{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:T7ZFHTWF5IV4BK55EHBUTMGYTU","short_pith_number":"pith:T7ZFHTWF","schema_version":"1.0","canonical_sha256":"9ff253cec5ea2bc0abbd21c349b0d89d1ab10ef8e7f0ae053f7fb7e39ab86af0","source":{"kind":"arxiv","id":"2411.08733","version":2},"attestation_state":"computed","paper":{"title":"Dynamic Rewarding with Prompt Optimization Enables Tuning-free Self-Alignment of Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abdullah Ashfaq, Eric P. Xing, Somanshu Singla, Tianyang Liu, Zhen Wang, Zhiting Hu","submitted_at":"2024-11-13T16:15:38Z","abstract_excerpt":"Aligning Large Language Models (LLMs) traditionally relies on costly training and human preference annotations. Self-alignment seeks to reduce these expenses by enabling models to align themselves. To further lower costs and achieve alignment without any expensive tuning or annotations, we introduce a new tuning-free approach for self-alignment, Dynamic Rewarding with Prompt Optimization (DRPO). Our approach leverages a search-based optimization framework that allows LLMs to iteratively self-improve and craft the optimal alignment instructions, all without additional training or human interven"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2411.08733","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-13T16:15:38Z","cross_cats_sorted":[],"title_canon_sha256":"df0b34a9d016312f2962328126e856f744a206342fe63b66e3fa3503d33dce41","abstract_canon_sha256":"0f707695a28f6162398c07ef62a090c512986684e8cfb8e578f13c416b4a6a4f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:35:15.441253Z","signature_b64":"+6zc/cdFcp1bI0PrltlC7HJ+I9uE9qwqiDm85plTiESoz3hG/ezrkJ45knucr/0+50anbl1E6JYcjqnIHQmtDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9ff253cec5ea2bc0abbd21c349b0d89d1ab10ef8e7f0ae053f7fb7e39ab86af0","last_reissued_at":"2026-07-05T09:35:15.440784Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:35:15.440784Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dynamic Rewarding with Prompt Optimization Enables Tuning-free Self-Alignment of Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abdullah Ashfaq, Eric P. Xing, Somanshu Singla, Tianyang Liu, Zhen Wang, Zhiting Hu","submitted_at":"2024-11-13T16:15:38Z","abstract_excerpt":"Aligning Large Language Models (LLMs) traditionally relies on costly training and human preference annotations. Self-alignment seeks to reduce these expenses by enabling models to align themselves. To further lower costs and achieve alignment without any expensive tuning or annotations, we introduce a new tuning-free approach for self-alignment, Dynamic Rewarding with Prompt Optimization (DRPO). Our approach leverages a search-based optimization framework that allows LLMs to iteratively self-improve and craft the optimal alignment instructions, all without additional training or human interven"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.08733","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/2411.08733/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2411.08733","created_at":"2026-07-05T09:35:15.440838+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.08733v2","created_at":"2026-07-05T09:35:15.440838+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.08733","created_at":"2026-07-05T09:35:15.440838+00:00"},{"alias_kind":"pith_short_12","alias_value":"T7ZFHTWF5IV4","created_at":"2026-07-05T09:35:15.440838+00:00"},{"alias_kind":"pith_short_16","alias_value":"T7ZFHTWF5IV4BK55","created_at":"2026-07-05T09:35:15.440838+00:00"},{"alias_kind":"pith_short_8","alias_value":"T7ZFHTWF","created_at":"2026-07-05T09:35:15.440838+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.24339","citing_title":"See Further, Think Deeper: Advancing VLM's Reasoning Ability with Low-level Visual Cues and Reflection","ref_index":43,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T7ZFHTWF5IV4BK55EHBUTMGYTU","json":"https://pith.science/pith/T7ZFHTWF5IV4BK55EHBUTMGYTU.json","graph_json":"https://pith.science/api/pith-number/T7ZFHTWF5IV4BK55EHBUTMGYTU/graph.json","events_json":"https://pith.science/api/pith-number/T7ZFHTWF5IV4BK55EHBUTMGYTU/events.json","paper":"https://pith.science/paper/T7ZFHTWF"},"agent_actions":{"view_html":"https://pith.science/pith/T7ZFHTWF5IV4BK55EHBUTMGYTU","download_json":"https://pith.science/pith/T7ZFHTWF5IV4BK55EHBUTMGYTU.json","view_paper":"https://pith.science/paper/T7ZFHTWF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.08733&json=true","fetch_graph":"https://pith.science/api/pith-number/T7ZFHTWF5IV4BK55EHBUTMGYTU/graph.json","fetch_events":"https://pith.science/api/pith-number/T7ZFHTWF5IV4BK55EHBUTMGYTU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T7ZFHTWF5IV4BK55EHBUTMGYTU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T7ZFHTWF5IV4BK55EHBUTMGYTU/action/storage_attestation","attest_author":"https://pith.science/pith/T7ZFHTWF5IV4BK55EHBUTMGYTU/action/author_attestation","sign_citation":"https://pith.science/pith/T7ZFHTWF5IV4BK55EHBUTMGYTU/action/citation_signature","submit_replication":"https://pith.science/pith/T7ZFHTWF5IV4BK55EHBUTMGYTU/action/replication_record"}},"created_at":"2026-07-05T09:35:15.440838+00:00","updated_at":"2026-07-05T09:35:15.440838+00:00"}