{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:KGRULLYAMHOJIAQOMAO6YNPWSP","short_pith_number":"pith:KGRULLYA","canonical_record":{"source":{"id":"2608.13426","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-08-13T16:16:04Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"f6a92ebdf8cb3c9e2f3b0cf0a6e428f43045720116fc8a03f2f117103369a301","abstract_canon_sha256":"452806c7a7d5c5fee8b972171a67656cebc3aa0eee2448a605d2da3f59d228fb"},"schema_version":"1.0"},"canonical_sha256":"51a345af0061dc94020e601dec35f693cca38f03f2ea5cc670f66054d63d2e7b","source":{"kind":"arxiv","id":"2608.13426","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.13426","created_at":"2026-08-14T01:04:25Z"},{"alias_kind":"arxiv_version","alias_value":"2608.13426v1","created_at":"2026-08-14T01:04:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.13426","created_at":"2026-08-14T01:04:25Z"},{"alias_kind":"pith_short_12","alias_value":"KGRULLYAMHOJ","created_at":"2026-08-14T01:04:25Z"},{"alias_kind":"pith_short_16","alias_value":"KGRULLYAMHOJIAQO","created_at":"2026-08-14T01:04:25Z"},{"alias_kind":"pith_short_8","alias_value":"KGRULLYA","created_at":"2026-08-14T01:04:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:KGRULLYAMHOJIAQOMAO6YNPWSP","target":"record","payload":{"canonical_record":{"source":{"id":"2608.13426","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-08-13T16:16:04Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"f6a92ebdf8cb3c9e2f3b0cf0a6e428f43045720116fc8a03f2f117103369a301","abstract_canon_sha256":"452806c7a7d5c5fee8b972171a67656cebc3aa0eee2448a605d2da3f59d228fb"},"schema_version":"1.0"},"canonical_sha256":"51a345af0061dc94020e601dec35f693cca38f03f2ea5cc670f66054d63d2e7b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-14T01:04:25.110156Z","signature_b64":"jyy/ALNsgUlbRkNKbLqurN7dH7UcHXZOh3Ez8wUKWHUwrXf9vDQ30eA8MIOJjXQblQhDT+RW1fE314CkwTo8BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"51a345af0061dc94020e601dec35f693cca38f03f2ea5cc670f66054d63d2e7b","last_reissued_at":"2026-08-14T01:04:25.108478Z","signature_status":"signed_v1","first_computed_at":"2026-08-14T01:04:25.108478Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.13426","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":"2026-08-14T01:04:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wNEeQ2JfkksC0R8lD+PPgv+/+G0D6qTaMkheKdEpJzKy+UugY4TeriVw+aAFuUxgAUJbokqoZMbBk05l6tYjCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T12:21:57.781806Z"},"content_sha256":"5d246a62725d6c6f28942bcb9b4fcf12d7e9c2d383fb58766aed2c7196828c87","schema_version":"1.0","event_id":"sha256:5d246a62725d6c6f28942bcb9b4fcf12d7e9c2d383fb58766aed2c7196828c87"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:KGRULLYAMHOJIAQOMAO6YNPWSP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reduced Matrix Multiplication: Input-Adaptive Matrix-Product Reduction for LLM Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jiawei Zhou, Yanhong Li, Zixuan Lan","submitted_at":"2026-08-13T16:16:04Z","abstract_excerpt":"Transformer-based language models achieve strong performance but incur substantial inference cost due to repeated high-dimensional matrix multiplications. We propose Reduced Matrix Multiplication (RMM), a training-free, input-adaptive inference method that reduces Transformer matrix products by selecting informative slices along their contraction dimensions, without modifying model weights. Under a simple retention-ratio control, RMM provides a smooth and predictable accuracy-efficiency trade-off. Across language models ranging from 1B to 70B parameters, we find that reduction tolerance depend"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.13426","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/2608.13426/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":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-14T01:04:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GD9VGY1AYx1nyIvUwHwMOFjNI1Tpxnrffr/InNrMIxBE4GKpaAZQBcdo84t5relu6/lcilNNkPBcgPu07LSxDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T12:21:57.782616Z"},"content_sha256":"389d62aa535644f3b2e7190426b472ddf8010ddf3c1b16ceb7e5353850946d78","schema_version":"1.0","event_id":"sha256:389d62aa535644f3b2e7190426b472ddf8010ddf3c1b16ceb7e5353850946d78"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KGRULLYAMHOJIAQOMAO6YNPWSP/bundle.json","state_url":"https://pith.science/pith/KGRULLYAMHOJIAQOMAO6YNPWSP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KGRULLYAMHOJIAQOMAO6YNPWSP/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":"2026-08-17T12:21:57Z","links":{"resolver":"https://pith.science/pith/KGRULLYAMHOJIAQOMAO6YNPWSP","bundle":"https://pith.science/pith/KGRULLYAMHOJIAQOMAO6YNPWSP/bundle.json","state":"https://pith.science/pith/KGRULLYAMHOJIAQOMAO6YNPWSP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KGRULLYAMHOJIAQOMAO6YNPWSP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:KGRULLYAMHOJIAQOMAO6YNPWSP","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":"452806c7a7d5c5fee8b972171a67656cebc3aa0eee2448a605d2da3f59d228fb","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-08-13T16:16:04Z","title_canon_sha256":"f6a92ebdf8cb3c9e2f3b0cf0a6e428f43045720116fc8a03f2f117103369a301"},"schema_version":"1.0","source":{"id":"2608.13426","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.13426","created_at":"2026-08-14T01:04:25Z"},{"alias_kind":"arxiv_version","alias_value":"2608.13426v1","created_at":"2026-08-14T01:04:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.13426","created_at":"2026-08-14T01:04:25Z"},{"alias_kind":"pith_short_12","alias_value":"KGRULLYAMHOJ","created_at":"2026-08-14T01:04:25Z"},{"alias_kind":"pith_short_16","alias_value":"KGRULLYAMHOJIAQO","created_at":"2026-08-14T01:04:25Z"},{"alias_kind":"pith_short_8","alias_value":"KGRULLYA","created_at":"2026-08-14T01:04:25Z"}],"graph_snapshots":[{"event_id":"sha256:389d62aa535644f3b2e7190426b472ddf8010ddf3c1b16ceb7e5353850946d78","target":"graph","created_at":"2026-08-14T01:04:25Z","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":[],"endpoint":"/pith/2608.13426/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Transformer-based language models achieve strong performance but incur substantial inference cost due to repeated high-dimensional matrix multiplications. We propose Reduced Matrix Multiplication (RMM), a training-free, input-adaptive inference method that reduces Transformer matrix products by selecting informative slices along their contraction dimensions, without modifying model weights. Under a simple retention-ratio control, RMM provides a smooth and predictable accuracy-efficiency trade-off. Across language models ranging from 1B to 70B parameters, we find that reduction tolerance depend","authors_text":"Jiawei Zhou, Yanhong Li, Zixuan Lan","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-08-13T16:16:04Z","title":"Reduced Matrix Multiplication: Input-Adaptive Matrix-Product Reduction for LLM Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.13426","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":"sha256:5d246a62725d6c6f28942bcb9b4fcf12d7e9c2d383fb58766aed2c7196828c87","target":"record","created_at":"2026-08-14T01:04:25Z","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":"452806c7a7d5c5fee8b972171a67656cebc3aa0eee2448a605d2da3f59d228fb","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-08-13T16:16:04Z","title_canon_sha256":"f6a92ebdf8cb3c9e2f3b0cf0a6e428f43045720116fc8a03f2f117103369a301"},"schema_version":"1.0","source":{"id":"2608.13426","kind":"arxiv","version":1}},"canonical_sha256":"51a345af0061dc94020e601dec35f693cca38f03f2ea5cc670f66054d63d2e7b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"51a345af0061dc94020e601dec35f693cca38f03f2ea5cc670f66054d63d2e7b","first_computed_at":"2026-08-14T01:04:25.108478Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-14T01:04:25.108478Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"jyy/ALNsgUlbRkNKbLqurN7dH7UcHXZOh3Ez8wUKWHUwrXf9vDQ30eA8MIOJjXQblQhDT+RW1fE314CkwTo8BA==","signature_status":"signed_v1","signed_at":"2026-08-14T01:04:25.110156Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.13426","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5d246a62725d6c6f28942bcb9b4fcf12d7e9c2d383fb58766aed2c7196828c87","sha256:389d62aa535644f3b2e7190426b472ddf8010ddf3c1b16ceb7e5353850946d78"],"state_sha256":"3292a675ea8457f1e4485e2bc0baf0a0b882e8c28db4537abacebad129e86783"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3e9dHt3340XNGY0zrUMX2ZXbhYZloOAelRiblG75AF+H3YiAT4kGrbof2xY4vj0j9jUJ9il09Qw6JzB8RhHNCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T12:21:57.789310Z","bundle_sha256":"9c269e73cad4ad79cfa62a864ed9279fccff14b1fce52a1f5dab1c5be064db8c"}}