud progressively builds a hierarchical skill tree to match edge SLM execution capability, and (3) a collaborative skill evolution scheme upon these two trees that evolves cloud LLM and edge SLM in a separated way to jointly improve task success rate. Experiments across ALFWorld and WebShop show that CCF reduces LLM+SLM tokens by 15.6%--41.9% over state-of-the-art hybrid methods while consistently improving 25.8%--76.4% task success rate.

cs.DC↗

arXiv2026

Prompt Injection in Automated Résumé Screening with Large Language Models: Single and Multi-Injection Settings

Large language models (LLMs) are increasingly used to screen and rank job applicants, creating incentives for candidates to strategically manipulate algorithmic hiring systems. We study prompt injection in automated résumé screening, defined as subtle self-promotional text that introduces no new qualifications but is designed to influence LLM evaluations. Using controlled experiments, we show that prompt injection reliably improves applicant rankings when résumé quality is homogeneous and few candidates inject. However, its effectiveness rapidly diminishes as more candidates inject, collapsing when manipulation becomes widespread. When candidate quality is heterogeneous, prompt injection is less effective on average, but can occasionally allow lower-quality candidates to outrank higher-quality ones, raising fairness concerns. Overall, LLM-based screening is most vulnerable when manipulation is rare and candidate quality differences are small. Code and resources are publicly available at: https://github.com/preetb1199/Prompt_Injection_ACL26

cs.AI↗

arXiv2026

AI Self-preferencing in Algorithmic Hiring: Empirical Evidence and Insights

As artificial intelligence (AI) tools become widely adopted, large language models (LLMs) are increasingly involved on both sides of decision-making processes, ranging from hiring to content moderation. This dual adoption raises a critical question: do LLMs systematically favor content that resembles their own outputs? Prior research in computer science has identified self-preference bias -- the tendency of LLMs to favor their own generated content -- but its real-world implications have not been empirically evaluated. We focus on the hiring context, where job applicants often rely on LLMs to refine resumes, while employers deploy them to screen those same resumes. Using a large-scale controlled resume correspondence experiment, we find that LLMs consistently prefer resumes generated by themselves over those written by humans or produced by alternative models, even when content quality is controlled. The bias against human-written resumes is particularly substantial, with self-preference bias ranging from 67% to 82% across major commercial and open-source models. To assess labor market impact, we simulate realistic hiring pipelines across 24 occupations. These simulations show that candidates using the same LLM as the evaluator are 23% to 60% more likely to be shortlisted than equally qualified applicants submitting human-written resumes, with the largest disadvantages observed in business-related fields such as sales and accounting. We further demonstrate that this bias can be reduced by more than 50% through simple interventions targeting LLMs' self-recognition capabilities. These findings highlight an emerging but previously overlooked risk in AI-assisted decision making and call for expanded frameworks of AI fairness that address not only demographic-based disparities, but also biases in AI-AI interactions.

cs.CY↗

arXiv2025

Which Demographic Features Are Relevant for Individual Fairness Evaluation of U.S. Recidivism Risk Assessment Tools?

Despite its constitutional relevance, the technical ``individual fairness'' criterion has not been operationalized in U.S. state or federal statutes/regulations. We conduct a human subjects experiment to address this gap, evaluating which demographic features are relevant for individual fairness evaluation of recidivism risk assessment (RRA) tools. Our analyses conclude that the individual similarity function should consider age and sex, but it should ignore race.

cs.CY↗

arXiv2025

Effort-aware Fairness: Incorporating a Philosophy-informed, Human-centered Notion of Effort into Algorithmic Fairness Metrics

Although popularized AI fairness metrics, e.g., demographic parity, have uncovered bias in AI-assisted decision-making outcomes, they do not consider how much effort one has spent to get to where one is today in the input feature space. However, the notion of effort is important in how Philosophy and humans understand fairness. We propose a philosophy-informed approach to conceptualize and evaluate Effort-aware Fairness (EaF), grounded in the concept of Force, which represents the temporal trajectory of predictive features coupled with inertia. Besides theoretical formulation, our empirical contributions include: (1) a pre-registered human subjects experiment, which shows that for both stages of the (individual) fairness evaluation process, people consider the temporal trajectory of a predictive feature more than its aggregate value; (2) pipelines to compute Effort-aware Individual/Group Fairness in the criminal justice and personal finance contexts. Our work may enable AI model auditors to uncover and potentially correct unfair decisions against individuals who have spent significant efforts to improve but are still stuck with systemic disadvantages outside their control.

cs.AI↗

arXiv2025

How May U.S. Courts Scrutinize Their Recidivism Risk Assessment Tools? Contextualizing AI Fairness Criteria on a Judicial Scrutiny-based Framework

The AI/HCI and legal communities have developed largely independent conceptualizations of fairness. This conceptual difference hinders the potential incorporation of technical fairness criteria (e.g., procedural, group, and individual fairness) into sustainable policies and designs, particularly for high-stakes applications like recidivism risk assessment. To foster common ground, we conduct legal research to identify if and how technical AI conceptualizations of fairness surface in primary legal sources. We find that while major technical fairness criteria can be linked to constitutional mandates such as ``Due Process'' and ``Equal Protection'' thanks to judicial interpretation, several challenges arise when operationalizing them into concrete statutes/regulations. These policies often adopt procedural and group fairness but ignore the major technical criterion of individual fairness. Regarding procedural fairness, judicial ``scrutiny'' categories are relevant but may not fully capture how courts scrutinize the use of demographic features in potentially discriminatory government tools like RRA. Furthermore, some policies contradict each other on whether to apply procedural fairness to certain demographic features. Thus, we propose a new framework, integrating U.S. demographics-related legal scrutiny concepts and technical fairness criteria, and contextualize it in three other major AI-adopting jurisdictions (EU, China, and India).

cs.CY↗

arXiv2023

Towards Conceptualization of "Fair Explanation": Disparate Impacts of anti-Asian Hate Speech Explanations on Content Moderators

Recent research at the intersection of AI explainability and fairness has focused on how explanations can improve human-plus-AI task performance as assessed by fairness measures. We propose to characterize what constitutes an explanation that is itself "fair" -- an explanation that does not adversely impact specific populations. We formulate a novel evaluation method of "fair explanations" using not just accuracy and label time, but also psychological impact of explanations on different user groups across many metrics (mental discomfort, stereotype activation, and perceived workload). We apply this method in the context of content moderation of potential hate speech, and its differential impact on Asian vs. non-Asian proxy moderators, across explanation approaches (saliency map and counterfactual explanation). We find that saliency maps generally perform better and show less evidence of disparate impact (group) and individual unfairness than counterfactual explanations. Content warning: This paper contains examples of hate speech and racially discriminatory language. The authors do not support such content. Please consider your risk of discomfort carefully before continuing reading!

cs.CL↗

arXiv2023

SkinGPT-4: An Interactive Dermatology Diagnostic System with Visual Large Language Model

Skin and subcutaneous diseases rank high among the leading contributors to the global burden of nonfatal diseases, impacting a considerable portion of the population. Nonetheless, the field of dermatology diagnosis faces three significant hurdles. Firstly, there is a shortage of dermatologists accessible to diagnose patients, particularly in rural regions. Secondly, accurately interpreting skin disease images poses a considerable challenge. Lastly, generating patient-friendly diagnostic reports is usually a time-consuming and labor-intensive task for dermatologists. To tackle these challenges, we present SkinGPT-4, which is the world's first interactive dermatology diagnostic system powered by an advanced visual large language model. SkinGPT-4 leverages a fine-tuned version of MiniGPT-4, trained on an extensive collection of skin disease images (comprising 52,929 publicly available and proprietary images) along with clinical concepts and doctors' notes. We designed a two-step training process to allow SkinGPT to express medical features in skin disease images with natural language and make accurate diagnoses of the types of skin diseases. With SkinGPT-4, users could upload their own skin photos for diagnosis, and the system could autonomously evaluate the images, identifies the characteristics and categories of the skin conditions, performs in-depth analysis, and provides interactive treatment recommendations. Meanwhile, SkinGPT-4's local deployment capability and commitment to user privacy also render it an appealing choice for patients in search of a dependable and precise diagnosis of their skin ailments. To demonstrate the robustness of SkinGPT-4, we conducted quantitative evaluations on 150 real-life cases, which were independently reviewed by certified dermatologists, and showed that SkinGPT-4 could provide accurate diagnoses of skin diseases.

eess.IV↗