The Commonplace
Home Papers Evidence Explore Trends Syntheses Digests About 🎲 Workforce Futures

Papers (1699)

Filter: Sort:
Orchestrated human–AI workflows sharply cut effort and defects: across three real modernization projects, the Chiron platform reduced modeled person-days from 1080 to 232.5 and senior-equivalent effort from 1080 to 139.5 days, halved validation issues and raised first-release coverage from 77% to 90.5%, with largest gains when AI was embedded in the end-to-end delivery flow rather than used as an isolated coding assistant.
medium quasi_experimental Full text arxiv
Orchestrating Human-AI Software Delivery: A Retrospective Longitudinal Field Study of Three Software Modernization Programs · Maximiliano Armesto, Christophe Kolb · Mar 20, 2026
Generative AI lifts task productivity by 15–50%—especially for junior workers—yet so far has not caused mass unemployment; instead it reshapes entry-level roles and demands organizational change before broad macro gains emerge.
medium review_meta Summary semantic_scholar
AI, Productivity, and Labor Markets: A Review of the Empirical Evidence · Eric Fruits, Kristian Stout · Fetched Mar 20, 2026
China's growth model is shifting from investment to innovation: total factor productivity's share rose from 18% to 26% in 2016–22, with the digital economy explaining about 40% of that increase; coastal provinces are innovation-driven while inland regions remain capital-intensive.
medium correlational Full text semantic_scholar
Analysis of China's Economic Growth Drivers: An Empirical Study Based on an Extended Cobb-Douglas Production Function (2010-2022) · Zihan Zhao · Fetched Mar 20, 2026
AIGQ’s generative query suggestions boost Taobao engagement: an interest-aware training and policy-optimization pipeline raises click-through and other business metrics in large-scale randomized homepage tests.
high rct Full text arxiv
AIGQ: An End-to-End Hybrid Generative Architecture for E-commerce Query Recommendation · Jingcao Xu, Jianyun Zou, Renkai Yang, Zili Geng, Qiang Liu, Haihong Tang · Mar 20, 2026
Inventory-theory regularization steadies deep-RL: grounding DRL policies in base-stock concepts speeds hyperparameter tuning and raises inventory performance, enabling a 100% deployment on Alibaba's Tmall; synthetic tests show regularization alters which DRL methods perform best.
medium descriptive Full text arxiv
DeepStock: Reinforcement Learning with Policy Regularizations for Inventory Management · Yaqi Xie, Xinru Hao, Jiaxi Liu, Will Ma, Linwei Xin, Lei Cao, Yidong Zhang · Mar 20, 2026
Human-crafted 'skills' let AI agents reliably program real IoT devices, whereas out-of-the-box LLM skills often fail due to timing and hardware-specific constraints; hardware-in-the-loop validation across platforms shows human expertise remains crucial for embedded software success.
medium quasi_experimental Full text arxiv
Skilled AI Agents for Embedded and IoT Systems Development · Yiming Li, Yuhan Cheng, Mingchen Ma, Yihang Zou, Ningyuan Yang, Wei Cheng, Hai "... · Mar 20, 2026
China’s digital economy boosts provincial carbon productivity both locally and across borders, while industrial intelligence depresses local carbon productivity but produces strong positive spillovers that make its net effect beneficial; impacts vary sharply by region.
low correlational Full text openalex
Study on the impact of industrial intelligence and the digital economy on China’s regional total factor carbon productivity under carbon neutrality · Dandan Xiao, Jinwang Liu · Mar 20, 2026
A 1,400-question benchmark finds modern LLMs can leverage retrieval to interpret company filings but largely fail at market trading-signal and time-series reasoning; this gap limits their reliability for real-world financial decision-making.
low descriptive Full text arxiv
FinTradeBench: A Financial Reasoning Benchmark for LLMs · Yogesh Agrawal, Aniruddha Dutta, Md Mahadi Hasan, Santu Karmaker, Aritra Dutta · Mar 19, 2026
A new benchmark reframes GPU kernel evaluation around hardware 'Speed-of-Light' limits: SOL-ExecBench measures 235 CUDA kernels from 124 AI models on NVIDIA Blackwell GPUs and scores optimizers by how much of the gap to analytically derived hardware bounds they close, with measurement safeguards to deter reward-hacking.
n/a descriptive Full text arxiv
SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits · Edward Lin, Sahil Modi, Siva Kumar Sastry Hari, Qijing Huang, Zhifan Ye, Nestor ... · Mar 19, 2026
Learner prerequisites — not information abundance — can bottleneck AI-driven knowledge adoption: unless teaching reaches the prerequisite depth of a target concept, extra instruction is futile, and one-size-fits-all broadcasts can be up to proportionally slower than tailored teaching across diverse learner types.
n/a theoretical Full text arxiv
A Mathematical Theory of Understanding · Bahar Taşkesen · Mar 19, 2026
AI agents fall short on domain-specific data science: in a 17-task benchmark across six industries, AI-only systems performed near or below the median human competitors, while human–AI teams produced the strongest solutions, underscoring the continued importance of human expertise.
medium descriptive Full text arxiv
AgentDS Technical Report: Benchmarking the Future of Human-AI Collaboration in Domain-Specific Data Science · An Luo, Jin Du, Xun Xian, Robert Specht, Fangqiao Tian, Ganghua Wang, Xuan Bi, C... · Mar 19, 2026
Simulated deployments of predictive-policing systems reveal extreme, city-specific racial disparities in police detection—Baltimore shows runaway amplification while Chicago exhibits systematic under‑detection of Black residents; algorithmic debiasing via CTGAN shifts but does not erase these structural imbalances.
medium descriptive Full text arxiv
Unmasking Algorithmic Bias in Predictive Policing: A GAN-Based Simulation Framework with Multi-City Temporal Analysis · Pronob Kumar Barman, Pronoy Kumar Barman · Mar 19, 2026
Natural-language rendering of a structured prompt protocol (PPS) meaningfully improves LLM alignment and usability—cutting follow-up prompts by about two-thirds—especially for ambiguous business tasks; however, benefits vary by task type and conventional evaluation metrics can mask these practical gains.
medium rct Full text arxiv
Evaluating 5W3H Structured Prompting for Intent Alignment in Human-AI Interaction · Peng Gang · Mar 19, 2026
Institutional design, not model brand, largely determines whether LLM agents misbehave: simulated government agents break rules much more under weak authority structures, and modest safeguards sometimes help but do not reliably prevent serious abuse.
medium quasi_experimental Full text arxiv
I Can't Believe It's Corrupt: Evaluating Corruption in Multi-Agent Governance Systems · Vedanta S P, Ponnurangam Kumaraguru · Mar 19, 2026
A new multi-hop spatial benchmark shows modern vision-language models still falter at compositional spatial reasoning and precise grounding; targeted reinforcement-learning post-training on a dedicated corpus measurably improves both benchmark scores and transfer to embodied manipulation tasks.
medium descriptive Full text arxiv
MultihopSpatial: Multi-hop Compositional Spatial Reasoning Benchmark for Vision-Language Model · Youngwan Lee, Soojin Jang, Yoorhim Cho, Seunghwan Lee, Yong-Ju Lee, Sung Ju Hwan... · Mar 19, 2026
Video filters and avatars erode viewers' trust and confidence but do not make them worse at spotting lies; AI-mediated appearance changes reduce perceived credibility without impairing actual detection of deception.
high rct Full text arxiv
Through the Looking-Glass: AI-Mediated Video Communication Reduces Interpersonal Trust and Confidence in Judgments · Nelson Navajas Fernández, Jeffrey T. Hancock, Maurice Jakesch · Mar 19, 2026
A new Agent Control Protocol codifies cryptographic admission control for enterprise autonomous agents, pairing 36 specification documents with a 22-package Go reference implementation and 51 signed conformance vectors; it aims to let institutions authorize, audit and revoke agent actions without replacing existing RBAC or Zero Trust systems.
n/a descriptive Full text arxiv
Agent Control Protocol: Admission Control for Agent Actions · Marcelo Fernandez · Mar 19, 2026
Prompt framing creates a confirmation-bias blind spot in LLM code reviewers: when changes are framed as bug-free or as security fixes, detection rates fall sharply and attackers can reintroduce known vulnerabilities in up to 88% of trials against an autonomous agent; removing metadata or adding explicit instructions largely undoes the effect.
medium quasi_experimental Full text arxiv
Measuring and Exploiting Confirmation Bias in LLM-Assisted Security Code Review · Dimitris Mitropoulos, Nikolaos Alexopoulos, Georgios Alexopoulos, Diomidis Spine... · Mar 19, 2026
A formal framework separates AI that amplifies human judgment from AI that progressively usurps it, proposing four operational metrics (CAI*, D, HRI, HCDR) to diagnose whether hybrid performance is sustainable; it warns that maximizing short-term capability can undermine long-term human expertise and calls for a cognitive sustainability constraint in system design.
n/a theoretical Full text arxiv
Cognitive Amplification vs Cognitive Delegation in Human-AI Systems: A Metric Framework · Eduardo Di Santi · Mar 19, 2026
Regulatory sandboxes offer a pragmatic route to balance AI innovation and safety by enabling iterative, evidence-based rulemaking; however, their promise hinges on clear boundaries, proportionality, and robust safeguards against capture and rent-seeking.
n/a theoretical Summary semantic_scholar
Experimentalism beyond ex ante regulation: A law and economics perspective on AI regulatory sandboxes · Antonella Zarra · Fetched Mar 19, 2026
Off-the-shelf reasoning AI agents converge to Nash-like behavior in repeated strategic games without post-training alignment, the authors prove and validate in simulations; this suggests many strategic interactions may attain stable equilibria without universal alignment procedures.
medium theoretical Full text arxiv
Reasonably reasoning AI agents can avoid game-theoretic failures in zero-shot, provably · Enoch Hyunwook Kang · Mar 19, 2026
Labeling products or strategy with 'AI' without substance appears to hurt green innovation: Chinese listed firms that AI-wash their annual reports show lower green patenting, a decline transmitted through reputation losses in product markets and increased financing constraints. The effect is strongest for private firms, SMEs and firms in highly competitive sectors, suggesting targeted disclosure and regulatory measures could restore market incentives.
low correlational Full text arxiv
The Spillover Effects of Peer AI Rinsing on Corporate Green Innovation · Li Wenxiu, Wen Zhanjie, Xia Jiechang, Guo Jingqiao · Mar 19, 2026
Zimbabwean banks find AI can broaden access—especially anomaly detection, which explains most measured differences—but rollout is constrained by digital illiteracy, poor internet and high costs, and customers remain wary of opaque credit-scoring and chatbot decisions.
low descriptive Full text openalex
The Impact of Artificial Intelligence on Financial Inclusion in Zimbabwe's Banking Sector: Challenges and Opportunities for Expanding Access to Financial Service · Justin Tinashe Muswaburi, Precious Mandava, Nyasha Mpondwe · Mar 19, 2026
Signalling AI adoption makes employers more attractive to jobseekers; the boost comes from perceptions of greater innovation and is strongest for applicants confident in AI skills.
medium rct Summary openalex
Signaling Organizational Artificial Intelligence Adoption in Recruitment Materials: Role of Perceived Innovation Ability in Organizational Attractiveness · Jialin Cheng, Shunhong Ji · Mar 19, 2026
Financial digital intelligence in China is linked to stronger innovation in strategic emerging industries, with gains driven by improved disclosure, lower transaction costs and deeper industry–university–research ties; effects vary substantially across regions and industry types.
medium correlational Full text openalex
Financial Digital Intelligence and Innovative Development of Strategic Emerging Industries: Impact Effects and Mechanism Tests · Wei XUE, Ziyan LI, Qianyi Lei · Mar 19, 2026