Evidence (8974 claims)
Search and filter individual claims pulled from the papers. Looking for a specific finding ("what's the effect on wages?"), you're in the right place. Want to compare whole outcome categories against each other instead? Use the Evidence Explorer.
The board below groups claims two ways: by broad theme (nine paper-level topics) and by outcome category (the 34 claim-level outcomes that the Explorer and Syntheses also use).
Browse by theme
Nine broad, paper-level topics. Click one to filter the claims below.
Adoption
10085 claims
Filter claims →
Productivity
8974 claims
Filtered →
Governance
8062 claims
Filter claims →
Human-AI Collaboration
7749 claims
Filter claims →
Org Design
5057 claims
Filter claims →
Innovation
4896 claims
Filter claims →
Labor Markets
4088 claims
Filter claims →
Skills & Training
3372 claims
Filter claims →
Inequality
2377 claims
Filter claims →
Claims by outcome category
Counts by direction of finding. These are the same 34 outcome categories the Explorer compares and the Syntheses are written for. A linked row has a published synthesis.
| Outcome | Positive | Negative | Mixed | Null | Total |
|---|---|---|---|---|---|
| Other | 882 | 244 | 117 | 1097 | 2424 |
| Governance & Regulation | 1010 | 469 | 229 | 135 | 1875 |
| Organizational Efficiency | 977 | 235 | 149 | 90 | 1462 |
| Technology Adoption Rate | 781 | 299 | 143 | 128 | 1362 |
| Research Productivity | 506 | 155 | 74 | 363 | 1110 |
| Output Quality | 555 | 219 | 71 | 70 | 915 |
| Decision Quality | 395 | 200 | 95 | 54 | 751 |
| Firm Productivity | 523 | 67 | 101 | 27 | 724 |
| AI Safety & Ethics | 262 | 309 | 75 | 36 | 688 |
| Market Structure | 195 | 201 | 135 | 30 | 566 |
| Task Allocation | 248 | 77 | 96 | 38 | 464 |
| Innovation Output | 300 | 34 | 55 | 20 | 411 |
| Skill Acquisition | 207 | 75 | 65 | 21 | 368 |
| Employment Level | 138 | 67 | 119 | 24 | 350 |
| Fiscal & Macroeconomic | 156 | 80 | 53 | 33 | 329 |
| Task Completion Time | 211 | 38 | 13 | 16 | 280 |
| Firm Revenue | 183 | 52 | 29 | 5 | 270 |
| Consumer Welfare | 131 | 77 | 48 | 13 | 269 |
| Inequality Measures | 50 | 141 | 54 | 9 | 254 |
| Worker Satisfaction | 104 | 85 | 25 | 13 | 227 |
| Error Rate | 87 | 112 | 11 | 5 | 215 |
| Automation Exposure | 69 | 69 | 37 | 20 | 198 |
| Wages & Compensation | 102 | 49 | 31 | 11 | 193 |
| Team Performance | 115 | 30 | 30 | 11 | 187 |
| Regulatory Compliance | 88 | 74 | 17 | 7 | 186 |
| Training Effectiveness | 109 | 22 | 14 | 21 | 168 |
| Developer Productivity | 116 | 21 | 15 | 8 | 161 |
| Job Displacement | 12 | 92 | 26 | 1 | 131 |
| Hiring & Recruitment | 57 | 12 | 9 | 5 | 83 |
| Skill Obsolescence | 6 | 59 | 10 | 2 | 77 |
| Social Protection | 43 | 17 | 8 | 2 | 70 |
| Creative Output | 35 | 21 | 9 | 4 | 70 |
| Labor Share of Income | 18 | 23 | 17 | 1 | 59 |
| Worker Turnover | 15 | 16 | — | 4 | 35 |
| Industry | — | — | — | 1 | 1 |
Productivity
Remove filter
A formative study (N = 8) and a within-subjects summative evaluation (N = 16) comparing Pista to a baseline agent demonstrated that active participation in execution influenced not only task outcomes but also users' comprehension of the task, their perception of the agent, and their sense of role within the workflow.
Empirical evaluation consisting of a formative study with N=8 and a within-subjects summative evaluation with N=16 comparing Pista to a baseline agent (authors report influence on task outcomes, comprehension, perception, and role).
We introduce Pista, a spreadsheet AI agent that decomposes execution into auditable, controllable actions, providing users with visibility into the agent's decision-making process and the capacity to intervene at each step.
System description / design contribution presented by the authors (implementation description rather than empirical evidence).
Selective forgetting should be considered a fundamental capability for next-generation LLM agents operating in real-world, resource-constrained scenarios.
Conclusion/argument in paper based on conceptual analysis and reported empirical benefits.
The work bridges cognitive neuroscience (hippocampal indexing/consolidation theory and Ebbinghaus forgetting curve) and AI systems to inform forgetting mechanisms.
Claimed theoretical grounding and cross-disciplinary framing in paper (stated in abstract).
Empirical results show security performance with 100% elimination of security risks.
Reported experimental result in abstract claiming full elimination of security risks.
Empirical results show content quality improved by +29.2% signal-to-noise ratio.
Reported experimental result in abstract (signal-to-noise ratio improvement).
Empirical results show access efficiency improved by +8.49%.
Reported experimental result in abstract.
Building on advances in LLM agent architectures and vector databases, the paper presents detailed specifications, implementation strategies, and empirical validation from controlled experiments.
Methodological claim in abstract indicating implementation and controlled experiments (no experimental details in abstract).
Selective forgetting improves security through active forgetting of malicious inputs, sensitive data, and privacy-compromising content.
Authors' taxonomy and safety-triggered forgetting mechanism; abstract reports empirical security performance (100% elimination of security risks).
Selective forgetting improves content quality by dynamically updating outdated preferences and context.
Conceptual claim supported by authors' implementation and empirical validation; abstract reports content quality improvement (signal-to-noise ratio).
A well-designed forgetting mechanism improves efficiency via intelligent memory pruning.
Claim supported by authors' framework and controlled experiments reported in the paper (abstract references empirical results for access efficiency).
In resource-constrained environments, a well-designed forgetting mechanism is as crucial as remembering.
Argument and conceptual analysis in paper; motivated by theoretical considerations and (claimed) empirical validation.
The findings point to a staged progression of AI utility from low-consequence assistance toward higher-order automation, as trust, infrastructure, and verification mature.
Synthesis of interview responses (over 30) indicating current use cases are lower-risk assistance and that stakeholders expect (or prefer) gradual progression toward automation contingent on trust/infrastructure/verification improvements.
Reliability, verification, and auditability are central requirements for adoption, driving human-in-the-loop frameworks and governance aligned with existing engineering reviews.
Consistent themes from interviews (over 30) indicating stakeholders prioritize reliability, verifiability, and audit trails, leading to preference for human-in-the-loop designs integrated with current review processes.
Higher-value agentic gains come from orchestrating multi-step workflows across tools.
Observed and reported in interviews (over 30) with stakeholders in engineering and manufacturing workflows describing value from agentic orchestration across tools.
Near-term AI gains cluster around structured, repetitive work and data-intensive synthesis.
Qualitative findings from an exploratory state-of-practice study based on over 30 semi-structured interviews across four stakeholder groups (large enterprises, small/medium firms, AI developers, and CAD/CAM/CAE vendors).
SWE-chat is a living dataset; our collection pipeline automatically and continually discovers and processes sessions from public repositories.
Description of the dataset collection infrastructure and pipeline provided in the paper; operational behavior asserted by authors.
The dataset currently contains 6,000 sessions, comprising more than 63,000 user prompts and 355,000 agent tool calls.
Descriptive statistics reported by the authors based on their dataset collection pipeline (dataset metadata).
We present SWE-chat, the first large-scale dataset of real coding agent sessions collected from open-source developers in the wild.
Paper authorship / dataset description; dataset curated and presented by the paper as a contribution. No external validation provided in excerpt.
Statelessness is the load-bearing property explaining enterprises' preference for weaker but replayable retrieval pipelines, and DPM demonstrates this property is attainable without the decisioning penalty retrieval pays.
Synthesis/conclusion based on theoretical argument and empirical results presented (architectural analysis + experiments showing DPM performance and auditability).
The audit surface follows the same one-versus-N pattern: DPM logs two LLM calls per decision while summarization logs 83-97 on LongHorizon-Bench.
Empirical measurement on LongHorizon-Bench reported in the paper: logged LLM calls per decision are 2 for DPM vs 83-97 for summarization.
DPM is additionally 7-15x faster at binding budgets, making one LLM call at decision time instead of N.
Empirical runtime/efficiency measurement reported in the paper (range 7-15x speedup) comparing number of LLM calls and latency under tight memory budgets.
At a 20x compression ratio, DPM improves reasoning coherence by +0.53 (Cohen's h=1.13, p=0.0034) compared to summarization-based memory (paired permutation, n=10).
Paired permutation test over 10 cases at a 20x compression ratio; reported effect +0.53 with Cohen's h=1.13 and p=0.0034.
At a 20x compression ratio, DPM improves factual precision by +0.52 (Cohen's h=1.17, p=0.0014) compared to summarization-based memory (paired permutation, n=10).
Paired permutation test over 10 cases at a 20x compression ratio; reported effect +0.52 with Cohen's h=1.17 and p=0.0014.
On ten regulated decisioning cases at three memory budgets, DPM matches summarization-based memory at generous budgets and substantially outperforms it when the budget binds.
Empirical evaluation on 10 decisioning cases across three memory budgets; comparison between DPM and summarization-based memory as reported in the paper (n=10).
We propose Deterministic Projection Memory (DPM): an append-only event log plus one task-conditioned projection at decision time.
Method/architectural proposal described in the paper.
Long-term prospects of agentic AI include catalyzing accelerated innovation in physical design via autonomous algorithm discovery, continuous tool improvement, and closed-loop learning from large design corpora.
Forward-looking conclusion in the paper; framed as the authors' projection based on survey synthesis rather than as an empirically demonstrated outcome in the abstract.
Interfaces between agentic systems and traditional EDA frameworks are a key area of focus and enable tighter integration of agent capabilities into existing design workflows.
Survey highlights interfaces between agents and EDA frameworks as a focus area; claim is descriptive of research direction rather than reporting empirical outcomes.
Autonomous agents can explore heuristic spaces for placement, routing, and partitioning, enabling autonomous exploration of design heuristics.
Presented as an emphasized capability/area of research in the survey; the abstract asserts this possibility but does not report empirical benchmarks or sample sizes.
Tool-integrated agents can be used for algorithm evolution, debugging, and workflow automation in physical design R&D.
Paper emphasizes this as a primary area of application in the survey; rationale and examples are discussed but no quantitative trial sizes are given in the abstract.
Agentic AI systems can comprehend user specifications, modify code, run EDA tools, analyze results, perform multi-step reasoning, and iteratively refine design heuristics—unlike earlier ML uses that focused narrowly on prediction or optimization subroutines.
Descriptive claim in the paper contrasting agentic AI capabilities with earlier ML approaches; presented as an overview of functional capabilities rather than empirical measurement.
Recent advances in large language models (LLMs) and tool-using autonomous agents present new opportunities for accelerating research and development in physical design.
Stated as a central thesis in the paper's abstract/survey; based on the authors' synthesis of recent advances and emerging applications (no empirical sample or quantified evaluation reported in the abstract).
The same user study (n=32) reports improvements in subjective measures including fluency and user preference for RAPIDDS over non-adaptive systems.
User study (n=32) reporting subjective questionnaire/ratings (fluency, preference) comparing RAPIDDS vs non-adaptive baselines.
A user study (n=32) shows significant plan improvement compared to non-adaptive systems across objective metrics such as efficiency and proximity.
User study reported in paper with sample size n=32 comparing RAPIDDS to non-adaptive systems on objective metrics (efficiency, proximity); significance claimed.
An ablation study in simulation and a physical robot scenario demonstrates the importance of dual (task + motion) adaptation.
Ablation experiments reported in paper (simulation and physical robot experiments comparing full RAPIDDS to ablated variants).
RAPIDDS jointly adapts task schedules and steers diffusion models of robot motions to maximize efficiency and minimize proximity accounting for individualized models.
Algorithmic method described in paper combining schedule optimization with motion steering (method section).
In the ICT industry, Tobin's Q significantly increased following AI adoption (heterogeneous positive effect).
Subgroup/heterogeneity analysis within the main sample (KOSDAQ firms 2018–2025), estimating the post-adoption effect of AI on Tobin's Q in firms classified as ICT.
Our baseline model finds evidence that AI is productivity enhancing.
Results from the paper's stated baseline empirical model using BEA industry-account-based measures; model specification described by authors.
ClawNet enables multiple users to collaborate securely through their respective agents.
Capability claim about the instantiated system (authors assert that ClawNet enables secure multi-user collaboration; excerpt contains no empirical security evaluation or user study).
We instantiate this paradigm in ClawNet, an identity-governed agent collaboration framework that enforces identity binding and authorization verification through a central orchestrator.
Implementation claim: authors state they built ClawNet as an instantiation of their paradigm (paper describes framework/architecture; no experimental evaluation included in excerpt).
Action-level accountability logs every operation against its owner's identity and authorization, ensuring full auditability.
Design claim describing an accountability primitive (paper asserts logging and auditability as a property; no audit or verification evidence shown in excerpt).
Scoped authorization enforces per-identity access control and escalates boundary violations to the owner.
Design/specification claim describing the scoped authorization governance primitive in the proposed paradigm (no empirical or security evaluation provided in excerpt).
The paradigm rests on three governance primitives: (1) a layered identity architecture that separates a Manager Agent from multiple context-specific Identity Agents; the Manager Agent holds global knowledge but is architecturally isolated from external communication.
Architectural/design claim describing the proposed layered identity primitive (presentation of design; no empirical validation in excerpt).
We propose a human-symbiotic agent paradigm in which each user owns a permanently bound agent system that collaborates on the owner's behalf, forming a network whose nodes are humans rather than agents.
Design proposal / conceptual architecture presented in the paper (no large-scale deployment or empirical evaluation described in excerpt).
The next frontier for AI agents lies not in stronger individual capability, but in the digitization of human collaborative relationships.
Normative/strategic claim advanced by the authors as the central thesis (conceptual argument, no empirical test reported).
Human productivity rests on the social and organizational relationships through which people coordinate, negotiate, and delegate.
Theoretical/argumentative claim presented as background motivation (conceptual reasoning, citation not provided in excerpt).
Time Series Augmented Generation (TSAG) enables LLM agents to delegate quantitative tasks to verifiable external tools.
Description of TSAG framework in paper stating delegation mechanism to external verifiable tools for quantitative computations.
We publicly release the evaluation framework and empirical insights to foster standardized research on reliable financial AI.
Paper states that the framework, benchmark, and empirical results are released publicly by the authors.
The results demonstrate that capable agents can achieve near-perfect tool-use accuracy with minimal hallucination, validating the tool-augmented paradigm.
Empirical results from the authors' experiments on the 100-question benchmark across multiple agents; paper states agents achieve 'near-perfect' tool-use accuracy and 'minimal' hallucination.
We apply this methodology in a large-scale empirical study using our framework, Time Series Augmented Generation (TSAG), where an LLM agent delegates quantitative tasks to verifiable, external tools.
Paper reports applying the TSAG framework in an empirical study in which agents call external tools to perform quantitative computations; described as 'large-scale' and implemented by the authors.