{"_canonicalization":{"envelope_id":"axm_ + sha256(envelope minus {signature, axiom_id, anchors})","envelope_signature":"ed25519(envelope minus {signature, axiom_id})","json":"sort_keys=True, separators=(',',':'), ensure_ascii=False, allow_nan=False, utf-8","leaf_hash":"sha256(0x00 || canonical_json(envelope_full))","seal_signature":"ed25519(seal minus {signature, sig_algorithm})"},"axiom_id":"axm_b59c0338595fd0ef3be6c1fc0534e0ce6df3941c2e221c9e765ed84b3b8a0011","bitcoin_anchor":{"bitcoin_attestations":[],"calendar_attestations":[],"ots_url":"","stamped_at":"","status":"pending_next_stamp"},"envelope":{"anchors":[{"chain":"crovia.axiom_graph","height":0,"merkle_proof":"spider_vendor_press_v1","root_at_anchor":"spider_vendor_press_v1"}],"axiom_id":"axm_b59c0338595fd0ef3be6c1fc0534e0ce6df3941c2e221c9e765ed84b3b8a0011","axiom_type":"AX.OBS","body":{"axiom_subtype":"news.vendor_press.v1","category":"news","fingerprint":"19af7e08084221ddd0bc2b288199d7e2aeff3628e7a27c98b01efe8b94827f63","published":"Thu, 16 Jul 2026 00:00:00 -0400","receipt_hash":"19af7e08084221ddd0bc2b288199d7e2aeff3628e7a27c98b01efe8b94827f63","schema":"spider.news.vendor_press.v1","spider":"vendor_press","spider_record":{"axiom_subtype":"news.vendor_press.v1","category":"news","decision_hint":"POSITIVE","envelope_target":"AX.OBS","fingerprint":"19af7e08084221ddd0bc2b288199d7e2aeff3628e7a27c98b01efe8b94827f63","observed_at":"2026-07-16T04:44:04.234592Z","parent_run_hash":"392cd8ef881a5a54dbb7fbd0e0c490e4811d7ed32379244b3196bd5cb9c5d632","published":"Thu, 16 Jul 2026 00:00:00 -0400","runtime_version":"0.1.0","schema":"spider.news.vendor_press.v1","source_status":200,"source_url":"https://export.arxiv.org/rss/cs.AI","spider":"vendor_press","summary_excerpt":"arXiv:2505.18231v3 Announce Type: replace-cross \nAbstract: Large Language Model (LLM) inference is typically memory-intensive, especially when processing large batch sizes and long sequences, due to the large size of key-value (KV) cache. Vector Quantization (VQ) is recently adopted to alleviate this issue, but we find that the existing approach is susceptible to distribution shift due to its reliance on calibration datasets. To address this limitation, we introduce NSNQuant, a calibration-free Vector Quantization (VQ) technique designed for low-bit compression of the KV cache. By applying a three-step transformation-1) a token-wise normalization (Normalize), 2) a channel-wise centering (Shift), and 3) a second token-wise normalization (Normalize)-with Hadamard transform, NSNQuant effectively aligns the token distribution with the standard normal distribution. This alignment enables robust, calibration-free vector quantization using a single reusable codebook. Extensive experiments sho","title":"NSNQuant: A Double Normalization Approach for Calibration-Free Low-Bit Vector Quantization of KV Cache","url":"https://arxiv.org/abs/2505.18231","vendor":"arxiv_cs_ai"},"summary":"arXiv:2505.18231v3 Announce Type: replace-cross \nAbstract: Large Language Model (LLM) inference is typically memory-intensive, especially when processing large batch sizes and long sequences, due to the large size of key-value (KV) cache. Vector Quantization (VQ) is recently adopted to alleviate this issue, but we find that the existing approach is susceptible to distribution shift due to its reliance on calibration datasets. To address this limitation, we introduce NSNQuant, a calibration-free Vector Quantization (VQ) technique designed for low-bit compression of the KV cache. By applying a three-step transformation-1) a token-wise normalization (Normalize), 2) a channel-wise centering (Shift), and 3) a second token-wise normalization (Normalize)-with Hadamard transform, NSNQuant effectively aligns the token distribution with the standard normal distribution. This alignment enables robust, calibration-free vector quantization using a single reusable codebook. Extensive experiments sho","title":"NSNQuant: A Double Normalization Approach for Calibration-Free Low-Bit Vector Quantization of KV Cache","vendor":"arxiv_cs_ai"},"confidence":{"method":"deterministic"},"decision":"POSITIVE","issued_at":"2026-07-16T04:44:04Z","notes":"Spider vendor_press (news) news.vendor_press.v1","object":{"captured_by":"crovia.spider.vendor_press","primary_source_url":"https://arxiv.org/abs/2505.18231"},"predecessors":[],"schema":"crovia.axiom.v1","signature":"ed25519:467d9f03d244686bc1558ed0bfbb6bf70c446f3c2fc8b93566924e4a6e6811c4f1016270c073e8021912c036c5657efe5d30b621178cc259057d5770ae7a9b08","signer":"crovia.substrate","subject":{"observed_at":"2026-07-16T04:44:04Z","source_collector":"spider:vendor_press","target_id":"https://arxiv.org/abs/2505.18231"},"tsa":{"authority":"crovia.substrate.bootstrap","rfc3161_token":"{\"kind\":\"crovia.bootstrap.tsa\",\"source_jsonl\":\"/opt/crovia/spider/data/news/vendor_press_v1.jsonl\",\"source_seal_merkle_root\":\"spider_vendor_press_v1\",\"upgrade_path\":\"Sessione H \\u2014 OpenTimestamps weekly anchor\"}"},"zk_mode":"clear","zk_proof":null},"ledger":{"leaf_hash":"f9ca2a78b83650c10ef88d6769a8efae4d960a080a5fb2a6c289ef189a4b12fb","leaf_index":319903,"ledger_path":"/opt/crovia/substrate/axiom_ledger.jsonl"},"merkle_proof":{"hash_alg":"sha256","leaf_prefix":"0x00","node_prefix":"0x01","odd_leaf_rule":"duplicate_last","path":[{"sibling":"5b918c3bb3b9c01e9d70d84922cc8537f7b6c65f87af244f55834df7b0f1b08a","side":"left"},{"sibling":"5ac1183529ecc8460b0b79d848cb43acbec56e2d7263ba3598cdd124e141a57b","side":"left"},{"sibling":"cabd5589d7c4b8ac8ebec1ae299af69b270ab675e4bc3ac6c3539e2efefb42a4","side":"left"},{"sibling":"3b643d27add0b4f72a8061f92b33310ac626d49ac14f684d6eb3494436fb2d7b","side":"left"},{"sibling":"83b44dfbeff4896d1c3548ce16bb68989434aa76eb6e74963505811bc3fc8951","side":"left"},{"sibling":"62e3b394632b6a6c803431d7367cf3239ed0a656a71708a95c3697a979e35cdb","side":"right"},{"sibling":"ee301003f62e32cb8340b9b36c3b96d6f2fbb9843fae27ab12f5db78a700fd40","side":"right"},{"sibling":"07cc977fce7e64af6ef759252372209956d49c5a495213666eef01b40c0f6d6f","side":"left"},{"sibling":"929619d7bb5abd9d4c015f749565f8d43a371ed382bafe122457d67edb9b8323","side":"left"},{"sibling":"1290775fa2a1fe2079ed83a9c60b8cb479714b6312967daa62f6a91dbbe1cac9","side":"right"},{"sibling":"95977bf2fb44d423026e874e6c275f78b1cf48666d2c472c5e4603d16f4faf7f","side":"right"},{"sibling":"6b83879fdb76b5b7270071f1edfe4d1a059fa9b92653e70fc8c8893214b8f237","side":"right"},{"sibling":"ff49c1d8749b258f58cc33cfaf20d723098ed6860fb62a6779db960d6250915b","side":"right"},{"sibling":"34d85f6ad6cc7dfa79d90e2b9ff99a561bcdc75b0301bbbd3e83861f54535c1e","side":"left"},{"sibling":"f302542c38ba7c3aab7c9280dd60259ecec777dca6e6f71b6f0729b0b8791b72","side":"left"},{"sibling":"d8b9143917b539c543cf4448cec00131f8b807bd8004979c54ebe09798748c66","side":"left"},{"sibling":"3ee6ab8db1878db0b324dd807ca233edd472899b8e90d2c5e018848ea5d12d99","side":"right"},{"sibling":"73a8d1605c15a74de720f0c54b5b4567e3eb8b5999bd8812dc66eedc461540fa","side":"right"},{"sibling":"1cecb7f447febd025aac272837c80de218aecc6485d2395a509b2a1f1b9c746e","side":"left"}]},"schema":"crovia.axiom_proof.v1","seal":{"first_collector_run_id":"","first_receipt_hash":"","jsonl_path":"/opt/crovia/substrate/axiom_ledger.jsonl","key_id":"430895f101d38164","last_collector_run_id":"","last_receipt_hash":"","leaf_count":320080,"merkle_root":"1e159bceacfdb1f3c930dae410f8759ddc4c762abcfa4a257b150d6f44ab16e0","public_key_hex":"cf742e26f75669dc673cb5c0786a1ae23ae8ca19c347317192ce40c28a7ff25c","run_id":"hourly_json_retrofit_20260716T053701Z","schema":"crovia.seal.v1","seal_family_version":"crovia-seal-family/1","seal_kind":"substrate_batch","sealed_at":"2026-07-16T05:38:31Z","sig_algorithm":"ed25519","signature":"c1cd1fc5704d4fec4dee47b29a0e9877f8cfb41eb61dcfd783d82344a15f6e44d80d6cd5128849547b62fe9d9724fc71433319ab6030445b4ac1c9e6bf8a3709","signer_version":"1.1.0"},"trust_root":{"key_id":"430895f101d38164","public_key_hex":"cf742e26f75669dc673cb5c0786a1ae23ae8ca19c347317192ce40c28a7ff25c","signature_algorithm":"ed25519","url":"/registry/canon/TRUST_ROOT.md"},"verifier":{"spec":"/registry/canon/AXIOM_RECEIPT_v1.md","url":"/v/axm_b59c0338595fd0ef3be6c1fc0534e0ce6df3941c2e221c9e765ed84b3b8a0011"}}