{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:VEIZUCNTJ46U25G6RDE625H3DT","short_pith_number":"pith:VEIZUCNT","canonical_record":{"source":{"id":"2605.08692","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-09T04:59:21Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"8c57d93e58903ac57c202268320d23ed69423890eeba882f70edb9ade98ca08f","abstract_canon_sha256":"4a3c29bde5d4d0a34a4c96096abc0625327cebf361251f84738d5cfe881bdfec"},"schema_version":"1.0"},"canonical_sha256":"a9119a09b34f3d4d74de88c9ed74fb1cef72d5a54ddf3f2519f6886e4ff8f52a","source":{"kind":"arxiv","id":"2605.08692","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.08692","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"arxiv_version","alias_value":"2605.08692v2","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.08692","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_12","alias_value":"VEIZUCNTJ46U","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_16","alias_value":"VEIZUCNTJ46U25G6","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_8","alias_value":"VEIZUCNT","created_at":"2026-06-08T01:04:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:VEIZUCNTJ46U25G6RDE625H3DT","target":"record","payload":{"canonical_record":{"source":{"id":"2605.08692","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-09T04:59:21Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"8c57d93e58903ac57c202268320d23ed69423890eeba882f70edb9ade98ca08f","abstract_canon_sha256":"4a3c29bde5d4d0a34a4c96096abc0625327cebf361251f84738d5cfe881bdfec"},"schema_version":"1.0"},"canonical_sha256":"a9119a09b34f3d4d74de88c9ed74fb1cef72d5a54ddf3f2519f6886e4ff8f52a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-08T01:04:06.971618Z","signature_b64":"ru1GaRgfiJ+aLoqprdqJ8kHJ1tw39LjdbzbfLaLikU4SoKtEohvD481jkfc509q0xg3endfFjryVLyUyVlNvCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a9119a09b34f3d4d74de88c9ed74fb1cef72d5a54ddf3f2519f6886e4ff8f52a","last_reissued_at":"2026-06-08T01:04:06.970717Z","signature_status":"signed_v1","first_computed_at":"2026-06-08T01:04:06.970717Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.08692","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":"2026-06-08T01:04:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aYoFw5ldDeHh/iGrCu7uyjQV3K0BzMmCj+gWv389Dbx9UVZPglX3t4TErismAP5o80yt/9twn2VHo8tTNoTECA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-09T07:57:00.375700Z"},"content_sha256":"8f8e875d14fe9914cc01533c42f0b7c7226c0e81791a4fb1b54b42542f32f859","schema_version":"1.0","event_id":"sha256:8f8e875d14fe9914cc01533c42f0b7c7226c0e81791a4fb1b54b42542f32f859"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:VEIZUCNTJ46U25G6RDE625H3DT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AAAC: Activation-Aware Adaptive Codebooks for 4-bit LLM Weight Quantization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Two learned scalar codebooks per layer, chosen by activation-weighted error and encoded with no extra storage, let 4-bit LLM quantization finish in minutes while beating fixed-codebook baselines.","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Beshr IslamBouli, David Jin","submitted_at":"2026-05-09T04:59:21Z","abstract_excerpt":"Post-training weight-only quantization to 4 bits is widely used to reduce the memory and compute costs of large language model inference. Existing PTQ methods, such as AWQ and GPTQ, improve how weights are mapped onto a fixed 4-bit grid through scaling, clipping, or error compensation. To further improve accuracy, methods such as OmniQuant and QuIP\\# uses gradient-assisted algorithms at the cost of hours of quantization time. In this work, we propose AAAC (Activation-Aware Adaptive Codebooks), a lightweight method for 4-bit LLM weight quantization. AAAC replaces the fixed scalar codebook used "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"AAAC outperforms baselines at orders-of-magnitude less quantization time.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That two learned scalar codebooks per layer, selected by activation-weighted reconstruction error, are sufficient to capture the necessary weight distributions without introducing new quantization artifacts or requiring per-model hyperparameter search.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"AAAC uses two adaptive 64-byte codebooks per layer for 4-bit LLM weight quantization, choosing the optimal one per group to minimize activation-weighted error with zero storage overhead and fast runtime.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Two learned scalar codebooks per layer, chosen by activation-weighted error and encoded with no extra storage, let 4-bit LLM quantization finish in minutes while beating fixed-codebook baselines.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"2ae38dd8f290e9cab21cd3c7452633df3589a9fc8329a57f75a895f4b27a1487"},"source":{"id":"2605.08692","kind":"arxiv","version":2},"verdict":{"id":"aabe7f83-35e7-4ae7-9bce-b72fd7ac0fb8","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T01:03:07.602846Z","strongest_claim":"AAAC outperforms baselines at orders-of-magnitude less quantization time.","one_line_summary":"AAAC uses two adaptive 64-byte codebooks per layer for 4-bit LLM weight quantization, choosing the optimal one per group to minimize activation-weighted error with zero storage overhead and fast runtime.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That two learned scalar codebooks per layer, selected by activation-weighted reconstruction error, are sufficient to capture the necessary weight distributions without introducing new quantization artifacts or requiring per-model hyperparameter search.","pith_extraction_headline":"Two learned scalar codebooks per layer, chosen by activation-weighted error and encoded with no extra storage, let 4-bit LLM quantization finish in minutes while beating fixed-codebook baselines."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.08692/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T09:02:01.979298Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T22:36:05.031696Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T14:31:17.785425Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T10:51:59.895111Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"c784765d5e7b178addd1fb8309c94c969e3ed0a71a24f29990951f997b615be1"},"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":"aabe7f83-35e7-4ae7-9bce-b72fd7ac0fb8"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-08T01:04:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kj/4rMq8mpBzcXzops0bB4Rijl7GveuvuvfHMoMgP4Ry0LdoJKapevTO8JRUuX76CattdOOIiBb9fQBQlmBtAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-09T07:57:00.376524Z"},"content_sha256":"7d5959834b518f56045a26be792f26abe9444a400a9c00c7a1f6021fa59c0a18","schema_version":"1.0","event_id":"sha256:7d5959834b518f56045a26be792f26abe9444a400a9c00c7a1f6021fa59c0a18"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VEIZUCNTJ46U25G6RDE625H3DT/bundle.json","state_url":"https://pith.science/pith/VEIZUCNTJ46U25G6RDE625H3DT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VEIZUCNTJ46U25G6RDE625H3DT/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":"2026-06-09T07:57:00Z","links":{"resolver":"https://pith.science/pith/VEIZUCNTJ46U25G6RDE625H3DT","bundle":"https://pith.science/pith/VEIZUCNTJ46U25G6RDE625H3DT/bundle.json","state":"https://pith.science/pith/VEIZUCNTJ46U25G6RDE625H3DT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VEIZUCNTJ46U25G6RDE625H3DT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:VEIZUCNTJ46U25G6RDE625H3DT","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":"4a3c29bde5d4d0a34a4c96096abc0625327cebf361251f84738d5cfe881bdfec","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-09T04:59:21Z","title_canon_sha256":"8c57d93e58903ac57c202268320d23ed69423890eeba882f70edb9ade98ca08f"},"schema_version":"1.0","source":{"id":"2605.08692","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.08692","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"arxiv_version","alias_value":"2605.08692v2","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.08692","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_12","alias_value":"VEIZUCNTJ46U","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_16","alias_value":"VEIZUCNTJ46U25G6","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_8","alias_value":"VEIZUCNT","created_at":"2026-06-08T01:04:06Z"}],"graph_snapshots":[{"event_id":"sha256:7d5959834b518f56045a26be792f26abe9444a400a9c00c7a1f6021fa59c0a18","target":"graph","created_at":"2026-06-08T01:04:06Z","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":"AAAC outperforms baselines at orders-of-magnitude less quantization time."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That two learned scalar codebooks per layer, selected by activation-weighted reconstruction error, are sufficient to capture the necessary weight distributions without introducing new quantization artifacts or requiring per-model hyperparameter search."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"AAAC uses two adaptive 64-byte codebooks per layer for 4-bit LLM weight quantization, choosing the optimal one per group to minimize activation-weighted error with zero storage overhead and fast runtime."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Two learned scalar codebooks per layer, chosen by activation-weighted error and encoded with no extra storage, let 4-bit LLM quantization finish in minutes while beating fixed-codebook baselines."}],"snapshot_sha256":"2ae38dd8f290e9cab21cd3c7452633df3589a9fc8329a57f75a895f4b27a1487"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T09:02:01.979298Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T22:36:05.031696Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T14:31:17.785425Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T10:51:59.895111Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.08692/integrity.json","findings":[],"snapshot_sha256":"c784765d5e7b178addd1fb8309c94c969e3ed0a71a24f29990951f997b615be1","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Post-training weight-only quantization to 4 bits is widely used to reduce the memory and compute costs of large language model inference. Existing PTQ methods, such as AWQ and GPTQ, improve how weights are mapped onto a fixed 4-bit grid through scaling, clipping, or error compensation. To further improve accuracy, methods such as OmniQuant and QuIP\\# uses gradient-assisted algorithms at the cost of hours of quantization time. In this work, we propose AAAC (Activation-Aware Adaptive Codebooks), a lightweight method for 4-bit LLM weight quantization. AAAC replaces the fixed scalar codebook used ","authors_text":"Beshr IslamBouli, David Jin","cross_cats":["cs.CL"],"headline":"Two learned scalar codebooks per layer, chosen by activation-weighted error and encoded with no extra storage, let 4-bit LLM quantization finish in minutes while beating fixed-codebook baselines.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-09T04:59:21Z","title":"AAAC: Activation-Aware Adaptive Codebooks for 4-bit LLM Weight Quantization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.08692","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-12T01:03:07.602846Z","id":"aabe7f83-35e7-4ae7-9bce-b72fd7ac0fb8","model_set":{"reader":"grok-4.3"},"one_line_summary":"AAAC uses two adaptive 64-byte codebooks per layer for 4-bit LLM weight quantization, choosing the optimal one per group to minimize activation-weighted error with zero storage overhead and fast runtime.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Two learned scalar codebooks per layer, chosen by activation-weighted error and encoded with no extra storage, let 4-bit LLM quantization finish in minutes while beating fixed-codebook baselines.","strongest_claim":"AAAC outperforms baselines at orders-of-magnitude less quantization time.","weakest_assumption":"That two learned scalar codebooks per layer, selected by activation-weighted reconstruction error, are sufficient to capture the necessary weight distributions without introducing new quantization artifacts or requiring per-model hyperparameter search."}},"verdict_id":"aabe7f83-35e7-4ae7-9bce-b72fd7ac0fb8"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8f8e875d14fe9914cc01533c42f0b7c7226c0e81791a4fb1b54b42542f32f859","target":"record","created_at":"2026-06-08T01:04:06Z","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":"4a3c29bde5d4d0a34a4c96096abc0625327cebf361251f84738d5cfe881bdfec","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-09T04:59:21Z","title_canon_sha256":"8c57d93e58903ac57c202268320d23ed69423890eeba882f70edb9ade98ca08f"},"schema_version":"1.0","source":{"id":"2605.08692","kind":"arxiv","version":2}},"canonical_sha256":"a9119a09b34f3d4d74de88c9ed74fb1cef72d5a54ddf3f2519f6886e4ff8f52a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a9119a09b34f3d4d74de88c9ed74fb1cef72d5a54ddf3f2519f6886e4ff8f52a","first_computed_at":"2026-06-08T01:04:06.970717Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-08T01:04:06.970717Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ru1GaRgfiJ+aLoqprdqJ8kHJ1tw39LjdbzbfLaLikU4SoKtEohvD481jkfc509q0xg3endfFjryVLyUyVlNvCw==","signature_status":"signed_v1","signed_at":"2026-06-08T01:04:06.971618Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.08692","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8f8e875d14fe9914cc01533c42f0b7c7226c0e81791a4fb1b54b42542f32f859","sha256:7d5959834b518f56045a26be792f26abe9444a400a9c00c7a1f6021fa59c0a18"],"state_sha256":"1b9306142256a8f15372948d5bc8a2dd1667694b835a4e5e0598425ce8049445"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lm7m2YZ053v28yo4r7yBEq0U09cMGmPkUViQ4e7isJlXHJwsg7U9Ka6sMTXBV1u4xIDRjTsD1s8+HxB0b5HlCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-09T07:57:00.380083Z","bundle_sha256":"62954d5e07005f07bad3faec0f678954a4c395db6a6fedc2a575c6a1116b54c5"}}