{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:F4MGJBXU27N65PGT3SJF4P4R3N","short_pith_number":"pith:F4MGJBXU","canonical_record":{"source":{"id":"2512.21651","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-25T12:39:36Z","cross_cats_sorted":[],"title_canon_sha256":"c96f8566f1bc3e12dbf3aa669a763aa8ce6b1dcd8fa9828b63e5c75139bdee0f","abstract_canon_sha256":"b7813dbbb929cfa9b34818e0ef81958d5bc03906d72ec113085f09350d188f7b"},"schema_version":"1.0"},"canonical_sha256":"2f186486f4d7dbeebcd3dc925e3f91db68b0b1662121c55f3eefd966275f28b2","source":{"kind":"arxiv","id":"2512.21651","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.21651","created_at":"2026-05-17T23:39:00Z"},{"alias_kind":"arxiv_version","alias_value":"2512.21651v2","created_at":"2026-05-17T23:39:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.21651","created_at":"2026-05-17T23:39:00Z"},{"alias_kind":"pith_short_12","alias_value":"F4MGJBXU27N6","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"F4MGJBXU27N65PGT","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"F4MGJBXU","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:F4MGJBXU27N65PGT3SJF4P4R3N","target":"record","payload":{"canonical_record":{"source":{"id":"2512.21651","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-25T12:39:36Z","cross_cats_sorted":[],"title_canon_sha256":"c96f8566f1bc3e12dbf3aa669a763aa8ce6b1dcd8fa9828b63e5c75139bdee0f","abstract_canon_sha256":"b7813dbbb929cfa9b34818e0ef81958d5bc03906d72ec113085f09350d188f7b"},"schema_version":"1.0"},"canonical_sha256":"2f186486f4d7dbeebcd3dc925e3f91db68b0b1662121c55f3eefd966275f28b2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:00.375753Z","signature_b64":"9u51UYruBykKpp/w6dtw5hLS8B8zljTgoJ/yaAbF9bycjJatI8cx/TP8wuzkCX1EqyGtre7dG/EClEaQj/DfBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f186486f4d7dbeebcd3dc925e3f91db68b0b1662121c55f3eefd966275f28b2","last_reissued_at":"2026-05-17T23:39:00.375078Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:00.375078Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2512.21651","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":"2026-05-17T23:39:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"achKvoQTjJKspcOpMia/uRQCMZbsvmFcbVev7cQQ1Ijda3A7FYs17FH4x6cMl4p9AXS0dD+nKgrgE380rgc1Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T10:30:17.797770Z"},"content_sha256":"0894d7b260be9533a8004e6d980611f4dcb8937b8f2d8795b251a2365c1676e3","schema_version":"1.0","event_id":"sha256:0894d7b260be9533a8004e6d980611f4dcb8937b8f2d8795b251a2365c1676e3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:F4MGJBXU27N65PGT3SJF4P4R3N","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Rethinking Output Alignment For 1-bit Post-Training Quantization of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Naive output alignment fails in 1-bit LLM quantization because errors accumulate across layers and distort the representation space unevenly.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Cuong Nguyen, Cuong Pham, Dung Anh Hoang, Jianfei Cai, Thanh-Toan Do, Trung Le","submitted_at":"2025-12-25T12:39:36Z","abstract_excerpt":"Large Language Models (LLMs) deliver strong performance across a wide range of NLP tasks, but their massive sizes hinder deployment on resource-constrained devices. To reduce their computational and memory burden, various compression techniques have been proposed, including quantization, pruning, and knowledge distillation. Among these, post-training quantization (PTQ) is widely adopted for its efficiency, as it requires no retraining and only a small dataset for calibration, enabling low-cost deployment. Recent advances for post-training quantization have demonstrated that even near 4-bit met"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"we show that this failure arises from two fundamental issues: error accumulation across layers and, more critically, anisotropic distortion of the representation space. Based on these insights, we propose a novel PTQ method for 1-bit LLMs that explicitly addresses these issues while maintaining computational efficiency.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that correcting error accumulation and anisotropic distortion on a small calibration set will generalize to the full test distribution without introducing new distortions or requiring architecture-specific tuning.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A post-training 1-bit quantization method for LLMs that fixes error accumulation and anisotropic representation distortion to outperform prior weight-driven and naive output-driven baselines.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Naive output alignment fails in 1-bit LLM quantization because errors accumulate across layers and distort the representation space unevenly.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"224ff5bc432eafc18df120726120f86c4c3cd7b46daa7e979c46ced76a7fe856"},"source":{"id":"2512.21651","kind":"arxiv","version":2},"verdict":{"id":"8b19c048-9dd2-4c3b-ae39-a8155570b87f","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T19:27:59.724906Z","strongest_claim":"we show that this failure arises from two fundamental issues: error accumulation across layers and, more critically, anisotropic distortion of the representation space. Based on these insights, we propose a novel PTQ method for 1-bit LLMs that explicitly addresses these issues while maintaining computational efficiency.","one_line_summary":"A post-training 1-bit quantization method for LLMs that fixes error accumulation and anisotropic representation distortion to outperform prior weight-driven and naive output-driven baselines.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that correcting error accumulation and anisotropic distortion on a small calibration set will generalize to the full test distribution without introducing new distortions or requiring architecture-specific tuning.","pith_extraction_headline":"Naive output alignment fails in 1-bit LLM quantization because errors accumulate across layers and distort the representation space unevenly."},"references":{"count":22,"sample":[{"doi":"10.1609/aaai.v34i05.6239","year":2005,"title":"Language Models are Few-Shot Learners","work_id":"214732c0-2edd-44a0-af9e-28184a2b8279","ref_index":1,"cited_arxiv_id":"2005.14165","is_internal_anchor":true},{"doi":"","year":null,"title":"Stbllm: Breaking the 1-bit barrier with structured binary llms","work_id":"f9cf4e8e-ce01-40e5-820a-ae805db27e12","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"Network sketching: Exploiting bi- nary structure in deep cnns.2017 IEEE Conference on Computer Vision and Pattern Recog- nition (CVPR), pp","work_id":"18dc22c6-7aeb-46ba-8eab-94ca65110314","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding","work_id":"d79f6a26-2d92-4f7c-aea0-8aabf3b668c8","ref_index":4,"cited_arxiv_id":"1510.00149","is_internal_anchor":true},{"doi":"","year":null,"title":"Distilling the Knowledge in a Neural Network","work_id":"d927ab1f-17b8-4002-9d09-c3d55764fbad","ref_index":5,"cited_arxiv_id":"1503.02531","is_internal_anchor":true}],"resolved_work":22,"snapshot_sha256":"1b2bbe69b439b266218239681008d63b393781439a83862d6e51b0eaac1b81b6","internal_anchors":10},"formal_canon":{"evidence_count":2,"snapshot_sha256":"0fa99115169d6e8517e200daa4364b1dcbb699592f7b4bc95a067dd42541b33e"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"8b19c048-9dd2-4c3b-ae39-a8155570b87f"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pjPerXGk2fLr4z44g7GBSjkValFcv0LRf3+gFhriNvE1piy+JEeo5LYw6jOdAZrkAh36k7GPrrSrYoLp1i/0Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T10:30:17.798305Z"},"content_sha256":"ed9a720217cff91ccf6326588e69582c86c729c2dae094244086e044957d35dc","schema_version":"1.0","event_id":"sha256:ed9a720217cff91ccf6326588e69582c86c729c2dae094244086e044957d35dc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/F4MGJBXU27N65PGT3SJF4P4R3N/bundle.json","state_url":"https://pith.science/pith/F4MGJBXU27N65PGT3SJF4P4R3N/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/F4MGJBXU27N65PGT3SJF4P4R3N/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":"2026-05-27T10:30:17Z","links":{"resolver":"https://pith.science/pith/F4MGJBXU27N65PGT3SJF4P4R3N","bundle":"https://pith.science/pith/F4MGJBXU27N65PGT3SJF4P4R3N/bundle.json","state":"https://pith.science/pith/F4MGJBXU27N65PGT3SJF4P4R3N/state.json","well_known_bundle":"https://pith.science/.well-known/pith/F4MGJBXU27N65PGT3SJF4P4R3N/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:F4MGJBXU27N65PGT3SJF4P4R3N","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":"b7813dbbb929cfa9b34818e0ef81958d5bc03906d72ec113085f09350d188f7b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-25T12:39:36Z","title_canon_sha256":"c96f8566f1bc3e12dbf3aa669a763aa8ce6b1dcd8fa9828b63e5c75139bdee0f"},"schema_version":"1.0","source":{"id":"2512.21651","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.21651","created_at":"2026-05-17T23:39:00Z"},{"alias_kind":"arxiv_version","alias_value":"2512.21651v2","created_at":"2026-05-17T23:39:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.21651","created_at":"2026-05-17T23:39:00Z"},{"alias_kind":"pith_short_12","alias_value":"F4MGJBXU27N6","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"F4MGJBXU27N65PGT","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"F4MGJBXU","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:ed9a720217cff91ccf6326588e69582c86c729c2dae094244086e044957d35dc","target":"graph","created_at":"2026-05-17T23:39:00Z","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":"we show that this failure arises from two fundamental issues: error accumulation across layers and, more critically, anisotropic distortion of the representation space. Based on these insights, we propose a novel PTQ method for 1-bit LLMs that explicitly addresses these issues while maintaining computational efficiency."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The assumption that correcting error accumulation and anisotropic distortion on a small calibration set will generalize to the full test distribution without introducing new distortions or requiring architecture-specific tuning."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A post-training 1-bit quantization method for LLMs that fixes error accumulation and anisotropic representation distortion to outperform prior weight-driven and naive output-driven baselines."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Naive output alignment fails in 1-bit LLM quantization because errors accumulate across layers and distort the representation space unevenly."}],"snapshot_sha256":"224ff5bc432eafc18df120726120f86c4c3cd7b46daa7e979c46ced76a7fe856"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"0fa99115169d6e8517e200daa4364b1dcbb699592f7b4bc95a067dd42541b33e"},"paper":{"abstract_excerpt":"Large Language Models (LLMs) deliver strong performance across a wide range of NLP tasks, but their massive sizes hinder deployment on resource-constrained devices. To reduce their computational and memory burden, various compression techniques have been proposed, including quantization, pruning, and knowledge distillation. Among these, post-training quantization (PTQ) is widely adopted for its efficiency, as it requires no retraining and only a small dataset for calibration, enabling low-cost deployment. Recent advances for post-training quantization have demonstrated that even near 4-bit met","authors_text":"Cuong Nguyen, Cuong Pham, Dung Anh Hoang, Jianfei Cai, Thanh-Toan Do, Trung Le","cross_cats":[],"headline":"Naive output alignment fails in 1-bit LLM quantization because errors accumulate across layers and distort the representation space unevenly.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-25T12:39:36Z","title":"Rethinking Output Alignment For 1-bit Post-Training Quantization of Large Language Models"},"references":{"count":22,"internal_anchors":10,"resolved_work":22,"sample":[{"cited_arxiv_id":"2005.14165","doi":"10.1609/aaai.v34i05.6239","is_internal_anchor":true,"ref_index":1,"title":"Language Models are Few-Shot Learners","work_id":"214732c0-2edd-44a0-af9e-28184a2b8279","year":2005},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Stbllm: Breaking the 1-bit barrier with structured binary llms","work_id":"f9cf4e8e-ce01-40e5-820a-ae805db27e12","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Network sketching: Exploiting bi- nary structure in deep cnns.2017 IEEE Conference on Computer Vision and Pattern Recog- nition (CVPR), pp","work_id":"18dc22c6-7aeb-46ba-8eab-94ca65110314","year":2017},{"cited_arxiv_id":"1510.00149","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding","work_id":"d79f6a26-2d92-4f7c-aea0-8aabf3b668c8","year":null},{"cited_arxiv_id":"1503.02531","doi":"","is_internal_anchor":true,"ref_index":5,"title":"Distilling the Knowledge in a Neural Network","work_id":"d927ab1f-17b8-4002-9d09-c3d55764fbad","year":null}],"snapshot_sha256":"1b2bbe69b439b266218239681008d63b393781439a83862d6e51b0eaac1b81b6"},"source":{"id":"2512.21651","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-16T19:27:59.724906Z","id":"8b19c048-9dd2-4c3b-ae39-a8155570b87f","model_set":{"reader":"grok-4.3"},"one_line_summary":"A post-training 1-bit quantization method for LLMs that fixes error accumulation and anisotropic representation distortion to outperform prior weight-driven and naive output-driven baselines.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Naive output alignment fails in 1-bit LLM quantization because errors accumulate across layers and distort the representation space unevenly.","strongest_claim":"we show that this failure arises from two fundamental issues: error accumulation across layers and, more critically, anisotropic distortion of the representation space. Based on these insights, we propose a novel PTQ method for 1-bit LLMs that explicitly addresses these issues while maintaining computational efficiency.","weakest_assumption":"The assumption that correcting error accumulation and anisotropic distortion on a small calibration set will generalize to the full test distribution without introducing new distortions or requiring architecture-specific tuning."}},"verdict_id":"8b19c048-9dd2-4c3b-ae39-a8155570b87f"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0894d7b260be9533a8004e6d980611f4dcb8937b8f2d8795b251a2365c1676e3","target":"record","created_at":"2026-05-17T23:39:00Z","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":"b7813dbbb929cfa9b34818e0ef81958d5bc03906d72ec113085f09350d188f7b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-25T12:39:36Z","title_canon_sha256":"c96f8566f1bc3e12dbf3aa669a763aa8ce6b1dcd8fa9828b63e5c75139bdee0f"},"schema_version":"1.0","source":{"id":"2512.21651","kind":"arxiv","version":2}},"canonical_sha256":"2f186486f4d7dbeebcd3dc925e3f91db68b0b1662121c55f3eefd966275f28b2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2f186486f4d7dbeebcd3dc925e3f91db68b0b1662121c55f3eefd966275f28b2","first_computed_at":"2026-05-17T23:39:00.375078Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:00.375078Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9u51UYruBykKpp/w6dtw5hLS8B8zljTgoJ/yaAbF9bycjJatI8cx/TP8wuzkCX1EqyGtre7dG/EClEaQj/DfBw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:00.375753Z","signed_message":"canonical_sha256_bytes"},"source_id":"2512.21651","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0894d7b260be9533a8004e6d980611f4dcb8937b8f2d8795b251a2365c1676e3","sha256:ed9a720217cff91ccf6326588e69582c86c729c2dae094244086e044957d35dc"],"state_sha256":"2fd81c875bec34a376c44111b43d7022ef813fdfbcae85ea1aeb0bcf792b88e4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wdGBS6sxaxIVVnWo49mGz+6Acec0oOEmK8S4XBa41QiEcHEBR3q1NLJUN8e9/KQYzbiByx859qDq7lVyMIaQAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T10:30:17.801043Z","bundle_sha256":"69e9b478d4be1fbf807cceed5d93b1ef1f6f71bdfbb1412be216c5492c57c67b"}}