{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:Z6FP4LKYHDNPPT6WJHNPTNZXMI","short_pith_number":"pith:Z6FP4LKY","canonical_record":{"source":{"id":"2505.20777","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-27T06:30:48Z","cross_cats_sorted":[],"title_canon_sha256":"25a07a2d4b7d0e1ec48939bc1bad67a6acd7117dec7f6acc23b071c31c15e587","abstract_canon_sha256":"98a1b57520d1c42afc7cd77692bd2409d5625bc5fa2f976ebb25f327a502902b"},"schema_version":"1.0"},"canonical_sha256":"cf8afe2d5838daf7cfd649daf9b737622ebafef34cac30aa610d6684ee2bc81e","source":{"kind":"arxiv","id":"2505.20777","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.20777","created_at":"2026-07-05T11:10:18Z"},{"alias_kind":"arxiv_version","alias_value":"2505.20777v1","created_at":"2026-07-05T11:10:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.20777","created_at":"2026-07-05T11:10:18Z"},{"alias_kind":"pith_short_12","alias_value":"Z6FP4LKYHDNP","created_at":"2026-07-05T11:10:18Z"},{"alias_kind":"pith_short_16","alias_value":"Z6FP4LKYHDNPPT6W","created_at":"2026-07-05T11:10:18Z"},{"alias_kind":"pith_short_8","alias_value":"Z6FP4LKY","created_at":"2026-07-05T11:10:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:Z6FP4LKYHDNPPT6WJHNPTNZXMI","target":"record","payload":{"canonical_record":{"source":{"id":"2505.20777","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-27T06:30:48Z","cross_cats_sorted":[],"title_canon_sha256":"25a07a2d4b7d0e1ec48939bc1bad67a6acd7117dec7f6acc23b071c31c15e587","abstract_canon_sha256":"98a1b57520d1c42afc7cd77692bd2409d5625bc5fa2f976ebb25f327a502902b"},"schema_version":"1.0"},"canonical_sha256":"cf8afe2d5838daf7cfd649daf9b737622ebafef34cac30aa610d6684ee2bc81e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:18.731618Z","signature_b64":"8BRFxCBQsFUqo8fOsmZ6I9bX0SWwxMRMrEbn3JAVUlYOdWy5xFEoS3P5nx9reU8yU5ZrMw8GImfNJxwzg5dcBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cf8afe2d5838daf7cfd649daf9b737622ebafef34cac30aa610d6684ee2bc81e","last_reissued_at":"2026-07-05T11:10:18.730911Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:18.730911Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.20777","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:10:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BM8xBXQPrtenYtWjRLeDghQwE6SW09MniDUpPXmKvEexyOnd2PeWTlHa5hWstHDZ4XWS/R21u7I5E6dxYLoDAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T22:52:36.495810Z"},"content_sha256":"6445c2d0cef7eb8f2b7972193e0ff505a175f984a1ac009d091aa449e0e5b75d","schema_version":"1.0","event_id":"sha256:6445c2d0cef7eb8f2b7972193e0ff505a175f984a1ac009d091aa449e0e5b75d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:Z6FP4LKYHDNPPT6WJHNPTNZXMI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TACO: Think-Answer Consistency for Optimized Long-Chain Reasoning and Efficient Data Learning via Reinforcement Learning in LVLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Deqiang Jiang, Haoyu Cao, Kun Yin, Qingmin Liao, Wenming Yang, Xinghua Jiang, Xing Sun, Xin Li, Yanlin Liu, Yinsong Liu, Zhehan Kan","submitted_at":"2025-05-27T06:30:48Z","abstract_excerpt":"DeepSeek R1 has significantly advanced complex reasoning for large language models (LLMs). While recent methods have attempted to replicate R1's reasoning capabilities in multimodal settings, they face limitations, including inconsistencies between reasoning and final answers, model instability and crashes during long-chain exploration, and low data learning efficiency. To address these challenges, we propose TACO, a novel reinforcement learning algorithm for visual reasoning. Building on Generalized Reinforcement Policy Optimization (GRPO), TACO introduces Think-Answer Consistency, which tigh"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.20777","kind":"arxiv","version":1},"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/2505.20777/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":"2026-07-05T11:10:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hGGPLgrYRaGnumrZAve4RG7Odg08xW/qgiuka0xJzp/ycRsZsSmlj6Jbjb3iyFUpQ+P42jKIPsbHyuGt5t/iDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T22:52:36.496325Z"},"content_sha256":"7ba8d58a0d74ec0384af4255a319f3ce42165909626df817695215cbb208601d","schema_version":"1.0","event_id":"sha256:7ba8d58a0d74ec0384af4255a319f3ce42165909626df817695215cbb208601d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Z6FP4LKYHDNPPT6WJHNPTNZXMI/bundle.json","state_url":"https://pith.science/pith/Z6FP4LKYHDNPPT6WJHNPTNZXMI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Z6FP4LKYHDNPPT6WJHNPTNZXMI/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":"2026-08-14T22:52:36Z","links":{"resolver":"https://pith.science/pith/Z6FP4LKYHDNPPT6WJHNPTNZXMI","bundle":"https://pith.science/pith/Z6FP4LKYHDNPPT6WJHNPTNZXMI/bundle.json","state":"https://pith.science/pith/Z6FP4LKYHDNPPT6WJHNPTNZXMI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Z6FP4LKYHDNPPT6WJHNPTNZXMI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:Z6FP4LKYHDNPPT6WJHNPTNZXMI","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":"98a1b57520d1c42afc7cd77692bd2409d5625bc5fa2f976ebb25f327a502902b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-27T06:30:48Z","title_canon_sha256":"25a07a2d4b7d0e1ec48939bc1bad67a6acd7117dec7f6acc23b071c31c15e587"},"schema_version":"1.0","source":{"id":"2505.20777","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.20777","created_at":"2026-07-05T11:10:18Z"},{"alias_kind":"arxiv_version","alias_value":"2505.20777v1","created_at":"2026-07-05T11:10:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.20777","created_at":"2026-07-05T11:10:18Z"},{"alias_kind":"pith_short_12","alias_value":"Z6FP4LKYHDNP","created_at":"2026-07-05T11:10:18Z"},{"alias_kind":"pith_short_16","alias_value":"Z6FP4LKYHDNPPT6W","created_at":"2026-07-05T11:10:18Z"},{"alias_kind":"pith_short_8","alias_value":"Z6FP4LKY","created_at":"2026-07-05T11:10:18Z"}],"graph_snapshots":[{"event_id":"sha256:7ba8d58a0d74ec0384af4255a319f3ce42165909626df817695215cbb208601d","target":"graph","created_at":"2026-07-05T11:10:18Z","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/2505.20777/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"DeepSeek R1 has significantly advanced complex reasoning for large language models (LLMs). While recent methods have attempted to replicate R1's reasoning capabilities in multimodal settings, they face limitations, including inconsistencies between reasoning and final answers, model instability and crashes during long-chain exploration, and low data learning efficiency. To address these challenges, we propose TACO, a novel reinforcement learning algorithm for visual reasoning. Building on Generalized Reinforcement Policy Optimization (GRPO), TACO introduces Think-Answer Consistency, which tigh","authors_text":"Deqiang Jiang, Haoyu Cao, Kun Yin, Qingmin Liao, Wenming Yang, Xinghua Jiang, Xing Sun, Xin Li, Yanlin Liu, Yinsong Liu, Zhehan Kan","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-27T06:30:48Z","title":"TACO: Think-Answer Consistency for Optimized Long-Chain Reasoning and Efficient Data Learning via Reinforcement Learning in LVLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.20777","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":"sha256:6445c2d0cef7eb8f2b7972193e0ff505a175f984a1ac009d091aa449e0e5b75d","target":"record","created_at":"2026-07-05T11:10:18Z","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":"98a1b57520d1c42afc7cd77692bd2409d5625bc5fa2f976ebb25f327a502902b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-27T06:30:48Z","title_canon_sha256":"25a07a2d4b7d0e1ec48939bc1bad67a6acd7117dec7f6acc23b071c31c15e587"},"schema_version":"1.0","source":{"id":"2505.20777","kind":"arxiv","version":1}},"canonical_sha256":"cf8afe2d5838daf7cfd649daf9b737622ebafef34cac30aa610d6684ee2bc81e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cf8afe2d5838daf7cfd649daf9b737622ebafef34cac30aa610d6684ee2bc81e","first_computed_at":"2026-07-05T11:10:18.730911Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:10:18.730911Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8BRFxCBQsFUqo8fOsmZ6I9bX0SWwxMRMrEbn3JAVUlYOdWy5xFEoS3P5nx9reU8yU5ZrMw8GImfNJxwzg5dcBA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:10:18.731618Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.20777","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6445c2d0cef7eb8f2b7972193e0ff505a175f984a1ac009d091aa449e0e5b75d","sha256:7ba8d58a0d74ec0384af4255a319f3ce42165909626df817695215cbb208601d"],"state_sha256":"7cc737a3195993e38e1d4ce2f61cf72ed271c7753383f55f9faf1f66cb734136"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Gvx+wqgtdlIKCcs/UaqmyNPiPT4g7ccx5oSoIzXxlkX9Y+5gXbEN5GcV7BW10iOUk0mX6paY60XlQhl04hjIBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T22:52:36.500786Z","bundle_sha256":"f0cb7895acb89b5f4a80fc81f3148d9beaa4916ef12b0f797209c5a428aee237"}}