{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:XYKCYTCWV7ZWQ464H6KKTGGY4I","short_pith_number":"pith:XYKCYTCW","canonical_record":{"source":{"id":"2605.18607","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-18T16:17:15Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9e42e53ba89737a3b6a63687b18e02f6cb21be5d6cc5e14b377ac05ed5195c78","abstract_canon_sha256":"1ee2220e1b835e32c44fe87fde6e49375e91105350c28a25d87dfddc6f000882"},"schema_version":"1.0"},"canonical_sha256":"be142c4c56aff36873dc3f94a998d8e21c356951dd9134dd5b3434322094790e","source":{"kind":"arxiv","id":"2605.18607","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.18607","created_at":"2026-05-20T00:06:10Z"},{"alias_kind":"arxiv_version","alias_value":"2605.18607v1","created_at":"2026-05-20T00:06:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18607","created_at":"2026-05-20T00:06:10Z"},{"alias_kind":"pith_short_12","alias_value":"XYKCYTCWV7ZW","created_at":"2026-05-20T00:06:10Z"},{"alias_kind":"pith_short_16","alias_value":"XYKCYTCWV7ZWQ464","created_at":"2026-05-20T00:06:10Z"},{"alias_kind":"pith_short_8","alias_value":"XYKCYTCW","created_at":"2026-05-20T00:06:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:XYKCYTCWV7ZWQ464H6KKTGGY4I","target":"record","payload":{"canonical_record":{"source":{"id":"2605.18607","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-18T16:17:15Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9e42e53ba89737a3b6a63687b18e02f6cb21be5d6cc5e14b377ac05ed5195c78","abstract_canon_sha256":"1ee2220e1b835e32c44fe87fde6e49375e91105350c28a25d87dfddc6f000882"},"schema_version":"1.0"},"canonical_sha256":"be142c4c56aff36873dc3f94a998d8e21c356951dd9134dd5b3434322094790e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:06:10.624548Z","signature_b64":"evdGQlLUgSVNjubhpCuofg/QAwL0U/jpVHiydjAk/Idl2crgpMnZbnVZRXh6FFKQHT02+MpYFniyAzVdlg9hDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be142c4c56aff36873dc3f94a998d8e21c356951dd9134dd5b3434322094790e","last_reissued_at":"2026-05-20T00:06:10.623661Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:06:10.623661Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.18607","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:06:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"i+KxRBfMgbv+g1QvK3sPeKO6EXP1huYVKEl2/051J/iHw7fe3dW/keBpp+lozr8NBfIyIOzE5keVr0Jd7NxzCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T07:12:25.694120Z"},"content_sha256":"33d284b06cbfb2fe8d2f0fe868e5ad31c95a35e6e493f258d770291865844bd7","schema_version":"1.0","event_id":"sha256:33d284b06cbfb2fe8d2f0fe868e5ad31c95a35e6e493f258d770291865844bd7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:XYKCYTCWV7ZWQ464H6KKTGGY4I","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Forecasting Downstream Performance of LLMs With Proxy Metrics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Arkil Patel, Dzmitry Bahdanau, Marius Mosbach, Siva Reddy","submitted_at":"2026-05-18T16:17:15Z","abstract_excerpt":"Progress in language model development is often driven by comparative decisions: which architecture to adopt, which pretraining corpus to use, or which training recipe to apply. Making these decisions well requires reliable performance forecasts, yet the two commonly used signals are fundamentally limited. Cross-entropy loss is poorly aligned with downstream capabilities, and direct downstream evaluation is expensive, sparse, and often uninformative at early training stages. Instead, we propose to construct proxy metrics by aggregating token-level statistics, such as entropy, top-k accuracy, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18607","kind":"arxiv","version":1},"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/2605.18607/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T00:01:59.243642Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"5145f996d8a7088c1abe8dfd88b63c61268775d3c3abd99dce9392b687660da9"},"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":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:06:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AEcx3rHsC6YCQVyyLqHp58qZDCAEFEIA2BXs939iWGxpo71jBZ5yHWNM+FiCwQ9upYlUKIo8fnboQjtjv33gCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T07:12:25.694850Z"},"content_sha256":"b62b7c50af21b3f4bbc86aba793d5857b9cd182d0bf10325e4c52c79159b98a8","schema_version":"1.0","event_id":"sha256:b62b7c50af21b3f4bbc86aba793d5857b9cd182d0bf10325e4c52c79159b98a8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XYKCYTCWV7ZWQ464H6KKTGGY4I/bundle.json","state_url":"https://pith.science/pith/XYKCYTCWV7ZWQ464H6KKTGGY4I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XYKCYTCWV7ZWQ464H6KKTGGY4I/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":"2026-05-26T07:12:25Z","links":{"resolver":"https://pith.science/pith/XYKCYTCWV7ZWQ464H6KKTGGY4I","bundle":"https://pith.science/pith/XYKCYTCWV7ZWQ464H6KKTGGY4I/bundle.json","state":"https://pith.science/pith/XYKCYTCWV7ZWQ464H6KKTGGY4I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XYKCYTCWV7ZWQ464H6KKTGGY4I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:XYKCYTCWV7ZWQ464H6KKTGGY4I","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":"1ee2220e1b835e32c44fe87fde6e49375e91105350c28a25d87dfddc6f000882","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-18T16:17:15Z","title_canon_sha256":"9e42e53ba89737a3b6a63687b18e02f6cb21be5d6cc5e14b377ac05ed5195c78"},"schema_version":"1.0","source":{"id":"2605.18607","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.18607","created_at":"2026-05-20T00:06:10Z"},{"alias_kind":"arxiv_version","alias_value":"2605.18607v1","created_at":"2026-05-20T00:06:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18607","created_at":"2026-05-20T00:06:10Z"},{"alias_kind":"pith_short_12","alias_value":"XYKCYTCWV7ZW","created_at":"2026-05-20T00:06:10Z"},{"alias_kind":"pith_short_16","alias_value":"XYKCYTCWV7ZWQ464","created_at":"2026-05-20T00:06:10Z"},{"alias_kind":"pith_short_8","alias_value":"XYKCYTCW","created_at":"2026-05-20T00:06:10Z"}],"graph_snapshots":[{"event_id":"sha256:b62b7c50af21b3f4bbc86aba793d5857b9cd182d0bf10325e4c52c79159b98a8","target":"graph","created_at":"2026-05-20T00:06:10Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T00:01:59.243642Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.18607/integrity.json","findings":[],"snapshot_sha256":"5145f996d8a7088c1abe8dfd88b63c61268775d3c3abd99dce9392b687660da9","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Progress in language model development is often driven by comparative decisions: which architecture to adopt, which pretraining corpus to use, or which training recipe to apply. Making these decisions well requires reliable performance forecasts, yet the two commonly used signals are fundamentally limited. Cross-entropy loss is poorly aligned with downstream capabilities, and direct downstream evaluation is expensive, sparse, and often uninformative at early training stages. Instead, we propose to construct proxy metrics by aggregating token-level statistics, such as entropy, top-k accuracy, a","authors_text":"Arkil Patel, Dzmitry Bahdanau, Marius Mosbach, Siva Reddy","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-18T16:17:15Z","title":"Forecasting Downstream Performance of LLMs With Proxy Metrics"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18607","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:33d284b06cbfb2fe8d2f0fe868e5ad31c95a35e6e493f258d770291865844bd7","target":"record","created_at":"2026-05-20T00:06:10Z","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":"1ee2220e1b835e32c44fe87fde6e49375e91105350c28a25d87dfddc6f000882","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-18T16:17:15Z","title_canon_sha256":"9e42e53ba89737a3b6a63687b18e02f6cb21be5d6cc5e14b377ac05ed5195c78"},"schema_version":"1.0","source":{"id":"2605.18607","kind":"arxiv","version":1}},"canonical_sha256":"be142c4c56aff36873dc3f94a998d8e21c356951dd9134dd5b3434322094790e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"be142c4c56aff36873dc3f94a998d8e21c356951dd9134dd5b3434322094790e","first_computed_at":"2026-05-20T00:06:10.623661Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:06:10.623661Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"evdGQlLUgSVNjubhpCuofg/QAwL0U/jpVHiydjAk/Idl2crgpMnZbnVZRXh6FFKQHT02+MpYFniyAzVdlg9hDA==","signature_status":"signed_v1","signed_at":"2026-05-20T00:06:10.624548Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.18607","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:33d284b06cbfb2fe8d2f0fe868e5ad31c95a35e6e493f258d770291865844bd7","sha256:b62b7c50af21b3f4bbc86aba793d5857b9cd182d0bf10325e4c52c79159b98a8"],"state_sha256":"4c05c0dd87dcbb972282ce0b3293390f920223e1144189752a13c16bf58104c1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gp1/x5VyAkUBGkJg28Kwe1X9hD8G+1ru0Wi9YLfFoozTQGOWvw2znRYZvo0u/xnKOaA8N/Oh/hskEX/ytocxDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T07:12:25.698635Z","bundle_sha256":"0ee5ff8b9e0fec4dadb4f80125d8a3091ba684bc52f35b6cc9d99c3717d44f0e"}}