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

Papers (1699)

Filter: Sort:
Human behavior variability reflects a dynamic internal 'state' that determines decision outcomes, and interventions targeted at that state can, in principle, control those outcomes; a new framework is supported by multiple literatures and two years of platform data from 200,000 users.
medium theoretical Full text arxiv
You Are in Control of Your State: Why Human Outcomes Are Controllable Through Causal State Intervention · Suraj Biswas, Saurav Gupta, Pritam Mukherjee · May 26, 2026
Per-query configuration selection slashes retrieval-agent serving costs without losing accuracy: BRANE picks pipeline settings at query time using LLM-derived features and lightweight predictors, matching top static configurations while reducing cost by up to 89% and outperforming routing and rule-based baselines.
medium other Full text arxiv
Natural Language Query to Configuration for Retrieval Agents · Melissa Z. Pan, Negar Arabzadeh, Mathew Jacob, Fiodar Kazhamiaka, Esha Choukse, ... · May 26, 2026
GENESIS proposes an agentic AI platform that could dramatically shorten cellular R&D cycles by converting specifications and anomalies into over-the-air-validated solutions and storing them in a compounding knowledge base; however, the framework's claims rest on architectural design with limited empirical evidence and unresolved risks from LLM hallucinations and sim-to-hardware transfer.
low descriptive Full text arxiv
GENESIS: Harnessing AI Agents for Autonomous 6G RAN Synthesis, Research, and Testing · Tamerlan Aghayev, Maxime Elkael, Michele Polese, Minh Dat Nguyen, Gabriele Gemmi... · May 26, 2026
Agentic AI imposes a persistent operating cost — a 'stochastic tax' — even when design debt is low, while accumulated 'agentic technical debt' amplifies that tax; the paper supplies a formal model and practical metrics managers can use to quantify and estimate these costs from operational data.
n/a theoretical Full text arxiv
Modeling Agentic Technical Debt and Stochastic Tax: A Standalone Framework for Measurement, Simulation, and Dashboarding · Muhammad Zia Hydari, Raja Iqbal, Narayan Ramasubbu · May 26, 2026
A self-improving loop that lets a language-model agent rewrite its scaffold and fine-tune weights substantially outperforms prior siloed approaches across three technical domains, cutting GPU kernel runtimes by ~92% and delivering major gains in legal classification and RNA denoising; combining harness changes (to make the agent more agentic) with weight updates (to build domain intuition) yields much larger improvements than changing either alone.
medium other Full text arxiv
SIA: Self Improving AI with Harness & Weight Updates · Prannay Hebbar, Yogendra Manawat, Samuel Verboomen, Alesia Ivanova, Selvam Palan... · May 26, 2026
AI drafts can make workers look faster while slowing down the system: under congestion reviewers cut scrutiny and AI errors cause costly rework, so AI only reduces backlogs when it handles enough tasks and review plus rework takes far less time than doing the task manually.
n/a theoretical Full text arxiv
Queue & AI: When Faster Tasks Slow Down the Workflow · Silvia Bartolucci, Pierpaolo Vivo · May 26, 2026
An AI-augmented hub-and-spoke data platform can reconcile domain autonomy with enterprise governance by automating quality controls, contracts, and reviews while empowering domain teams; measured by adoption and time-to-insight metrics, the model promises productivity gains without centralized bottlenecks.
low descriptive Full text arxiv
Beyond the Data Mesh Illusion: Designing Modern AI-augmented Lakehouses to Bridge the Gap Between Theory and Practice · Oliver Angélil, Jan Migon · May 26, 2026
ReasonOps reframes AI reasoning as an operational lifecycle—combining autoformalization, theorem proving and runtime assurance—to improve reliability of LLM-driven reasoning in safety-critical systems; the proposal is a conceptual blueprint without empirical validation.
n/a theoretical Full text arxiv
ReasonOps: A Unified Operational Paradigm for Trustworthy Verified LLM Reasoning · Adnan Rashid · May 26, 2026
A four-month case study of a persistent research agent shows workflows are cache-dominated—82.9% of recorded tokens were cache reads—implying the economic unit may shift from cost-per-token to cost-per-completed-artifact.
low descriptive Full text arxiv
Persistent AI Agents in Academic Research: A Single-Investigator Implementation Case Study · Anas H. Alzahrani · May 26, 2026
Teaching language models when and how to show emotion improves their bargaining payoffs in simulation. The EmoDistill pipeline — an IQL-based emotion selector plus LoRA-tuned expression policies — outperforms vanilla models and ablated variants across four negotiation domains, though results stem from agent-only simulations and await human validation.
medium other Full text arxiv
EmoDistill: Offline Emotion Skill Distillation for Language Model Agents in Adversarial Negotiation · Yunbo Long, Haolang Zhao, Lukas Beckenbauer, Liming Xu, Alexandra Brintrup · May 26, 2026
Human–AI collaboration is reshaping India's IT workforce: it boosts efficiency and innovation but risks widening skill gaps unless firms invest in continuous reskilling and governance.
low review_meta Full text semantic_scholar
Human–AI Collaboration in the Indian IT Industry: A Qualitative Exploration of the Future of Work · Vakhil Rajith Nair · Fetched May 26, 2026
AI raises manufacturing efficiency but deepens labor-market polarization by disproportionately displacing routine and mid-skilled jobs. New high-skill roles appear, yet limited accessibility and insufficient reskilling risk structural unemployment without human-centered policies.
medium review_meta Summary semantic_scholar
Artificial Intelligence in Manufacturing · Aysel Arslan, Roland Y. H. Silitonga · Fetched May 26, 2026
Iran's intellectual property laws leave AI-generated art and inventions in legal limbo; revising statutes or creating tailored rights is needed to protect innovation without eroding human creators' interests.
n/a review_meta Full text arxiv
Examining the Challenges of Intellectual Property in AI-Generated Productions · Ali Mazhar, Mohammad Zare, Marjan Veysi · May 26, 2026
A per-action insurance layer can make autonomous agents economically accountable: the authors derive a time-consistent counterfactual risk toll and prove no-splitting within underwriting boundaries, identify a strictly positive irreversible-authority premium, and show how high-probability toll envelopes yield executed-action budget guarantees—providing the mathematical base for operational runtime underwriting.
n/a theoretical Full text arxiv
Foundations of a Time-Consistent Counterfactual Actuarial Runtime for Autonomous AI Agents · Hao-Hsuan Chen · May 26, 2026
Multinationals reshuffle AI systems and signals in response to fragmented rules: firms with broader regulatory exposure and higher AI maturity show markedly stronger bifurcation, modular architectures, ethical signaling and compartmentalization across EU, US and China. The pattern is concentrated in tri-jurisdictional and tech firms, with governance exposure statistically predicting adaptation intensity.
medium correlational Full text openalex
Research on the adaptation path of corporate strategy based on the differences in AI governance visions between the Atlantic and China · Min Xu · May 26, 2026
AI can noticeably strengthen SMEs' financial management—improving cash‑flow forecasting, credit risk assessment and fraud detection—but real gains require data, staff skills and proper governance; cloud solutions, targeted training and explainable AI are key enablers.
n/a review_meta Full text openalex
The Role of Artificial Intelligence in Strengthening Financial Practices of SMEs · Aneta Cugova, Sumana Chaudhuri · May 26, 2026
AI is creating a new form of 'co-intelligence' in which human and machine minds jointly shape thinking and production, promising major gains but demanding fresh ethical standards and governance to manage displacement, inequality and institutional risk.
n/a theoretical Full text openalex
Co-Intelligence: Human-AI Coexistence in the Age of Thinking Machines · Divyansh Mishra, Rajesh Kumar Mishra, Rekha Agarwal · May 26, 2026
JobBench tests AI agents on 130 real-work tasks across 35 occupations using extensive fact-anchored rubrics and finds leading models succeed on less than half the required criteria, implying current agents are far from reliably performing the professional tasks humans actually want delegated.
n/a descriptive Full text arxiv
JobBench: Aligning Agent Work With Human Will · Yuetai Li, Yichen Feng, Zhangchen Xu, Zixian Ma, Kaiyuan Zheng, Fengqing Jiang, ... · May 25, 2026
AgentSociety shows autonomous LLM agents can be organized with liquid-democracy-style delegation and selective information disclosure so self-interested agents produce consensus routing and payoffs tied to marginal contribution. The authors prove incentive compatibility and Nash properties and demonstrate improved collaborative outcomes in simulated benchmarks using open and proprietary language models.
medium theoretical Full text arxiv
AgentSociety: Incentivizing Agentic Social Intelligence · Aditya Vema Reddy Kesari, Krishna Reddy Kesari · May 25, 2026
Fast, error-prone AI prompts reflexive human compliance that collapses team accuracy, while slower, accurate AI creates hesitation but enables recovery; dynamically gating AI interventions by neural timing restores and speeds up team performance.
medium rct Full text arxiv
The Timing Dependencies of Trust: Speed, Accuracy, and cBCI Neuro-Decoupling in Human-AI Teams · Christopher Baker, Stephen Hinton, Akashdeep Nijjar, Riccardo Poli, Caterina Cin... · May 25, 2026
Embedding data-centre water impacts directly into power dispatch reduces generation-side freshwater withdrawals by about 3–5% in IEEE test networks; a differentiable optimization layer and fixed-point coordination align virtual water attribution with physical withdrawals to enable water-aware workload relocation.
low theoretical Full text arxiv
From Accounting to Coordination: A Virtual Water-Aware Electricity-Computation-Water Nexus Framework for Data Center Dispatch · Haiyang You, Chengwei Lou, Jin Zhao, Yue Zhou, Lu Zhang, Jin Yang · May 25, 2026
A leading agent-to-agent marketplace rewards quantity over quality: 98% of shared assets on EvoMap are never reused and credits accrue to a small minority, while the platform's quality score and validation checks are easily gamed so most approved assets bypass meaningful tests.
medium descriptive Full text arxiv
Behind EvoMap: Characterizing a Self-Evolving Agent-to-Agent Collaboration Network · Qiming Ye, Peixain Zhang, Yupeng He, Zifan Peng, Gareth Tyson · May 25, 2026
An actuarial runtime contract lets AI 'buy' authority: pricing and reserving capital for side-effecting actions reveals how much autonomy can be safely released. Tests across four environments and a live Postgres panel show common low-reserve refusal and intermediate-release behaviour, while required reserve capital varies widely (22×) across domains and models.
low theoretical Full text arxiv
Insuring Every Action: An Authority Frontier Framework for Runtime Actuarial Control of Autonomous AI Agents · Hao-Hsuan Chen · May 25, 2026
European regulators should treat emotionally interactive conversational AI as a distinct governance challenge, expanding risk assessments to cover interaction design, psychological safety and potential gatekeeping roles; coordinated monitoring of global rules and practical guidance for firms operating across divergent jurisdictions are needed to avoid fragmentation.
n/a commentary Full text semantic_scholar
Governing Relational AI: China’s Regulation of Anthropomorphic AI and the European Governance Challenge · Łukasz Gacek · Fetched May 25, 2026
Regional AI infrastructure choices inevitably trade off rapid progress, environmental sustainability, and equitable outcomes; absent deliberate governance, market power and regulatory inertia will default allocation toward narrow priorities.
low theoretical Full text semantic_scholar
The AI Infrastructure Triad in Regional Governance: How Regions Balance Progress, Sustainability, and Equity · Paulo Carvao, Tushar Kanade · Fetched May 25, 2026