Recommendation
-
LangChoiceBench: Measuring and Explaining Programming-Language Choice in LLMs
2026-08-06 13:52:16Lukas Twist, Twm Stone, Helen Yannakoudakis, Jie M. ZhangAbstract
Large language models (LLMs) have been shown to exhibit strong Python preferences when generating project-level code, but there is currently no systematic way to measure this behaviour across new models. To bridge this gap, we introduce LangChoiceBench, a project-level code-generation benchmark for measuring Python preference, recommendation-implementation consistency, and language diversity. LangChoiceBench covers 28 projects across seven software areas where Python is often a poor default. We evaluate 25 diverse LLMs and find that Python remains heavily over-selected, recommendation-implementation consistency is low, and smaller open-weight models generally show stronger Python preference and lower language diversity. We further analyse 9,826 reasoning traces and find that most Python choices are automatic or driven primarily by ease, rather than explicit consideration of project requirements. In a smaller but important set of cases, models fabricate contextual support for choosing Python - a failure mode we call phantom evidence - or produce code that contradicts the language selected in their own reasoning.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06041
PDF
-
Studying People to Study AI: Expert Perspectives on the Epistemic Fit and Barriers of Human Research in AI Safety & Ethics
2026-08-06 06:52:52Jessica Y. Bo, Paula Akemi Aoyagui, Shalaleh Rismani, Dipto Das, Syed Ishtiaque Ahmed, Ashton AndersonAbstract
Safety risks of AI are becoming increasingly evident in human interactions with AI technologies. The prominent approaches to evaluating these risks favor technical methods, such as model benchmarks and LLM simulations, often sidelining empirical research with human subjects. To examine this apparent gap in the acceptance of human research, we conduct an expert survey (n=93) and expert interviews (n=17) with AI Safety & Ethics (AISE) researchers from Technical, Sociotechnical, Governance, and Normative backgrounds. Our findings suggest that although there is a consensus that human research is valuable for generating evidence for AISE, its adoption and acceptance are constrained by perceived validity issues, tangible resource barriers, epistemic and personal preferences in methods, and infrastructural constraints from the broader research community. In particular, Technical researchers tend to value human research less and collaborate across disciplines less, suggesting an epistemic tension towards human methods. We propose recommendations for establishing the epistemic fit of human research within AISE and bridging the prohibitive limitations that researchers face, while avoiding performative 'human-washing'.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05656
PDF
-
Who Checks the Citations? Benchmarking Legal Hallucination Detection
2026-08-06 05:48:02Patty Liu, Dominik Stammbach, Peter HendersonAbstract
Attorneys, judges, and pro se filers increasingly use AI to draft legal documents, yet these tools frequently fabricate citations. Despite predictions that newer models would hallucinate less or that court sanctions would deter negligent filers, we found over 1,000 filings containing fabricated citations---with this number growing year-over-year. This study evaluates whether AI-based systems can mitigate these errors by automatically detecting hallucinations. We propose a taxonomy of legal citation hallucinations grounded in actual court filings and introduce a dataset of 1,300 brief excerpts containing injected errors. Benchmarking five models in agentic and non-agentic settings as well as Claude Code reveals that while the latest iterations perform better---GPT-5 achieves 84.4% recall and a 55.0% F1 score in an agentic framework---all models struggle with subtle error categories. Agentic verification remains resource-intensive, with GPT-5 averaging 15.3 steps per excerpt. Furthermore, restricted information access limits the efficacy of even the best agents. This gap creates policy concerns, as it disadvantages both AI systems and litigants who lack subscriptions to commercial legal databases. Together, our dataset, tools, and policy recommendations provide a foundation for building and auditing reliable legal citation checking tools.
Abstract (translated)
URL
https://arxiv.org/abs/2606.21155
PDF
-
CoPlan: A Trustworthy Co-Intelligence Interface for Care Planning through Role-Based Contestable Argument Graphs
2026-08-05 17:43:51Hung Truong Thanh Nguyen, H\'el\`ene Fournier, Piper Jackson, Makoto Itoh, Shannon Freeman, Rene Richard, Hung CaoAbstract
AI-supported care planning can help clinicians, patients, caregivers, and care teams coordinate complex decisions across clinical, functional, psychosocial, and environmental needs. However, many AI systems present recommendations as fixed outputs, limiting stakeholders' ability to inspect, challenge, and revise plans when they conflict with clinical judgment, patient values, or real-world feasibility. We present CoPlan - a Co-Intelligent and Contestable Interface for Human-AI Care Planning. CoPlan uses a multi-agent workflow in which specialized AI agents generate candidate interventions and supporting or challenging arguments, while human care planners can accept, reject, modify, or add arguments before final plan generation. Through this design, CoPlan combines co-intelligence, in which humans and AI agents contribute complementary expertise, with contestability, where recommendations remain open to inspection, revision, and justification. We demonstrate CoPlan in an aging-in-place care planning scenario. The system supports adaptive care team recruitment, role-based argument review, final care plan generation, and practical follow-up through scheduling agents. This work contributes a contestable care planning interface and a design framing for trustworthy human-AI care planning that preserves human agency and clinical accountability.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05107
PDF
-
An Emerging Retail Portfolio Management Application: Personalized, Tax-Aware Reinforcement Learning with Natural Language Goals
2026-08-05 16:20:11Ramin PishehvarAbstract
Retail investors lack access to the kind of personalized, tax-aware portfolio management that institutional clients take for granted -- existing robo-advisors use static, rule-based allocation, and institutional-grade systems require account minimums and technology stacks unavailable to individual investors. We present a fully built, integration-tested application that closes this gap: a FastAPI backend and web dashboard that let a user describe an investment goal in plain language (e.g. "I want steady growth but need to sell some shares next month for a down payment"), routes that goal to one of six investment mandates, and produces a live, broker-integrated portfolio recommendation from athree-phase reinforcement learning system -- a self-supervised cross-asset encoder, a Mixture-of-Experts (MoE) allocation policy with a learned intent router, and a lightweight LoRA adapter that personalizes recommendations from an individual's revealed brokerage behavior without retraining the shared model. The system is functionally complete and integration-tested end-to-end against a live brokerage API (Alpaca, paper-trading mode), including multi-user authentication, a trust first preview-before-apply confirmation flow, daily email digests, and an auditable action-integrity chain, but has not yet been opened to real end-users; we report this honestly as an emerging, pre-deployment application with a concrete path to full deployment, alongside 14-day walk-forward backtests (bootstrapped confidence intervals included) as preliminary, pre-deployment validation rather than production performance. We also report several practical engineering lessons -- silently-inactive integration paths, hanging third-party API calls, and the value of end-to-end empirical verification over trusting checkpoint metadata -- that we believe generalize to other applied RL systems built on external, live data sources.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05255
PDF
-
Quality Diversity for Reliable Data Driven Time-Use Optimization
2026-08-05 12:41:52Aneta Neumann, Ty Stanford, Dorothea Dumuid, Frank NeumannAbstract
The daily allocation of the finite 24-hour time budget is strongly associated with physical, mental, and cognitive health. While predictive models can estimate the relationship between time-use compositions and health outcomes such as body mass index, life satisfaction, and cognition, most optimization approaches focus only on maximizing expected benefit and do not consider the uncertainty inherent in data-driven prediction. Ignoring uncertainty in health-related decisions can lead to unrealistic time-use recommendations. To address this gap, we introduce an uncertainty quantification Quality Diversity (QD) framework for a more reliable time-use recommendation. Objective functions are derived using compositional data analysis using a large child cohort dataset n > 1000, to capture the relationship between daily activity compositions and multiple health indicators. We develop a new approach that incorporates predictive uncertainty into QD processes and produces more reliable recommendations that balance the expected health benefits with the confidence of the model. We explore the solution space through variable-based and objective-based behavioral representations, revealing diverse high-quality time-use composition and explicit relationships between health outcomes under uncertainty. By embedding uncertainty directly into optimization, our framework shifts the time-use recommendations toward regions of lower uncertainty while preserving high-quality structures for more reliable decision-making in behavioral health.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05230
PDF
-
A/B Agent: A Self-Evolving Agent for Strategy Iteration in Industrial A/B Testing
2026-08-05 09:28:43Zhuohang Jiang, Yuxin Chen, Yongsen Pan, Zheng Hu, Wenqi Fan, Qing Li, Hongyang Wang, Jun Wang, Wenwu OuAbstract
Industrial recommendation strategy iteration heavily relies on large-scale A/B experimentation. Traditional tuning requires experts to repeatedly design strategies, configure experiments, analyze results, and adjust parameters, making the process labor-intensive and time-consuming. Meanwhile, valuable knowledge from historical experiments is often fragmented, making systematic reuse difficult through manual expert effort alone. Existing RAG agents partially alleviate this burden by retrieving prior strategies, but typically organize experience in a flat manner, overlooking the hierarchical relationships among business scenarios, recommendation stages, optimization objectives, and experimental contexts. This often results in mismatched retrieval and limited cross-scenario transfer, while preventing agents from continuously refining strategies and parameters through sequential A/B feedback. % To address these limitations, we propose A/B Agent, a closed-loop A/B agent for industrial recommendation strategy optimization. The framework comprises three tightly coupled core components: Historical Strategy Knowledge Organization, Autonomous Target-Aware Strategy Generation, and Experiment-Guided Strategy Self-Evolution. It organizes historical strategies into a hierarchical experience tree, retrieves transferable evidence through multi-path Tree-RAG to generate executable strategies, and continuously analyzes online A/B feedback to guide autonomous tuning and update the experience tree for self-evolution. Extensive offline and online evaluations demonstrate its effectiveness, including a 4.829% improvement in GMV in a real-world short-video e-commerce recommendation system while maintaining positive gains across all guardrail metrics.
Abstract (translated)
URL
https://arxiv.org/abs/2608.04625
PDF
-
Compass: Continuously Aligning Social Media Feeds via In-Situ Reflections
2026-08-04 23:04:48Aadit Barua, Leijie Wang, Amy X. ZhangAbstract
Social media recommendation feeds often optimize for users' immediate impulses rather than preferences they would hold after deeper reflection. Some systems address this misalignment by incorporating users' explicit preferences via a configuration page or in-feed controls instead of just behavioral signals. However, users typically have evolving preferences, and their stated preferences and behavior naturally diverge, necessitating continuous reflection and feed realignment. But existing strategies require the user to take initiative and are often effortful; as a result, in practice they are rarely invoked. We present Compass, a system that aligns a user's feed with their reflective preferences by helping users reflect on and articulate their preferences given their behavior. To enable continuous reflection during everyday browsing, Compass surfaces in-situ reflections via lightweight notifications, while feed alignment is achieved by periodically simulating behavioral signals and directly manipulating feed content. We embedded Compass within YouTube Shorts and compared it against a baseline without continuous support through a 10-day field study (N=15). We found that Compass promoted more reflective and purposeful feed consumption, iterative preference adjustment, and stronger feed alignment, without sacrificing the casual nature of feed browsing.
Abstract (translated)
URL
https://arxiv.org/abs/2608.04274
PDF
-
SiMDex: Mining Similar Egocentric Videos for Cross-Embodiment Dexterous Manipulation
2026-08-04 19:55:12Nie Lin, Takehiko Ohkawa, Sijin Chen, Ruoshi Wen, Zhuohang Li, Liqun Huang, Zhengming Zhu, Yiming Bao, Yunfei Li, Minjie Cai, Xiao Ma, Wei Xu, Yoichi SatoAbstract
Recent years have witnessed an explosive trend of scaling ego-centric human videos for robot manipulation, yet it remains unclear which data actually benefits dexterous manipulation. We present SiMDex, a similarity-based data mining framework that casts human data selection for VLA post-training in dexterous manipulation as a recommendation problem. For each robot demonstration, SiMDex employs a three-layer recall-ranking-re-ranking pipeline to extract task-relevant subsets from a pool of ~32M egocentric human samples, operating in a morphology-agnostic action space that requires no changes to VLA architecture or training. Against a strong baseline trained with an equal amount of randomly sampled human data, SiMDex uses only ~1.49M mined samples (<5% of the pool) yet improves the overall success rate from 47.7% to 61.1%, showing that selective curation outperforms indiscriminate data mixing.
Abstract (translated)
URL
https://arxiv.org/abs/2608.04196
PDF
-
When Attention Goes Blind: Numerical Failure in ALiBi Positional Encodings
2026-08-04 17:54:01Christopher Schr\"oder, Lukas Gienapp, Ferdinand Schlatt, Martin Potthast, Gerhard HeyerAbstract
We identify a previously overlooked failure mode of ALiBi positional encoding: its linear bias scaling underflows floating-point precision, which zeroes out a large fraction of attention weights and renders the affected attention heads partially blind. We analyze this failure mode, characterize its impact, and examine four mitigation strategies. We further demonstrate its occurrence in state-of-the-art pretrained models based on ALiBi. Comprehensive pretraining experiments with 148M-parameter decoder models help us to disentangle its effects from out-of-context degradation. We find that ALiBi's failure mode can substantially impair token retrieval while having only a minor effect on standard decoder benchmarks. We propose four training-time mitigation strategies and evaluate them individually and in combinations, finding that log-scaled distances yield the most consistent improvements in passkey retrieval. Despite this problem, default ALiBi slopes remain a surprisingly strong baseline, particularly for needle-in-a-haystack retrieval. Based on these findings we provide concrete recommendations on how to train models with ALiBi.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03994
PDF
-
Intertemporal Preference Steering in Qwen3 via Contrastive Activation Addition
Abstract
We study linear representations of temporal horizon in the large language model Qwen3-32B and use them to change the model's time-related preferences, recommendations, and capabilities. We train contrastive linear probes on teacher-forced temporal-choice answers to find a short-term versus long-term direction in the model's residual stream, and evaluate contrastive activation-addition steering on a held-out binary temporal-choice task, an out-of-distribution monetary intertemporal-choice task, and a TravelPlanner capability benchmark. The central result is that temporal-horizon directions can be identified with simple contrastive linear probes and then used for steering to induce large, bidirectional preference changes. On an out-of-distribution monetary choice task that varies reward size and delay, steering strongly shifts the model's indifference threshold between smaller-sooner and larger-later rewards in both directions. We further show improvements on a planning-related capability metric under moderate temporal steering. These results suggest that model intertemporal preferences are measurable and steerable, which is relevant for AI systems that give advice involving delayed costs and benefits, and for safety questions about long-horizon planning.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03892
PDF
-
ADMITBench: A Safety-Governed Reference Framework for Evaluating the Admissibility of Industrial LLM Advisories
2026-08-04 16:07:56Yash Misra, Javal Vyas, Siddharth Gutta, Mehmet Mercang\"ozAbstract
This white paper presents ADMITBench, a reference framework for evaluating industrial LLM advisories at the level of the proposed action. The framework implements a versioned, safety-governed evaluation contract that checks whether a recommendation is supported by the available evidence, permitted under the stated authority and procedure, and acceptable under the plant-specific consequence checks encoded in the selected evaluation profile. In this report, \emph{safety-governed} means that eligibility is determined through explicit, non-compensatory checks derived from a versioned plant profile; it does not mean that the evaluator, model, or plant has been safety-certified. Release 0.1.0 is a public reference implementation for technical and research evaluation, not an authorisation for physical execution.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03866
PDF
-
Designing Social Robots for Inclusive Child Wellbeing Assessment: Insights from Communities Supporting Developmental Language Disorder and Forced Migration
2026-08-04 15:28:41Fethiye Irmak Dogan, Yue Lou, Alva Markelius, Emma Geijer-Simpson, Gustaf Gredeb\"ack, Tamsin Jane Ford, Ginevra Castellano, Hatice Gunes, Georgina Warner, Jenny L. GibsonAbstract
Assessing children's wellbeing and mental health can be particularly challenging for children experiencing communication barriers, such as children with Developmental Language Disorder (DLD) and children with forced migration backgrounds. During the assessment process, traditional self-report questionnaires place substantial demands on language comprehension and verbal expression. In this context, social robots have emerged as a promising tool for supporting wellbeing assessment without solely relying on self-report questionnaires, yet limited research has examined how such interactions can be designed to be inclusive, appropriate, and ethically acceptable for children with diverse communication needs. To address this gap, we created candidate child--robot interaction activities as design probes and conducted focus groups with parents and professionals supporting children with DLD and children with forced migration backgrounds. Through thematic analysis, we identified considerations relating to robot role and capabilities, interactional dynamics, individual differences, and child agency, alongside population-specific considerations shaped by children's communication needs and lived experiences. Based on these findings, we derive a set of ethical and inclusive design recommendations for robot-mediated wellbeing assessment. By foregrounding these considerations and recommendations, this work contributes design guidance for inclusive robot-mediated wellbeing assessments for children with diverse communication needs.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03820
PDF
-
How Closely Do LLM Reviews Align with Human Peer Review?
2026-08-04 13:39:36Abraham Camelo-Guerrero, Jairo Diaz-RodriguezAbstract
Large language models (LLMs) are increasingly used to generate scientific reviews, yet existing evaluations rarely examine whether different providers align with both conference decisions and human reviewing priorities within the same controlled setting. We compare reviews from OpenAI GPT-5.4, Google Gemini 3.1 Pro Preview, and Anthropic Claude Opus 4.6 with human reviews and final decisions for 300 topic-matched ICLR 2026 submissions, equally divided among oral, poster, and rejected papers. Each model reviewed every paper using identical instructions and rating scales after decision information was removed. Our study contributes a cross-provider analysis of three complementary dimensions: alignment with broad and fine-grained decision categories, differences in recommendation-scale usage, and thematic agreement in identified weaknesses. All three LLMs distinguished accepted from rejected papers, but none reproduced the oral versus poster distinction present in human ratings. Scoring patterns were provider-specific: Gemini assigned systematically higher ratings, while OpenAI and Claude were closer to humans for rejected and poster papers but more critical of oral papers. Human and LLM reviews also differed in emphasis, with LLMs more frequently identifying missing baseline comparisons and humans more often raising computational-efficiency concerns. These results show that broad decision alignment does not imply agreement with finer human judgments or reviewing priorities.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03659
PDF
-
AI-Assisted Peer Review Across Research Communities: From Reviewer AI Policies to LLM Review Quality
2026-08-04 12:34:47Alexander M. Fichtl, Lukas Ellinger, Josefin Kelber, Kry\v{s}tof Ol\'ik, Georg GrohAbstract
AI-assisted peer review is increasingly discussed and adopted as a tool to support the scientific publishing process, yet there is little systematic understanding of how publication venues regulate its use or of how capable current AI review systems are. We address these questions by first surveying reviewer-facing AI policies across 111 leading AI/NLP conferences and medical journals, revealing substantial regulation differences between the two communities. Second, we evaluate AI-generated peer reviews at ICLR 2026 and Nature Communications using a novel dataset comprising original manuscript submissions and several hundred human- and machine-generated reviews. We compare reviews produced by open-source and proprietary models using complementary evaluation metrics, including LLM-as-a-Judge, score alignment, granularity, and overlap with human reviewers' concerns. Our results show that current LLMs can generate detailed and fluent reviews but exhibit systematic weaknesses, such as overly positive recommendations, generic criticism, and uneven evidence grounding. We demonstrate that aggregate quality scores alone can overestimate review quality and argue for multi-dimensional evaluation of AI-generated peer reviews.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03581
PDF
-
MinerU.Chem: A High-Precision System for Optical Chemical Structure and Reaction Recognition
2026-08-04 12:09:02Haote Yang, Jiang Wu, Jingchao Wang, Xingjian Wei, Lixin Ma, Linye Li, Chen Zhu, Xiaolong Wu, Yuheng Lu, Ziran Zhu, Junyuan Gao, Lingli Ge, Yuan Xu, Huijie Ao, QianQian Wu, Dechen Lin, Huaiyu Gu, Lu Chen, Shengxin Lu, ShaSha Wang, Yuanyuan Cao, Zhejia Yu, Ruijie Zhang, Zimai Tian, Jiaxing Sun, Yinfan Wang, Jiahe Song, Chuang Wang, Yubin Wang, Rui Nie, Hao Zheng, Bowen Jiang, Hongbin Lai, Yifan He, Chengjin Liu, Tingting Zhang, Liqun Wei, Lijun Wu, Bin Wang, Yuqiang Li, Guangyu Wang, Wei Li, Bowen Zhou, Dahua Lin, Conghui HeAbstract
In organic chemistry papers and patents, molecular structures, reaction schemes, and experimental conditions are often presented as molecular structure depictions, reaction diagrams, and complex tables or figures. Such information is difficult for general-purpose document parsing systems to directly convert into machine-readable data. This limits data production for organic chemistry knowledge base construction and for AI for Chemistry tasks such as reaction prediction, retrosynthesis, condition recommendation, molecular property prediction, and drug molecule design. This report introduces this http URL, a document parsing system for organic chemistry literature integrated into the MinerU online platform. Built on top of MinerU's general document parsing pipeline, this http URL adds five chemistry-specific modules: chemistry relevance filtering, molecular structure detection, molecule identifier extraction, molecular structure recognition, and reaction scheme parsing. Together, these modules convert organic-chemistry-related image regions in documents into a Molecule Summary List and a Reaction Summary List. For molecular structure recognition, this http URL uses CARBON (Complex Atomic Representation and Bonding Object Notation) as its core representation. CARBON enables recognition results to preserve both the visual layout of the original image and complex chemical semantics, while supporting the export of standard downstream formats such as MolFile and SMILES. On the SMILES-evaluable subset of MolRecBench-Wild (N=2,392), this http URL's molecular structure recognition module achieves a SMILES exact-match accuracy of 93.02%, outperforming the best evaluated comparison system, GPT-5.6-Sol (74.87%), by 18.15 percentage points. The system has been integrated into the MinerU online platform and is available at this https URL .
Abstract (translated)
URL
https://arxiv.org/abs/2608.03525
PDF
-
How Many Labels Are Enough? ALDA: Active Learning Deployment Advisor for Medical Image Classification
2026-08-04 11:53:02Julia Machnio, Mads Nielsen, Mostafa Mehdipour GhaziAbstract
Active learning (AL) promises to reduce the cost of medical imaging projects by lowering the number of clinical labels required. However, practical deployment requires committing to a sampling strategy before the full annotation budget is spent, and choosing the wrong strategy can increase rather than decrease costs. We propose Active-Learning Deployment Advisor (ALDA), a deployment-oriented framework for AL method selection under clinical performance constraints. Given a short pilot phase, ALDA fits a parametric learning-curve model to each candidate strategy, estimates whether that strategy is expected to reach a required clinical performance target, and predicts the number of expert annotations needed to do so. In addition to absolute annotation cost, ALDA introduces a deployment window that quantifies the sensitivity of this cost estimate to uncertainty in the clinical threshold. The final recommendation follows a risk-aware rule: among strategies with near-optimal predicted cost, ALDA prefers the strategy with the narrowest deployment window, the most robust to threshold revisions. Experiments on four medical imaging classification domains show that ALDA predicts the deployment-optimal method from a pilot of 15-30% of the intended budget and reduces annotation costs by up to 82% compared with a poor strategy choice. Rather than introducing a new sampling heuristic, ALDA provides a practical decision layer that answers a deployment-critical question: how many labels are enough?
Abstract (translated)
URL
https://arxiv.org/abs/2608.03511
PDF
-
TumorBoard: Evidence-Grounded Multi-Agent Decision Support for Longitudinal Neuro-Oncology
2026-08-04 06:32:25Yantong Liu, Zheyu Zhang, Runpeng Liu, Mu Xitang, Seong-Yoon Shin, Hyun-Ae LeeAbstract
Neuro-oncology decisions require coordinated interpretation of serial MRI, pathology, molecular markers, treatment history, performance status, and evolving guidelines. We present TumorBoard, a multi-agent decision-support system built around a shared longitudinal case state and an auditable claim-evidence ledger. Specialist agents for radiology, neuropathology, molecular diagnosis, guidelines, and therapy planning produce atomic claims with provenance. An adversarial critic exposes contradictions, and a safety governor releases, qualifies, or defers recommendations according to evidence sufficiency and temporal validity. On a 360-case hidden benchmark at a matched token budget, TumorBoard achieved an action F1 of 0.772 and evidence entailment of 0.914. It exceeded the strongest typed-council baseline by 3.1 percentage points (95% CI: 1.6 to 4.7, adjusted p = 0.0012), while recommendation-to-evidence coverage reached 0.927. Under evidence deletion, the system deferred 84.2% of unsafe cases and limited harmful recommendations to 5.8%. The safety governor reduced harmful release by 7.8 percentage points at a false-deferral cost of 4.3 percentage points. Ablation studies of the ledger, critic, and governor produced the predicted failure patterns, establishing structured coordination as the source of the measured multi-agent advantage.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03190
PDF
-
Optimal Liability Design for Medical AI
2026-08-04 04:33:10Rui Mao, Tingliang Huang, Houcai ShenAbstract
Artificial intelligence (AI) is increasingly integrated into medical decision-making, yet its liability implications remain complex, particularly when physicians differ in diagnostic skills and their quality is unobservable. This paper develops a principal-agent model in which a social planner designs medical liability to regulate a physician with private quality information who chooses between a standard treatment, a personalized judgment-based treatment, or following an imperfect AI recommendation. Our analysis yields several novel insights. First, we show that the optimal mechanism under asymmetric information is surprisingly simple: a uniform, one-size-fits-all liability level for all physician types who deviate from the standard of care. Despite physician heterogeneity, this simple policy often achieves the full-information first-best outcome, particularly when standard care is reliable or AI is highly accurate. Second, the relationship between AI accuracy and optimal liability is non-monotonic. Contrary to common intuition, better AI does not always imply more relaxed liability. As AI accuracy increases, the optimal liability either decreases monotonically or follows an inverted-U pattern, depending on the uncertainty of the standard treatment. Third, asymmetric information does not universally reduce social welfare. Welfare loss arises only when standard care is unreliable and AI accuracy is too low; even then, its magnitude follows an inverted U-shape, initially increasing as AI complicates the regulatory problem, but declining as more accurate AI helps mitigate it. Finally, we find that information asymmetry is a double-edged sword in the presence of AI, and greater transparency does not benefit all stakeholders equally.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03114
PDF
-
SeqLLM: Augmenting LLMs with Behavioral-Sequence Modeling for High-Stakes Decisions at WeChat Pay
2026-08-04 03:23:12Guilin Li, Jiaxing Zhang, Matthias Hwai Yong Tan, Bo Wang, Weiran HuangAbstract
Merchant risk control at large payment platforms screens tens of millions of merchants daily, where false positives harm legitimate merchants and false negatives leave harmful activity undetected. The hardest cases require jointly understanding a merchant's textual profile and long behavioral sequence. Large language models (LLMs) excel at text but cannot natively model such sequences, while adapting them often causes catastrophic forgetting. We present SeqLLM, a framework that adds behavioral-sequence modeling to a pretrained LLM while preserving its language ability. SeqLLM combines three components: a compact discrete vocabulary that represents behavioral events as native tokens; a lightweight projector, trained with a two-stage alignment curriculum, that grounds these tokens in the LLM's semantic space; and prefix-guided capability injection, which acquires sequence-modeling ability through task-prefixed supervised fine-tuning rather than continual pre-training. SeqLLM is deployed at WeChat Pay, screening millions of merchants daily. Against the production DeepSeek-based LLM baseline, it raises screening precision from 92.0% to 97.5%. Its pretrained behavior-token embeddings also improve Precision@Top-0.01% by 26.8 percentage points in a production fraud detector serving billion-scale transaction traffic. Beyond payments, SeqLLM achieves state-of-the-art results on public recommendation benchmarks. On MovieLens and Amazon, it surpasses the strong User-LLM baseline by up to 32% relative Recall@5 while retaining markedly stronger language ability. On RecIF, it improves Pass@32 by 14.2% over the full OneRec-8B pipeline using only one-fifth of its GPU-days.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03063
PDF
