{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:COB3EHCQNSZCGGJSNJOXVGSKKR","short_pith_number":"pith:COB3EHCQ","schema_version":"1.0","canonical_sha256":"1383b21c506cb22319326a5d7a9a4a546ae928460fffc7746a95d715f10910a7","source":{"kind":"arxiv","id":"2411.13543","version":2},"attestation_state":"computed","paper":{"title":"BALROG: Benchmarking Agentic LLM and VLM Reasoning On Games","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Akbir Khan, Bart{\\l}omiej Cupia{\\l}, Davide Paglieri, Eduardo Pignatelli, Jack Parker-Holder, Jakob Nicolaus Foerster, Lerrel Pinto, {\\L}ukasz Kuci\\'nski, Maciej Wolczyk, Rob Fergus, Samuel Coward, Tim Rockt\\\"aschel, Ulyana Piterbarg","submitted_at":"2024-11-20T18:54:32Z","abstract_excerpt":"Large Language Models (LLMs) and Vision Language Models (VLMs) possess extensive knowledge and exhibit promising reasoning abilities, however, they still struggle to perform well in complex, dynamic environments. Real-world tasks require handling intricate interactions, advanced spatial reasoning, long-term planning, and continuous exploration of new strategies-areas in which we lack effective methodologies for comprehensively evaluating these capabilities. To address this gap, we introduce BALROG, a novel benchmark designed to assess the agentic capabilities of LLMs and VLMs through a diverse"},"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":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2411.13543","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-11-20T18:54:32Z","cross_cats_sorted":[],"title_canon_sha256":"b9a97646981814e2e98e8c3bf34eedb7bb5142867040266c8bf1e16d51eea563","abstract_canon_sha256":"87b9b42c92f9f03a221359f6441059d80652400597663ecec06b9de06f55b090"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:42:39.280541Z","signature_b64":"Az4aefS2sTS/xspZcxWZpBtYKvzfVr+tkl6ydYv+yHq71PjfXDpqjRyQr1wscNOuFd5CV0DVE4gmlI+xnC/vBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1383b21c506cb22319326a5d7a9a4a546ae928460fffc7746a95d715f10910a7","last_reissued_at":"2026-07-05T10:42:39.280009Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:42:39.280009Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BALROG: Benchmarking Agentic LLM and VLM Reasoning On Games","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Akbir Khan, Bart{\\l}omiej Cupia{\\l}, Davide Paglieri, Eduardo Pignatelli, Jack Parker-Holder, Jakob Nicolaus Foerster, Lerrel Pinto, {\\L}ukasz Kuci\\'nski, Maciej Wolczyk, Rob Fergus, Samuel Coward, Tim Rockt\\\"aschel, Ulyana Piterbarg","submitted_at":"2024-11-20T18:54:32Z","abstract_excerpt":"Large Language Models (LLMs) and Vision Language Models (VLMs) possess extensive knowledge and exhibit promising reasoning abilities, however, they still struggle to perform well in complex, dynamic environments. Real-world tasks require handling intricate interactions, advanced spatial reasoning, long-term planning, and continuous exploration of new strategies-areas in which we lack effective methodologies for comprehensively evaluating these capabilities. To address this gap, we introduce BALROG, a novel benchmark designed to assess the agentic capabilities of LLMs and VLMs through a diverse"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.13543","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2411.13543/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"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":"2411.13543","created_at":"2026-07-05T10:42:39.280074+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.13543v2","created_at":"2026-07-05T10:42:39.280074+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.13543","created_at":"2026-07-05T10:42:39.280074+00:00"},{"alias_kind":"pith_short_12","alias_value":"COB3EHCQNSZC","created_at":"2026-07-05T10:42:39.280074+00:00"},{"alias_kind":"pith_short_16","alias_value":"COB3EHCQNSZCGGJS","created_at":"2026-07-05T10:42:39.280074+00:00"},{"alias_kind":"pith_short_8","alias_value":"COB3EHCQ","created_at":"2026-07-05T10:42:39.280074+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":23,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18950","citing_title":"RTSGameBench: An RTS Benchmark for Strategic Reasoning by Vision-Language Models","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19338","citing_title":"Beyond the Current Observation: Evaluating Multimodal Large Language Models in Controllable Non-Markov Games","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13608","citing_title":"AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12191","citing_title":"Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application","ref_index":114,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09826","citing_title":"OmniGameArena: A Unified UE5 Benchmark for VLM Game Agents with Improvement Dynamics","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05896","citing_title":"Resonant Minds: Closed-Loop Social Avatars with Theory of Mind","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01224","citing_title":"AutoMem: Automated Learning of Memory as a Cognitive Skill","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24539","citing_title":"DemoEvolve: Overcoming Sparse Feedback in Agentic Harness Evolution with Demonstrations","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07021","citing_title":"Behavior Cue Reasoning: Monitorable Reasoning Improves Efficiency and Safety through Oversight","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17933","citing_title":"AtlasVA: Self-Evolving Visual Skill Memory for Teacher-Free VLM Agents","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2506.03610","citing_title":"Orak: A Foundational Benchmark for Training and Evaluating LLM Agents on Diverse Video Games","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2511.00739","citing_title":"Towards Understanding, Analyzing, and Optimizing Agentic AI Execution: A CPU-Centric Perspective","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13875","citing_title":"Common-agency Games for Multi-Objective Test-Time Alignment","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06869","citing_title":"Agentick: A Unified Benchmark for General Sequential Decision-Making Agents","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09965","citing_title":"Towards Generalist Game Players: An Investigation of Foundation Models in the Game Multiverse","ref_index":128,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11223","citing_title":"Do Vision-Language-Models show human-like logical problem-solving capability in point and click puzzle games?","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11223","citing_title":"Do Vision-Language-Models show human-like logical problem-solving capability in point and click puzzle games?","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09965","citing_title":"Towards Generalist Game Players: An Investigation of Foundation Models in the Game Multiverse","ref_index":128,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24558","citing_title":"Hierarchical Behaviour Spaces","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08340","citing_title":"PokeGym: A Visually-Driven Long-Horizon Benchmark for Vision-Language Models","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07021","citing_title":"Behavior Cue Reasoning: Monitorable Reasoning Improves Efficiency and Safety through Oversight","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06869","citing_title":"Agentick: A Unified Benchmark for General Sequential Decision-Making Agents","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20987","citing_title":"Co-Evolving LLM Decision and Skill Bank Agents for Long-Horizon Tasks","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/COB3EHCQNSZCGGJSNJOXVGSKKR","json":"https://pith.science/pith/COB3EHCQNSZCGGJSNJOXVGSKKR.json","graph_json":"https://pith.science/api/pith-number/COB3EHCQNSZCGGJSNJOXVGSKKR/graph.json","events_json":"https://pith.science/api/pith-number/COB3EHCQNSZCGGJSNJOXVGSKKR/events.json","paper":"https://pith.science/paper/COB3EHCQ"},"agent_actions":{"view_html":"https://pith.science/pith/COB3EHCQNSZCGGJSNJOXVGSKKR","download_json":"https://pith.science/pith/COB3EHCQNSZCGGJSNJOXVGSKKR.json","view_paper":"https://pith.science/paper/COB3EHCQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.13543&json=true","fetch_graph":"https://pith.science/api/pith-number/COB3EHCQNSZCGGJSNJOXVGSKKR/graph.json","fetch_events":"https://pith.science/api/pith-number/COB3EHCQNSZCGGJSNJOXVGSKKR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/COB3EHCQNSZCGGJSNJOXVGSKKR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/COB3EHCQNSZCGGJSNJOXVGSKKR/action/storage_attestation","attest_author":"https://pith.science/pith/COB3EHCQNSZCGGJSNJOXVGSKKR/action/author_attestation","sign_citation":"https://pith.science/pith/COB3EHCQNSZCGGJSNJOXVGSKKR/action/citation_signature","submit_replication":"https://pith.science/pith/COB3EHCQNSZCGGJSNJOXVGSKKR/action/replication_record"}},"created_at":"2026-07-05T10:42:39.280074+00:00","updated_at":"2026-07-05T10:42:39.280074+00:00"}