{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:YELJ2MS2SRZZ7EQWSNRSEUZPWY","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":"ed612ac693b3f6b72a2966c4ee21ce2349173a9c7f2695217200644d5f6b8697","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-09T06:36:06Z","title_canon_sha256":"58d287ceba5e5e1ad4f02ca241526ed24e0dce68af02a3302fd4855d8e1cb421"},"schema_version":"1.0","source":{"id":"2606.09850","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.09850","created_at":"2026-06-10T00:08:28Z"},{"alias_kind":"arxiv_version","alias_value":"2606.09850v1","created_at":"2026-06-10T00:08:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09850","created_at":"2026-06-10T00:08:28Z"},{"alias_kind":"pith_short_12","alias_value":"YELJ2MS2SRZZ","created_at":"2026-06-10T00:08:28Z"},{"alias_kind":"pith_short_16","alias_value":"YELJ2MS2SRZZ7EQW","created_at":"2026-06-10T00:08:28Z"},{"alias_kind":"pith_short_8","alias_value":"YELJ2MS2","created_at":"2026-06-10T00:08:28Z"}],"graph_snapshots":[{"event_id":"sha256:df3cc4623abc8c4d0c8d77eafc3dbd75feda043126c6e8ba0384a451a79b0440","target":"graph","created_at":"2026-06-10T00:08:28Z","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/2606.09850/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Post-training alignment algorithms are predominantly evaluated as black boxes, obscuring how they reshape language models' internal computations. We present a systematic mechanistic analysis of six preference-optimization methods: PPO, DPO, SimPO, ORPO, GRPO, and KTO across three open-weight model families. By integrating layer-wise linear probing, Sparse Autoencoders, and crosscoders, we localize preference representations and quantify alignment-induced geometric transformations in latent space. We find that preference signals consistently concentrate in early--mid or mid--late layers, but di","authors_text":"Aarush Sinha, Arth Singh, Ishan Garg, Kushal Garg, Veeraraju Elluru","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-09T06:36:06Z","title":"Mechanistic Analysis of Alignment Algorithms in Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09850","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":"sha256:5603fb76ec66ddd84e34f720addfc2db940a4a51d584cec1ed695b0ec49b94d7","target":"record","created_at":"2026-06-10T00:08:28Z","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":"ed612ac693b3f6b72a2966c4ee21ce2349173a9c7f2695217200644d5f6b8697","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-09T06:36:06Z","title_canon_sha256":"58d287ceba5e5e1ad4f02ca241526ed24e0dce68af02a3302fd4855d8e1cb421"},"schema_version":"1.0","source":{"id":"2606.09850","kind":"arxiv","version":1}},"canonical_sha256":"c1169d325a94739f9216936322532fb632c6ae938c3b429bc23bfaa55590d955","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c1169d325a94739f9216936322532fb632c6ae938c3b429bc23bfaa55590d955","first_computed_at":"2026-06-10T00:08:28.478102Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-10T00:08:28.478102Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QWY6gPqm0KVC/OdsUA5gFj3WV/AK2JOIb4PekfP4UK6ex2EaNaKzRM2lCB5XWGu2ywVWoSkWZRmzr+XSATQ8AA==","signature_status":"signed_v1","signed_at":"2026-06-10T00:08:28.479046Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.09850","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5603fb76ec66ddd84e34f720addfc2db940a4a51d584cec1ed695b0ec49b94d7","sha256:df3cc4623abc8c4d0c8d77eafc3dbd75feda043126c6e8ba0384a451a79b0440"],"state_sha256":"8f26c91dba031779682c856b976d428b141469ec86939592be4e5f90119dc001"}