{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:TRUGLNEYCVTCWBGR5NFDZCCGZC","short_pith_number":"pith:TRUGLNEY","canonical_record":{"source":{"id":"2604.10825","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-12T21:37:26Z","cross_cats_sorted":[],"title_canon_sha256":"8f76097af32e619e69f6c0dad621447ae762ed1b9987f839b3732ab83cd5a33d","abstract_canon_sha256":"35e2ab0f93727970a58e0e2eca64eecb02bbba26445ccaf8c784cae24d4deca6"},"schema_version":"1.0"},"canonical_sha256":"9c6865b49815662b04d1eb4a3c8846c8a88940d2d5b183bd1745e0fc9bbe1c1c","source":{"kind":"arxiv","id":"2604.10825","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.10825","created_at":"2026-05-20T00:02:11Z"},{"alias_kind":"arxiv_version","alias_value":"2604.10825v2","created_at":"2026-05-20T00:02:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.10825","created_at":"2026-05-20T00:02:11Z"},{"alias_kind":"pith_short_12","alias_value":"TRUGLNEYCVTC","created_at":"2026-05-20T00:02:11Z"},{"alias_kind":"pith_short_16","alias_value":"TRUGLNEYCVTCWBGR","created_at":"2026-05-20T00:02:11Z"},{"alias_kind":"pith_short_8","alias_value":"TRUGLNEY","created_at":"2026-05-20T00:02:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:TRUGLNEYCVTCWBGR5NFDZCCGZC","target":"record","payload":{"canonical_record":{"source":{"id":"2604.10825","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-12T21:37:26Z","cross_cats_sorted":[],"title_canon_sha256":"8f76097af32e619e69f6c0dad621447ae762ed1b9987f839b3732ab83cd5a33d","abstract_canon_sha256":"35e2ab0f93727970a58e0e2eca64eecb02bbba26445ccaf8c784cae24d4deca6"},"schema_version":"1.0"},"canonical_sha256":"9c6865b49815662b04d1eb4a3c8846c8a88940d2d5b183bd1745e0fc9bbe1c1c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:02:11.384760Z","signature_b64":"zfVKHfnlzVQEhFiFPvLVuZWCNRm6kTGxYpX14bYI5/UuokyN5+eD9GTjmkJBXt/YBI/4F3LelI4ViC/puvTbAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c6865b49815662b04d1eb4a3c8846c8a88940d2d5b183bd1745e0fc9bbe1c1c","last_reissued_at":"2026-05-20T00:02:11.383748Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:02:11.383748Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.10825","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:02:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OgfDi9WVSAe3p6CMseyMv+gKwVQTSOt7RwGSiv9plh51/kfEVfCNfSBB0h4wMZUBgyG041oNlv50UG+cEabnDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T18:57:57.996804Z"},"content_sha256":"44e2f5eb93148a296d325ac6265d4ecd4fcad60a5bc9564222786d3b4417a67d","schema_version":"1.0","event_id":"sha256:44e2f5eb93148a296d325ac6265d4ecd4fcad60a5bc9564222786d3b4417a67d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:TRUGLNEYCVTCWBGR5NFDZCCGZC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CheeseBench: Evaluating Large Language Models on Rodent Behavioral Neuroscience Paradigms","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Open-weight LLMs reach only 53 percent success on ASCII versions of classic rodent tasks where animals average 79 percent.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Zacharie Bugaud","submitted_at":"2026-04-12T21:37:26Z","abstract_excerpt":"We introduce CheeseBench, a benchmark that evaluates large language models (LLMs) on nine classical behavioral neuroscience paradigms (Morris water maze, Barnes maze, T-maze, radial arm maze, star maze, operant chamber, shuttle box, conditioned place preference, and delayed non-match to sample), spanning six cognitive dimensions. Each task is grounded in peer-reviewed rodent protocols with approximate animal baselines. The agent receives a unified system prompt with no task-specific instructions and must discover goals purely from ASCII text observations and reward signals, much like a rodent "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Under this unified zero-shot ASCII protocol, current open-weight LLM agents remain well below approximate rodent reference values, particularly on tasks requiring spatial navigation and within-trial state tracking.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The ASCII text renderings of the tasks accurately capture the core cognitive and perceptual demands of the original rodent behavioral paradigms without introducing artifacts that alter difficulty.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"LLMs reach 52.6% average success on text-based rodent neuroscience tasks, above random agents at 32.1% but below approximate rodent baselines at 78.9%.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Open-weight LLMs reach only 53 percent success on ASCII versions of classic rodent tasks where animals average 79 percent.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"cf72cf3f28c4c7f6aebf5c99fa4759044dabc47f5a1727f61d4425940cedc3d7"},"source":{"id":"2604.10825","kind":"arxiv","version":2},"verdict":{"id":"cca5bbe6-e395-4aa8-ad0e-a20bbc9712bd","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T15:17:28.384818Z","strongest_claim":"Under this unified zero-shot ASCII protocol, current open-weight LLM agents remain well below approximate rodent reference values, particularly on tasks requiring spatial navigation and within-trial state tracking.","one_line_summary":"LLMs reach 52.6% average success on text-based rodent neuroscience tasks, above random agents at 32.1% but below approximate rodent baselines at 78.9%.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The ASCII text renderings of the tasks accurately capture the core cognitive and perceptual demands of the original rodent behavioral paradigms without introducing artifacts that alter difficulty.","pith_extraction_headline":"Open-weight LLMs reach only 53 percent success on ASCII versions of classic rodent tasks where animals average 79 percent."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.10825/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"},"verdict_id":"cca5bbe6-e395-4aa8-ad0e-a20bbc9712bd"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:02:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Fi0CV/uS5SxWAwSmvXYn3XQfn2wXYUsK4KpZc7uAFgw7fgWAGWHL///rSjyBxAZmlNeGjSy/kJY4lrsMBBAbAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T18:57:57.997793Z"},"content_sha256":"07bed6f11c5e61941bf2425db0e4a5a6ca3529b77a899ad31c2aee895ca09cee","schema_version":"1.0","event_id":"sha256:07bed6f11c5e61941bf2425db0e4a5a6ca3529b77a899ad31c2aee895ca09cee"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TRUGLNEYCVTCWBGR5NFDZCCGZC/bundle.json","state_url":"https://pith.science/pith/TRUGLNEYCVTCWBGR5NFDZCCGZC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TRUGLNEYCVTCWBGR5NFDZCCGZC/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-25T18:57:57Z","links":{"resolver":"https://pith.science/pith/TRUGLNEYCVTCWBGR5NFDZCCGZC","bundle":"https://pith.science/pith/TRUGLNEYCVTCWBGR5NFDZCCGZC/bundle.json","state":"https://pith.science/pith/TRUGLNEYCVTCWBGR5NFDZCCGZC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TRUGLNEYCVTCWBGR5NFDZCCGZC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TRUGLNEYCVTCWBGR5NFDZCCGZC","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"35e2ab0f93727970a58e0e2eca64eecb02bbba26445ccaf8c784cae24d4deca6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-12T21:37:26Z","title_canon_sha256":"8f76097af32e619e69f6c0dad621447ae762ed1b9987f839b3732ab83cd5a33d"},"schema_version":"1.0","source":{"id":"2604.10825","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.10825","created_at":"2026-05-20T00:02:11Z"},{"alias_kind":"arxiv_version","alias_value":"2604.10825v2","created_at":"2026-05-20T00:02:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.10825","created_at":"2026-05-20T00:02:11Z"},{"alias_kind":"pith_short_12","alias_value":"TRUGLNEYCVTC","created_at":"2026-05-20T00:02:11Z"},{"alias_kind":"pith_short_16","alias_value":"TRUGLNEYCVTCWBGR","created_at":"2026-05-20T00:02:11Z"},{"alias_kind":"pith_short_8","alias_value":"TRUGLNEY","created_at":"2026-05-20T00:02:11Z"}],"graph_snapshots":[{"event_id":"sha256:07bed6f11c5e61941bf2425db0e4a5a6ca3529b77a899ad31c2aee895ca09cee","target":"graph","created_at":"2026-05-20T00:02:11Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Under this unified zero-shot ASCII protocol, current open-weight LLM agents remain well below approximate rodent reference values, particularly on tasks requiring spatial navigation and within-trial state tracking."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The ASCII text renderings of the tasks accurately capture the core cognitive and perceptual demands of the original rodent behavioral paradigms without introducing artifacts that alter difficulty."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"LLMs reach 52.6% average success on text-based rodent neuroscience tasks, above random agents at 32.1% but below approximate rodent baselines at 78.9%."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Open-weight LLMs reach only 53 percent success on ASCII versions of classic rodent tasks where animals average 79 percent."}],"snapshot_sha256":"cf72cf3f28c4c7f6aebf5c99fa4759044dabc47f5a1727f61d4425940cedc3d7"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.10825/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce CheeseBench, a benchmark that evaluates large language models (LLMs) on nine classical behavioral neuroscience paradigms (Morris water maze, Barnes maze, T-maze, radial arm maze, star maze, operant chamber, shuttle box, conditioned place preference, and delayed non-match to sample), spanning six cognitive dimensions. Each task is grounded in peer-reviewed rodent protocols with approximate animal baselines. The agent receives a unified system prompt with no task-specific instructions and must discover goals purely from ASCII text observations and reward signals, much like a rodent ","authors_text":"Zacharie Bugaud","cross_cats":[],"headline":"Open-weight LLMs reach only 53 percent success on ASCII versions of classic rodent tasks where animals average 79 percent.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-12T21:37:26Z","title":"CheeseBench: Evaluating Large Language Models on Rodent Behavioral Neuroscience Paradigms"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.10825","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-10T15:17:28.384818Z","id":"cca5bbe6-e395-4aa8-ad0e-a20bbc9712bd","model_set":{"reader":"grok-4.3"},"one_line_summary":"LLMs reach 52.6% average success on text-based rodent neuroscience tasks, above random agents at 32.1% but below approximate rodent baselines at 78.9%.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Open-weight LLMs reach only 53 percent success on ASCII versions of classic rodent tasks where animals average 79 percent.","strongest_claim":"Under this unified zero-shot ASCII protocol, current open-weight LLM agents remain well below approximate rodent reference values, particularly on tasks requiring spatial navigation and within-trial state tracking.","weakest_assumption":"The ASCII text renderings of the tasks accurately capture the core cognitive and perceptual demands of the original rodent behavioral paradigms without introducing artifacts that alter difficulty."}},"verdict_id":"cca5bbe6-e395-4aa8-ad0e-a20bbc9712bd"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:44e2f5eb93148a296d325ac6265d4ecd4fcad60a5bc9564222786d3b4417a67d","target":"record","created_at":"2026-05-20T00:02:11Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"35e2ab0f93727970a58e0e2eca64eecb02bbba26445ccaf8c784cae24d4deca6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-12T21:37:26Z","title_canon_sha256":"8f76097af32e619e69f6c0dad621447ae762ed1b9987f839b3732ab83cd5a33d"},"schema_version":"1.0","source":{"id":"2604.10825","kind":"arxiv","version":2}},"canonical_sha256":"9c6865b49815662b04d1eb4a3c8846c8a88940d2d5b183bd1745e0fc9bbe1c1c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9c6865b49815662b04d1eb4a3c8846c8a88940d2d5b183bd1745e0fc9bbe1c1c","first_computed_at":"2026-05-20T00:02:11.383748Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:02:11.383748Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zfVKHfnlzVQEhFiFPvLVuZWCNRm6kTGxYpX14bYI5/UuokyN5+eD9GTjmkJBXt/YBI/4F3LelI4ViC/puvTbAQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:02:11.384760Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.10825","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:44e2f5eb93148a296d325ac6265d4ecd4fcad60a5bc9564222786d3b4417a67d","sha256:07bed6f11c5e61941bf2425db0e4a5a6ca3529b77a899ad31c2aee895ca09cee"],"state_sha256":"193e2274ce8bd1cf5c969acac24891314af95f64006e9856e869ec1cd8d6f4b4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SKKx5AzSvvdPPjMiZOAPkfSBH4kqR+sw25Xuj947+T5cEno8RuDZVCfRfH37BcNEpEMFnNMK7l3ERRY47BJ8DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T18:57:58.002136Z","bundle_sha256":"eba3737fd9d56d15265d5bf3952c8a0914cc068c7bd853166b39e86e155ce0a7"}}