{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:3AKINPDPEY5E4DLPOLPFFZDOXI","short_pith_number":"pith:3AKINPDP","schema_version":"1.0","canonical_sha256":"d81486bc6f263a4e0d6f72de52e46eba1eacab61c38fac9bd7279d6350bc4f6b","source":{"kind":"arxiv","id":"2504.19678","version":2},"attestation_state":"computed","paper":{"title":"From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A review organizes roughly 60 benchmarks for large language models and autonomous agents into one taxonomy covering reasoning, code, and real-world tasks.","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Merouane Debbah, Mohamed Amine Ferrag, Norbert Tihanyi","submitted_at":"2025-04-28T11:08:22Z","abstract_excerpt":"Large language models and autonomous AI agents have evolved rapidly, resulting in a diverse array of evaluation benchmarks, frameworks, and collaboration protocols. Driven by the growing need for standardized evaluation and integration, we systematically consolidate these fragmented efforts into a unified framework. However, the landscape remains fragmented and lacks a unified taxonomy or comprehensive survey. Therefore, we present a side-by-side comparison of benchmarks developed between 2019 and 2025 that evaluate these models and agents across multiple domains. In addition, we propose a tax"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2504.19678","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-04-28T11:08:22Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4dfa9e24fcc765bf15db0d4e228dd847b5d56f8dd817458a312a013cd5948841","abstract_canon_sha256":"fdacaba0a601f052017bef4adecc766750331cb6d24f763c7c1e3ddc86303bda"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:53.742358Z","signature_b64":"+huIwtPLopH0xkWG88XWhCdsv8D9/dKG1JDhgCA8cXXrWgF0y5GkAAtgEZu4XM+0bYlhV2wyEjsh75QvlNO9DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d81486bc6f263a4e0d6f72de52e46eba1eacab61c38fac9bd7279d6350bc4f6b","last_reissued_at":"2026-05-17T23:38:53.741627Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:53.741627Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A review organizes roughly 60 benchmarks for large language models and autonomous agents into one taxonomy covering reasoning, code, and real-world tasks.","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Merouane Debbah, Mohamed Amine Ferrag, Norbert Tihanyi","submitted_at":"2025-04-28T11:08:22Z","abstract_excerpt":"Large language models and autonomous AI agents have evolved rapidly, resulting in a diverse array of evaluation benchmarks, frameworks, and collaboration protocols. Driven by the growing need for standardized evaluation and integration, we systematically consolidate these fragmented efforts into a unified framework. However, the landscape remains fragmented and lacks a unified taxonomy or comprehensive survey. Therefore, we present a side-by-side comparison of benchmarks developed between 2019 and 2025 that evaluate these models and agents across multiple domains. In addition, we propose a tax"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"we present a side-by-side comparison of benchmarks developed between 2019 and 2025 that evaluate these models and agents across multiple domains... we propose a taxonomy of approximately 60 benchmarks that cover general and academic knowledge reasoning, mathematical problem-solving, code generation and software engineering, factual grounding and retrieval, domain-specific evaluations, multimodal and embodied tasks, task orchestration, and interactive assessments.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"the landscape remains fragmented and lacks a unified taxonomy or comprehensive survey, which the authors' proposed taxonomy of approximately 60 benchmarks is assumed to resolve without major omissions or selection bias in the covered works.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A survey consolidating benchmarks, agent frameworks, real-world applications, and protocols for LLM-based autonomous agents into a proposed taxonomy with recommendations for future research.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A review organizes roughly 60 benchmarks for large language models and autonomous agents into one taxonomy covering reasoning, code, and real-world tasks.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"12155fb9a0f94458fa89bacc37f0cce8fac32ae1941962551607bff5de458453"},"source":{"id":"2504.19678","kind":"arxiv","version":2},"verdict":{"id":"e97de2dc-ed32-4712-8b7f-1ded3d7cb153","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T02:53:18.046467Z","strongest_claim":"we present a side-by-side comparison of benchmarks developed between 2019 and 2025 that evaluate these models and agents across multiple domains... we propose a taxonomy of approximately 60 benchmarks that cover general and academic knowledge reasoning, mathematical problem-solving, code generation and software engineering, factual grounding and retrieval, domain-specific evaluations, multimodal and embodied tasks, task orchestration, and interactive assessments.","one_line_summary":"A survey consolidating benchmarks, agent frameworks, real-world applications, and protocols for LLM-based autonomous agents into a proposed taxonomy with recommendations for future research.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"the landscape remains fragmented and lacks a unified taxonomy or comprehensive survey, which the authors' proposed taxonomy of approximately 60 benchmarks is assumed to resolve without major omissions or selection bias in the covered works.","pith_extraction_headline":"A review organizes roughly 60 benchmarks for large language models and autonomous agents into one taxonomy covering reasoning, code, and real-world tasks."},"references":{"count":236,"sample":[{"doi":"","year":2024,"title":"OpenAI o1 System Card","work_id":"68d3c334-0fc9-49e3-b7b0-a69afae933e2","ref_index":1,"cited_arxiv_id":"2412.16720","is_internal_anchor":true},{"doi":"","year":2025,"title":"Qwen2.5-Omni Technical Report","work_id":"438f105c-fa9b-44aa-ad52-43acb8045cda","ref_index":2,"cited_arxiv_id":"2503.20215","is_internal_anchor":true},{"doi":"","year":2025,"title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","work_id":"e6b75ad5-2877-4168-97c8-710407094d20","ref_index":3,"cited_arxiv_id":"2501.12948","is_internal_anchor":true},{"doi":"","year":2024,"title":"The Llama 3 Herd of Models","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","ref_index":4,"cited_arxiv_id":"2407.21783","is_internal_anchor":true},{"doi":"","year":2024,"title":"Understanding the planning of LLM agents: A survey","work_id":"6daa5311-7f56-401d-94db-17be43e95cbc","ref_index":5,"cited_arxiv_id":"2402.02716","is_internal_anchor":true}],"resolved_work":236,"snapshot_sha256":"d9a8e56106c8959ad2eeb33339464e4cc83c02bd5819d7e3eb8b20a0074ce7b8","internal_anchors":43},"formal_canon":{"evidence_count":1,"snapshot_sha256":"df92f97a9cc1f0285d2d10bcc1fb6d1a80591f55a84b879debac6c25f0c3f401"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2504.19678","created_at":"2026-05-17T23:38:53.741752+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.19678v2","created_at":"2026-05-17T23:38:53.741752+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.19678","created_at":"2026-05-17T23:38:53.741752+00:00"},{"alias_kind":"pith_short_12","alias_value":"3AKINPDPEY5E","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"3AKINPDPEY5E4DLP","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"3AKINPDP","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":33,"internal_anchor_count":33,"sample":[{"citing_arxiv_id":"2601.11650","citing_title":"Large Language Model Agent for User-friendly Chemical Process Simulations","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2603.24858","citing_title":"Context-Mediated Domain Adaptation in Multi-Agent Sensemaking Systems","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21967","citing_title":"Reinforced Preference Optimization for Reasoning-Augmented Recommendations","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19418","citing_title":"Conflict-Resilient Multi-Agent Reasoning via Signed Graph Modeling","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10057","citing_title":"STAR: Failure-Aware Markovian Routing for Multi-Agent Spatiotemporal Reasoning","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2506.04565","citing_title":"From Standalone LLMs to Integrated Intelligence: A Survey of Compound Al Systems","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2508.13024","citing_title":"WebMall -- A Multi-Shop Benchmark for Evaluating Web Agents","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2510.04465","citing_title":"Autonomy Reshapes How Personalization Affects Privacy Concerns and Trust in LLM Agents","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2510.19973","citing_title":"A Tutorial on Cognitive Biases in Agentic AI-Driven 6G Autonomous Networks","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2511.19175","citing_title":"LLM-Based Agentic Negotiation for 6G: Addressing Uncertainty Neglect and Tail-Event Risk","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2511.23281","citing_title":"MCP vs RAG vs NLWeb vs HTML: A Comparison of the Effectiveness and Efficiency of Different Agent Interfaces to the Web (Technical Report)","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2602.23452","citing_title":"CiteAudit: You Cited It, But Did You Read It? A Benchmark for Verifying Scientific References in the LLM Era","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14460","citing_title":"Exploiting LLM Agent Supply Chains via Payload-less Skills","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13579","citing_title":"Position: Assistive Agents Need Accessibility Alignment","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2604.02870","citing_title":"Token Warping Helps MLLMs Look from Nearby Viewpoints","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2507.13334","citing_title":"A Survey of Context Engineering for Large Language Models","ref_index":289,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10057","citing_title":"STAR: Failure-Aware Markovian Routing for Multi-Agent Spatiotemporal Reasoning","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2501.09136","citing_title":"Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG","ref_index":58,"is_internal_anchor":true},{"citing_arxiv_id":"2604.26235","citing_title":"LATTICE: Evaluating Decision Support Utility of Crypto Agents","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10057","citing_title":"STAR: Failure-Aware Markovian Routing for Multi-Agent Spatiotemporal Reasoning","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2604.25189","citing_title":"AgentDID: Trustless Identity Authentication for AI Agents","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2604.23691","citing_title":"Intention-Aware Semantic Agent Communications for AI Glasses","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22577","citing_title":"QuantClaw: Precision Where It Matters for OpenClaw","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2605.01143","citing_title":"A Low-Latency Fraud Detection Layer for Detecting Adversarial Interaction Patterns in LLM-Powered Agents","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19260","citing_title":"Understanding the Mechanism of Altruism in Large Language Models","ref_index":246,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":1,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3AKINPDPEY5E4DLPOLPFFZDOXI","json":"https://pith.science/pith/3AKINPDPEY5E4DLPOLPFFZDOXI.json","graph_json":"https://pith.science/api/pith-number/3AKINPDPEY5E4DLPOLPFFZDOXI/graph.json","events_json":"https://pith.science/api/pith-number/3AKINPDPEY5E4DLPOLPFFZDOXI/events.json","paper":"https://pith.science/paper/3AKINPDP"},"agent_actions":{"view_html":"https://pith.science/pith/3AKINPDPEY5E4DLPOLPFFZDOXI","download_json":"https://pith.science/pith/3AKINPDPEY5E4DLPOLPFFZDOXI.json","view_paper":"https://pith.science/paper/3AKINPDP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.19678&json=true","fetch_graph":"https://pith.science/api/pith-number/3AKINPDPEY5E4DLPOLPFFZDOXI/graph.json","fetch_events":"https://pith.science/api/pith-number/3AKINPDPEY5E4DLPOLPFFZDOXI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3AKINPDPEY5E4DLPOLPFFZDOXI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3AKINPDPEY5E4DLPOLPFFZDOXI/action/storage_attestation","attest_author":"https://pith.science/pith/3AKINPDPEY5E4DLPOLPFFZDOXI/action/author_attestation","sign_citation":"https://pith.science/pith/3AKINPDPEY5E4DLPOLPFFZDOXI/action/citation_signature","submit_replication":"https://pith.science/pith/3AKINPDPEY5E4DLPOLPFFZDOXI/action/replication_record"}},"created_at":"2026-05-17T23:38:53.741752+00:00","updated_at":"2026-05-17T23:38:53.741752+00:00"}