Collaborating with AI Agents: Field Experiments on Teamwork, Productivity, and Performancecs.AI updates on arXiv.org arXiv:2503.18238v3 Announce Type: replace-cross
Abstract: We examined the mechanisms underlying productivity and performance gains from AI agents using a large-scale experiment on Pairit, a platform we developed to study human-AI collaboration. We randomly assigned 2,234 participants to human-human and human-AI teams that produced 11,024 ads for a think tank. We evaluated the ads using independent human ratings and a field experiment on X which garnered ~5M impressions. We found human-AI teams produced 50% more ads per worker and higher text quality, while human-human teams produced higher image quality, suggesting a jagged frontier of AI agent capability. Human-AI teams also produced more homogeneous, or self-similar, outputs. The field experiment revealed higher text quality improved click-through rates and view-through duration, while higher image quality improved cost-per-click rates. We found three mechanisms explained these effects. First, human-AI collaboration was more task-oriented, with 25% more task-oriented messages and 18% fewer interpersonal messages. Second, human-AI collaboration displayed more delegation, as participants delegated 17% more work to AI agents than to human partners and performed 62% fewer direct text edits when working with AI. Third, recognition that the collaborator was an AI moderated these effects as participants who correctly identified they were working with AI were more task-oriented and more likely to delegate work. These mechanisms then explained performance as task-oriented communication improved ad quality, specifically when working with AI, while interpersonal communication reduced ad quality; delegation improved text quality but had no effect on image quality and was positively associated with diversity collapse, creating homogeneous outputs of higher average quality. The results suggest AI agents drive changes in productivity, performance, and output diversity by reshaping teamwork.
arXiv:2503.18238v3 Announce Type: replace-cross
Abstract: We examined the mechanisms underlying productivity and performance gains from AI agents using a large-scale experiment on Pairit, a platform we developed to study human-AI collaboration. We randomly assigned 2,234 participants to human-human and human-AI teams that produced 11,024 ads for a think tank. We evaluated the ads using independent human ratings and a field experiment on X which garnered ~5M impressions. We found human-AI teams produced 50% more ads per worker and higher text quality, while human-human teams produced higher image quality, suggesting a jagged frontier of AI agent capability. Human-AI teams also produced more homogeneous, or self-similar, outputs. The field experiment revealed higher text quality improved click-through rates and view-through duration, while higher image quality improved cost-per-click rates. We found three mechanisms explained these effects. First, human-AI collaboration was more task-oriented, with 25% more task-oriented messages and 18% fewer interpersonal messages. Second, human-AI collaboration displayed more delegation, as participants delegated 17% more work to AI agents than to human partners and performed 62% fewer direct text edits when working with AI. Third, recognition that the collaborator was an AI moderated these effects as participants who correctly identified they were working with AI were more task-oriented and more likely to delegate work. These mechanisms then explained performance as task-oriented communication improved ad quality, specifically when working with AI, while interpersonal communication reduced ad quality; delegation improved text quality but had no effect on image quality and was positively associated with diversity collapse, creating homogeneous outputs of higher average quality. The results suggest AI agents drive changes in productivity, performance, and output diversity by reshaping teamwork. Read More
Automated Customization of LLMs for Enterprise Code Repositories Using Semantic Scopescs.AI updates on arXiv.org arXiv:2602.05780v1 Announce Type: cross
Abstract: Code completion (CC) is a task frequently used by developers when working in collaboration with LLM-based programming assistants. Despite the increased performance of LLMs on public benchmarks, out of the box LLMs still have a hard time generating code that aligns with a private code repository not previously seen by the model’s training data. Customizing code LLMs to a private repository provides a way to improve the model performance. In this paper we present our approach for automated LLM customization based on semantic scopes in the code. We evaluate LLMs on real industry cases with two private enterprise code repositories with two customization strategies: Retrieval-Augmented Generation (RAG) and supervised Fine-Tuning (FT). Our mechanism for ingesting the repository’s data and formulating the training data pairs with semantic scopes helps models to learn the underlying patterns specific to the repository, providing more precise code to developers and helping to boost their productivity. The code completions of moderately sized customized models can be significantly better than those of uncustomized models of much larger capacity. We also include an analysis of customization on two public benchmarks and present opportunities for future work.
arXiv:2602.05780v1 Announce Type: cross
Abstract: Code completion (CC) is a task frequently used by developers when working in collaboration with LLM-based programming assistants. Despite the increased performance of LLMs on public benchmarks, out of the box LLMs still have a hard time generating code that aligns with a private code repository not previously seen by the model’s training data. Customizing code LLMs to a private repository provides a way to improve the model performance. In this paper we present our approach for automated LLM customization based on semantic scopes in the code. We evaluate LLMs on real industry cases with two private enterprise code repositories with two customization strategies: Retrieval-Augmented Generation (RAG) and supervised Fine-Tuning (FT). Our mechanism for ingesting the repository’s data and formulating the training data pairs with semantic scopes helps models to learn the underlying patterns specific to the repository, providing more precise code to developers and helping to boost their productivity. The code completions of moderately sized customized models can be significantly better than those of uncustomized models of much larger capacity. We also include an analysis of customization on two public benchmarks and present opportunities for future work. Read More
STELLAR: Structure-guided LLM Assertion Retrieval and Generation for Formal Verificationcs.AI updates on arXiv.org arXiv:2601.19903v2 Announce Type: replace-cross
Abstract: Formal Verification (FV) relies on high-quality SystemVerilog Assertions (SVAs), but the manual writing process is slow and error-prone. Existing LLM-based approaches either generate assertions from scratch or ignore structural patterns in hardware designs and expert-crafted assertions. This paper presents STELLAR, the first framework that guides LLM-based SVA generation with structural similarity. STELLAR represents RTL blocks as AST structural fingerprints, retrieves structurally relevant (RTL, SVA) pairs from a knowledge base, and integrates them into structure-guided prompts. Experiments show that STELLAR achieves superior syntax correctness, stylistic alignment, and functional correctness, highlighting structure-aware retrieval as a promising direction for industrial FV.
arXiv:2601.19903v2 Announce Type: replace-cross
Abstract: Formal Verification (FV) relies on high-quality SystemVerilog Assertions (SVAs), but the manual writing process is slow and error-prone. Existing LLM-based approaches either generate assertions from scratch or ignore structural patterns in hardware designs and expert-crafted assertions. This paper presents STELLAR, the first framework that guides LLM-based SVA generation with structural similarity. STELLAR represents RTL blocks as AST structural fingerprints, retrieves structurally relevant (RTL, SVA) pairs from a knowledge base, and integrates them into structure-guided prompts. Experiments show that STELLAR achieves superior syntax correctness, stylistic alignment, and functional correctness, highlighting structure-aware retrieval as a promising direction for industrial FV. Read More
Enhancing Personality Recognition by Comparing the Predictive Power of Traits, Facets, and Nuancescs.AI updates on arXiv.org arXiv:2602.05650v1 Announce Type: cross
Abstract: Personality is a complex, hierarchical construct typically assessed through item-level questionnaires aggregated into broad trait scores. Personality recognition models aim to infer personality traits from different sources of behavioral data. However, reliance on broad trait scores as ground truth, combined with limited training data, poses challenges for generalization, as similar trait scores can manifest through diverse, context dependent behaviors. In this work, we explore the predictive impact of the more granular hierarchical levels of the Big-Five Personality Model, facets and nuances, to enhance personality recognition from audiovisual interaction data. Using the UDIVA v0.5 dataset, we trained a transformer-based model including cross-modal (audiovisual) and cross-subject (dyad-aware) attention mechanisms. Results show that nuance-level models consistently outperform facet and trait-level models, reducing mean squared error by up to 74% across interaction scenarios.
arXiv:2602.05650v1 Announce Type: cross
Abstract: Personality is a complex, hierarchical construct typically assessed through item-level questionnaires aggregated into broad trait scores. Personality recognition models aim to infer personality traits from different sources of behavioral data. However, reliance on broad trait scores as ground truth, combined with limited training data, poses challenges for generalization, as similar trait scores can manifest through diverse, context dependent behaviors. In this work, we explore the predictive impact of the more granular hierarchical levels of the Big-Five Personality Model, facets and nuances, to enhance personality recognition from audiovisual interaction data. Using the UDIVA v0.5 dataset, we trained a transformer-based model including cross-modal (audiovisual) and cross-subject (dyad-aware) attention mechanisms. Results show that nuance-level models consistently outperform facet and trait-level models, reducing mean squared error by up to 74% across interaction scenarios. Read More
VibeCodeHPC: An Agent-Based Iterative Prompting Auto-Tuner for HPC Code Generation Using LLMscs.AI updates on arXiv.org arXiv:2510.00031v2 Announce Type: replace-cross
Abstract: We propose VibeCodeHPC, an automatic tuning system for HPC programs based on multi-agent LLMs for code generation. VibeCodeHPC tunes programs through multi-agent role allocation and iterative prompt refinement. We describe the system configuration with four roles: Project Manager (PM), System Engineer (SE), Programmer (PG), and Continuous Delivery (CD). We introduce dynamic agent deployment and activity monitoring functions to facilitate effective multi-agent collaboration. In our case study, we convert and optimize CPU-based matrix-matrix multiplication code written in C to GPU code using CUDA. The multi-agent configuration of VibeCodeHPC achieved higher-quality code generation per unit time compared to a solo-agent configuration. Additionally, the dynamic agent deployment and activity monitoring capabilities facilitated more effective identification of requirement violations and other issues.
arXiv:2510.00031v2 Announce Type: replace-cross
Abstract: We propose VibeCodeHPC, an automatic tuning system for HPC programs based on multi-agent LLMs for code generation. VibeCodeHPC tunes programs through multi-agent role allocation and iterative prompt refinement. We describe the system configuration with four roles: Project Manager (PM), System Engineer (SE), Programmer (PG), and Continuous Delivery (CD). We introduce dynamic agent deployment and activity monitoring functions to facilitate effective multi-agent collaboration. In our case study, we convert and optimize CPU-based matrix-matrix multiplication code written in C to GPU code using CUDA. The multi-agent configuration of VibeCodeHPC achieved higher-quality code generation per unit time compared to a solo-agent configuration. Additionally, the dynamic agent deployment and activity monitoring capabilities facilitated more effective identification of requirement violations and other issues. Read More
Learning to Share: Selective Memory for Efficient Parallel Agentic Systemscs.AI updates on arXiv.org arXiv:2602.05965v1 Announce Type: cross
Abstract: Agentic systems solve complex tasks by coordinating multiple agents that iteratively reason, invoke tools, and exchange intermediate results. To improve robustness and solution quality, recent approaches deploy multiple agent teams running in parallel to explore diverse reasoning trajectories. However, parallel execution comes at a significant computational cost: when different teams independently reason about similar sub-problems or execute analogous steps, they repeatedly perform substantial overlapping computation. To address these limitations, in this paper, we propose Learning to Share (LTS), a learned shared-memory mechanism for parallel agentic frameworks that enables selective cross-team information reuse while controlling context growth. LTS introduces a global memory bank accessible to all teams and a lightweight controller that decides whether intermediate agent steps should be added to memory or not. The controller is trained using stepwise reinforcement learning with usage-aware credit assignment, allowing it to identify information that is globally useful across parallel executions. Experiments on the AssistantBench and GAIA benchmarks show that LTS significantly reduces overall runtime while matching or improving task performance compared to memory-free parallel baselines, demonstrating that learned memory admission is an effective strategy for improving the efficiency of parallel agentic systems. Project page: https://joefioresi718.github.io/LTS_webpage/
arXiv:2602.05965v1 Announce Type: cross
Abstract: Agentic systems solve complex tasks by coordinating multiple agents that iteratively reason, invoke tools, and exchange intermediate results. To improve robustness and solution quality, recent approaches deploy multiple agent teams running in parallel to explore diverse reasoning trajectories. However, parallel execution comes at a significant computational cost: when different teams independently reason about similar sub-problems or execute analogous steps, they repeatedly perform substantial overlapping computation. To address these limitations, in this paper, we propose Learning to Share (LTS), a learned shared-memory mechanism for parallel agentic frameworks that enables selective cross-team information reuse while controlling context growth. LTS introduces a global memory bank accessible to all teams and a lightweight controller that decides whether intermediate agent steps should be added to memory or not. The controller is trained using stepwise reinforcement learning with usage-aware credit assignment, allowing it to identify information that is globally useful across parallel executions. Experiments on the AssistantBench and GAIA benchmarks show that LTS significantly reduces overall runtime while matching or improving task performance compared to memory-free parallel baselines, demonstrating that learned memory admission is an effective strategy for improving the efficiency of parallel agentic systems. Project page: https://joefioresi718.github.io/LTS_webpage/ Read More
Interpretability by Design for Efficient Multi-Objective Reinforcement Learningcs.AI updates on arXiv.org arXiv:2506.04022v2 Announce Type: replace
Abstract: Multi-objective reinforcement learning (MORL) aims at optimising several, often conflicting goals to improve the flexibility and reliability of RL in practical tasks. This is typically achieved by finding a set of diverse, non-dominated policies that form a Pareto front in the performance space. We introduce LLE-MORL, an approach that achieves interpretability by design by utilising a training scheme based on the local relationship between the parameter space and the performance space. By exploiting a locally linear map between these spaces, our method provides an interpretation of policy parameters in terms of the objectives, and this structured representation enables an efficient search within contiguous solution domains, allowing for the rapid generation of high-quality solutions without extensive retraining. Experiments across diverse continuous control domains demonstrate that LLE-MORL consistently achieves higher Pareto front quality and efficiency than state-of-the-art approaches.
arXiv:2506.04022v2 Announce Type: replace
Abstract: Multi-objective reinforcement learning (MORL) aims at optimising several, often conflicting goals to improve the flexibility and reliability of RL in practical tasks. This is typically achieved by finding a set of diverse, non-dominated policies that form a Pareto front in the performance space. We introduce LLE-MORL, an approach that achieves interpretability by design by utilising a training scheme based on the local relationship between the parameter space and the performance space. By exploiting a locally linear map between these spaces, our method provides an interpretation of policy parameters in terms of the objectives, and this structured representation enables an efficient search within contiguous solution domains, allowing for the rapid generation of high-quality solutions without extensive retraining. Experiments across diverse continuous control domains demonstrate that LLE-MORL consistently achieves higher Pareto front quality and efficiency than state-of-the-art approaches. Read More
Vibe AIGC: A New Paradigm for Content Generation via Agentic Orchestrationcs.AI updates on arXiv.org arXiv:2602.04575v2 Announce Type: replace
Abstract: For the past decade, the trajectory of generative artificial intelligence (AI) has been dominated by a model-centric paradigm driven by scaling laws. Despite significant leaps in visual fidelity, this approach has encountered a “usability ceiling” manifested as the Intent-Execution Gap (i.e., the fundamental disparity between a creator’s high-level intent and the stochastic, black-box nature of current single-shot models). In this paper, inspired by the Vibe Coding, we introduce the textbf{Vibe AIGC}, a new paradigm for content generation via agentic orchestration, which represents the autonomous synthesis of hierarchical multi-agent workflows.
Under this paradigm, the user’s role transcends traditional prompt engineering, evolving into a Commander who provides a Vibe, a high-level representation encompassing aesthetic preferences, functional logic, and etc. A centralized Meta-Planner then functions as a system architect, deconstructing this “Vibe” into executable, verifiable, and adaptive agentic pipelines. By transitioning from stochastic inference to logical orchestration, Vibe AIGC bridges the gap between human imagination and machine execution. We contend that this shift will redefine the human-AI collaborative economy, transforming AI from a fragile inference engine into a robust system-level engineering partner that democratizes the creation of complex, long-horizon digital assets.
arXiv:2602.04575v2 Announce Type: replace
Abstract: For the past decade, the trajectory of generative artificial intelligence (AI) has been dominated by a model-centric paradigm driven by scaling laws. Despite significant leaps in visual fidelity, this approach has encountered a “usability ceiling” manifested as the Intent-Execution Gap (i.e., the fundamental disparity between a creator’s high-level intent and the stochastic, black-box nature of current single-shot models). In this paper, inspired by the Vibe Coding, we introduce the textbf{Vibe AIGC}, a new paradigm for content generation via agentic orchestration, which represents the autonomous synthesis of hierarchical multi-agent workflows.
Under this paradigm, the user’s role transcends traditional prompt engineering, evolving into a Commander who provides a Vibe, a high-level representation encompassing aesthetic preferences, functional logic, and etc. A centralized Meta-Planner then functions as a system architect, deconstructing this “Vibe” into executable, verifiable, and adaptive agentic pipelines. By transitioning from stochastic inference to logical orchestration, Vibe AIGC bridges the gap between human imagination and machine execution. We contend that this shift will redefine the human-AI collaborative economy, transforming AI from a fragile inference engine into a robust system-level engineering partner that democratizes the creation of complex, long-horizon digital assets. Read More
How to Build a Production-Grade Agentic AI System with Hybrid Retrieval, Provenance-First Citations, Repair Loops, and Episodic MemoryMarkTechPost In this tutorial, we build an ultra-advanced agentic AI workflow that behaves like a production-grade research and reasoning system rather than a single prompt call. We ingest real web sources asynchronously, split them into provenance-tracked chunks, and run hybrid retrieval using both TF-IDF (sparse) and OpenAI embeddings (dense), then fuse results for higher recall and
The post How to Build a Production-Grade Agentic AI System with Hybrid Retrieval, Provenance-First Citations, Repair Loops, and Episodic Memory appeared first on MarkTechPost.
In this tutorial, we build an ultra-advanced agentic AI workflow that behaves like a production-grade research and reasoning system rather than a single prompt call. We ingest real web sources asynchronously, split them into provenance-tracked chunks, and run hybrid retrieval using both TF-IDF (sparse) and OpenAI embeddings (dense), then fuse results for higher recall and
The post How to Build a Production-Grade Agentic AI System with Hybrid Retrieval, Provenance-First Citations, Repair Loops, and Episodic Memory appeared first on MarkTechPost. Read More
The European Commission said today that TikTok is facing a fine because its addictive features, including infinite scroll, autoplay, push notifications, and personalized recommendation systems, are breaching the EU’s Digital Services Act (DSA). […] Read More
