{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/evaluating-the-world-model-implicit-in-a","title":"Evaluating the World Model Implicit in a Generative Model","arxiv_id":"2406.03689","date":"2024-06-06","proceeding":null,"authors":["Keyon Vafa","Justin Y. Chen","Ashesh Rambachan","Jon Kleinberg","Sendhil Mullainathan"],"abstract":"Recent work suggests that large language models may implicitly learn world models. How should we assess this possibility? We formalize this question for the case where the underlying reality is governed by a deterministic finite automaton. This includes problems as diverse as simple logical reasoning, geographic navigation, game-playing, and chemistry. We propose new evaluation metrics for world model recovery inspired by the classic Myhill-Nerode theorem from language theory. We illustrate their utility in three domains: game playing, logic puzzles, and navigation. In all domains, the generative models we consider do well on existing diagnostics for assessing world models, but our evaluation metrics reveal their world models to be far less coherent than they appear. Such incoherence creates fragility: using a generative model to solve related but subtly different tasks can lead to failures. Building generative models that meaningfully capture the underlying logic of the domains they model would be immensely valuable; our results suggest new ways to assess how close a given model is to that goal.","url_abs":"https://arxiv.org/abs/2406.03689v3","url_pdf":"https://arxiv.org/pdf/2406.03689v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"evaluating-the-world-model-implicit-in-a","repo_url":"https://github.com/keyonvafa/world-model-evaluation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"logical-reasoning","task_name":"Logical Reasoning"},{"task_slug":"model","task_name":"model"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2406.03689","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2406.03689"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/keyonvafa/world-model-evaluation","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":6},"by_repo_kind":{"official":{"samples":6,"ran":6,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"441ba8cbd04d2d26","entry":"collate_fn","repo":"keyonvafa/world-model-evaluation","repo_kind":"official","path":"model.py","file_url":"https://github.com/keyonvafa/world-model-evaluation/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"441ba8cbd04d2d26"}},{"code_sha256_prefix":"1a0f3a6c603d52ed","entry":"permit","repo":"keyonvafa/world-model-evaluation","repo_kind":"official","path":"othello_world/mechanistic_interpretability/mech_interp_othello_utils.py","file_url":"https://github.com/keyonvafa/world-model-evaluation/blob/HEAD/othello_world/mechanistic_interpretability/mech_interp_othello_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1a0f3a6c603d52ed"}},{"code_sha256_prefix":"1484fe530bc561e5","entry":"permit_reverse","repo":"keyonvafa/world-model-evaluation","repo_kind":"official","path":"othello_world/mechanistic_interpretability/mech_interp_othello_utils.py","file_url":"https://github.com/keyonvafa/world-model-evaluation/blob/HEAD/othello_world/mechanistic_interpretability/mech_interp_othello_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1484fe530bc561e5"}},{"code_sha256_prefix":"03fcfa1178193c7f","entry":"sample_game_from_board","repo":"keyonvafa/world-model-evaluation","repo_kind":"official","path":"othello_world/utils.py","file_url":"https://github.com/keyonvafa/world-model-evaluation/blob/HEAD/othello_world/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"03fcfa1178193c7f"}},{"code_sha256_prefix":"371f505cf3526d45","entry":"str_to_int","repo":"keyonvafa/world-model-evaluation","repo_kind":"official","path":"othello_world/mechanistic_interpretability/tl_initial_exploration.py","file_url":"https://github.com/keyonvafa/world-model-evaluation/blob/HEAD/othello_world/mechanistic_interpretability/tl_initial_exploration.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"371f505cf3526d45"}},{"code_sha256_prefix":"0a2c0f55c57d9f48","entry":"to_board_label","repo":"keyonvafa/world-model-evaluation","repo_kind":"official","path":"othello_world/mechanistic_interpretability/mech_interp_othello_utils.py","file_url":"https://github.com/keyonvafa/world-model-evaluation/blob/HEAD/othello_world/mechanistic_interpretability/mech_interp_othello_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0a2c0f55c57d9f48"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}