{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ASAMY472MDZ3K2YLEO73XOMVWW","short_pith_number":"pith:ASAMY472","canonical_record":{"source":{"id":"2512.12677","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-12-14T13:02:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"07b4f42c7c7fd67c6a94237c3a92cf0901ce57aeedb5c677ba14ec68852938c6","abstract_canon_sha256":"bcf6e6008a93594f127c6e2271ce0bd22b7f2e3b8cd02cb969dca5b8414ea48d"},"schema_version":"1.0"},"canonical_sha256":"0480cc73fa60f3b56b0b23bfbbb995b5a627cbc09dca28b4c1532858135fb896","source":{"kind":"arxiv","id":"2512.12677","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.12677","created_at":"2026-05-25T02:01:10Z"},{"alias_kind":"arxiv_version","alias_value":"2512.12677v2","created_at":"2026-05-25T02:01:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.12677","created_at":"2026-05-25T02:01:10Z"},{"alias_kind":"pith_short_12","alias_value":"ASAMY472MDZ3","created_at":"2026-05-25T02:01:10Z"},{"alias_kind":"pith_short_16","alias_value":"ASAMY472MDZ3K2YL","created_at":"2026-05-25T02:01:10Z"},{"alias_kind":"pith_short_8","alias_value":"ASAMY472","created_at":"2026-05-25T02:01:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ASAMY472MDZ3K2YLEO73XOMVWW","target":"record","payload":{"canonical_record":{"source":{"id":"2512.12677","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-12-14T13:02:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"07b4f42c7c7fd67c6a94237c3a92cf0901ce57aeedb5c677ba14ec68852938c6","abstract_canon_sha256":"bcf6e6008a93594f127c6e2271ce0bd22b7f2e3b8cd02cb969dca5b8414ea48d"},"schema_version":"1.0"},"canonical_sha256":"0480cc73fa60f3b56b0b23bfbbb995b5a627cbc09dca28b4c1532858135fb896","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:01:10.843004Z","signature_b64":"r+qMCKJKSp0d6RAWUk+uYQpO+41qOMwK4ZYCGoTO/qCjoK6YcZ4vWyJIed/FfN6XI35nF9f6NMglQVG+r8ECBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0480cc73fa60f3b56b0b23bfbbb995b5a627cbc09dca28b4c1532858135fb896","last_reissued_at":"2026-05-25T02:01:10.842431Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:01:10.842431Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2512.12677","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":"2026-05-25T02:01:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pDzxfvXLGablHIIU/DhgpA42KgWFt6g3CSb9E2CZ0KQjVFvO6ISfaF8b39i5rr/Qn/7KyBYGoE00pCN7HmVqDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T19:26:50.775083Z"},"content_sha256":"8e45f55f1dc7ff74002521669aa043c900b1fed8e4150753802906ffb3a0a0f9","schema_version":"1.0","event_id":"sha256:8e45f55f1dc7ff74002521669aa043c900b1fed8e4150753802906ffb3a0a0f9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ASAMY472MDZ3K2YLEO73XOMVWW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Fine-Tuning Causal LLMs for Text Classification: Embedding-Based vs. Instruction-Based Approaches","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Amirhossein Yousefiramandi, Ciaran Cooney","submitted_at":"2025-12-14T13:02:06Z","abstract_excerpt":"We explore efficient strategies to fine-tune decoder-only Large Language Models (LLMs) for downstream text classification under resource constraints. Two approaches are investigated: (1) attaching a classification head to a pre-trained causal LLM and fine-tuning on the task using the LLM's final-token embedding as a sequence representation, and (2) instruction-tuning the LLM in a prompt-to-response format for classification. To enable single-GPU fine-tuning of models up to 8B parameters, we combine 4-bit model quantization with Low-Rank Adaptation (LoRA) for parameter-efficient training. Exper"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.12677","kind":"arxiv","version":2},"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/2512.12677/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":"2026-05-25T02:01:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"W7XNcG0PFCOr20N0QSyhuYa3W4MWzdsY1RIgc3urz5eMMm3ugKi+0QtFcAcgYwZAqM9C9icN1iiihXxiRErzBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T19:26:50.775491Z"},"content_sha256":"bad44d38f641cf7a563c5c8c318023d1424d2e2f3c194576073b5ff723e2cbd8","schema_version":"1.0","event_id":"sha256:bad44d38f641cf7a563c5c8c318023d1424d2e2f3c194576073b5ff723e2cbd8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ASAMY472MDZ3K2YLEO73XOMVWW/bundle.json","state_url":"https://pith.science/pith/ASAMY472MDZ3K2YLEO73XOMVWW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ASAMY472MDZ3K2YLEO73XOMVWW/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":"2026-06-29T19:26:50Z","links":{"resolver":"https://pith.science/pith/ASAMY472MDZ3K2YLEO73XOMVWW","bundle":"https://pith.science/pith/ASAMY472MDZ3K2YLEO73XOMVWW/bundle.json","state":"https://pith.science/pith/ASAMY472MDZ3K2YLEO73XOMVWW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ASAMY472MDZ3K2YLEO73XOMVWW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ASAMY472MDZ3K2YLEO73XOMVWW","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":"bcf6e6008a93594f127c6e2271ce0bd22b7f2e3b8cd02cb969dca5b8414ea48d","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-12-14T13:02:06Z","title_canon_sha256":"07b4f42c7c7fd67c6a94237c3a92cf0901ce57aeedb5c677ba14ec68852938c6"},"schema_version":"1.0","source":{"id":"2512.12677","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.12677","created_at":"2026-05-25T02:01:10Z"},{"alias_kind":"arxiv_version","alias_value":"2512.12677v2","created_at":"2026-05-25T02:01:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.12677","created_at":"2026-05-25T02:01:10Z"},{"alias_kind":"pith_short_12","alias_value":"ASAMY472MDZ3","created_at":"2026-05-25T02:01:10Z"},{"alias_kind":"pith_short_16","alias_value":"ASAMY472MDZ3K2YL","created_at":"2026-05-25T02:01:10Z"},{"alias_kind":"pith_short_8","alias_value":"ASAMY472","created_at":"2026-05-25T02:01:10Z"}],"graph_snapshots":[{"event_id":"sha256:bad44d38f641cf7a563c5c8c318023d1424d2e2f3c194576073b5ff723e2cbd8","target":"graph","created_at":"2026-05-25T02:01:10Z","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":[],"endpoint":"/pith/2512.12677/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We explore efficient strategies to fine-tune decoder-only Large Language Models (LLMs) for downstream text classification under resource constraints. Two approaches are investigated: (1) attaching a classification head to a pre-trained causal LLM and fine-tuning on the task using the LLM's final-token embedding as a sequence representation, and (2) instruction-tuning the LLM in a prompt-to-response format for classification. To enable single-GPU fine-tuning of models up to 8B parameters, we combine 4-bit model quantization with Low-Rank Adaptation (LoRA) for parameter-efficient training. Exper","authors_text":"Amirhossein Yousefiramandi, Ciaran Cooney","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-12-14T13:02:06Z","title":"Fine-Tuning Causal LLMs for Text Classification: Embedding-Based vs. Instruction-Based Approaches"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.12677","kind":"arxiv","version":2},"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":"sha256:8e45f55f1dc7ff74002521669aa043c900b1fed8e4150753802906ffb3a0a0f9","target":"record","created_at":"2026-05-25T02:01:10Z","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":"bcf6e6008a93594f127c6e2271ce0bd22b7f2e3b8cd02cb969dca5b8414ea48d","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-12-14T13:02:06Z","title_canon_sha256":"07b4f42c7c7fd67c6a94237c3a92cf0901ce57aeedb5c677ba14ec68852938c6"},"schema_version":"1.0","source":{"id":"2512.12677","kind":"arxiv","version":2}},"canonical_sha256":"0480cc73fa60f3b56b0b23bfbbb995b5a627cbc09dca28b4c1532858135fb896","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0480cc73fa60f3b56b0b23bfbbb995b5a627cbc09dca28b4c1532858135fb896","first_computed_at":"2026-05-25T02:01:10.842431Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-25T02:01:10.842431Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"r+qMCKJKSp0d6RAWUk+uYQpO+41qOMwK4ZYCGoTO/qCjoK6YcZ4vWyJIed/FfN6XI35nF9f6NMglQVG+r8ECBA==","signature_status":"signed_v1","signed_at":"2026-05-25T02:01:10.843004Z","signed_message":"canonical_sha256_bytes"},"source_id":"2512.12677","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8e45f55f1dc7ff74002521669aa043c900b1fed8e4150753802906ffb3a0a0f9","sha256:bad44d38f641cf7a563c5c8c318023d1424d2e2f3c194576073b5ff723e2cbd8"],"state_sha256":"26d1d0ddabedf7b4e4242feba9d7de507def3937feaa297bf1f4ea0ce3d3583d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Fucrw7weHfKANekU+CUlNhYmbwcvhofJkcir5hwTPmdnDe9n5jqSfoBP7DkdGrsshxPro20ZpZUBdHDQ2etbCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-29T19:26:50.780354Z","bundle_sha256":"d8d56a9b40d6ff45633f4f8ab12914709e4ad0375bd9a6c1f9777639ecbe6c09"}}