{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:H5VPMRH3ZWEZIRVA3JURJ36NSU","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":"c1c2ced4f400f5c1fce21efc76027fb18be7750b51ced415b4e1c73bacb6dbbc","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T17:59:55Z","title_canon_sha256":"402a185441e65509516ec4e2243c4b86cb4d33c0e299ea49cffc3afaf9d513b1"},"schema_version":"1.0","source":{"id":"2605.15198","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15198","created_at":"2026-05-17T21:18:32Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15198v1","created_at":"2026-05-17T21:18:32Z"},{"alias_kind":"pith_short_12","alias_value":"H5VPMRH3ZWEZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"H5VPMRH3ZWEZIRVA","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"H5VPMRH3","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:3eb42b46da8ff91e22ce9751b401a69cf8da82aa516a30e82d3718cd3ab80706","target":"graph","created_at":"2026-05-17T21:57:18Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"ATLAS achieves superior performance on challenging benchmarks while maintaining clear interpretability by using a single discrete functional token that serves as both an agentic operation and a latent visual reasoning unit."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That a single discrete functional token can effectively internalize visual operations without any visual supervision and still generalize across tasks when generated via standard next-token prediction."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"ATLAS uses a single functional token to unify agentic and latent visual reasoning without image generation or external execution."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"One discrete functional token suffices for both agentic operations and latent visual reasoning."}],"snapshot_sha256":"54db4534ab32f0ea5bf9dc95c6c87a3c4b3e69a860ce4d6570c579b438f4a032"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"affdd43ea14ce12d209cda4b098642c2992f038d3c83e22bfc1f7f6cec067d5e"},"paper":{"abstract_excerpt":"Visual reasoning, often interleaved with intermediate visual states, has emerged as a promising direction in the field. A straightforward approach is to directly generate images via unified models during reasoning, but this is computationally expensive and architecturally non-trivial. Recent alternatives include agentic reasoning through code or tool calls, and latent reasoning with learnable hidden embeddings. However, agentic methods incur context-switching latency from external execution, while latent methods lack task generalization and are difficult to train with autoregressive paralleliz","authors_text":"Pheng-Ann Heng, Rain Liu, Xinyan Chen, Ziyu Guo","cross_cats":["cs.AI","cs.CL"],"headline":"One discrete functional token suffices for both agentic operations and latent visual reasoning.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T17:59:55Z","title":"ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both"},"references":{"count":24,"internal_anchors":18,"resolved_work":24,"sample":[{"cited_arxiv_id":"2509.23661","doi":"","is_internal_anchor":true,"ref_index":1,"title":"LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Training","work_id":"41c2802e-aff9-482f-b506-10955ff0838d","year":null},{"cited_arxiv_id":"2511.21631","doi":"","is_internal_anchor":true,"ref_index":2,"title":"Qwen3-VL Technical Report","work_id":"1fe243aa-e3c0-4da6-b391-4cbcfc88d5c0","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Anole: An open, autoregressive, native large multimodal models for interleaved image-text generation","work_id":"31ce9d99-2071-41a0-9f51-51b8c5e3ba7e","year":null},{"cited_arxiv_id":"2507.06261","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","work_id":"008df105-2fdd-45d8-857a-8e35868aecb6","year":null},{"cited_arxiv_id":"2505.14683","doi":"","is_internal_anchor":true,"ref_index":5,"title":"Emerging Properties in Unified Multimodal Pretraining","work_id":"e0cfd82c-f5d4-44fd-b531-ec73ab0a805b","year":null}],"snapshot_sha256":"59531425264020a4256c48ab840b47299f63cfc2a769c710677630731a0adee2"},"source":{"id":"2605.15198","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T03:05:18.872222Z","id":"9f14be86-a865-4e9e-9c3a-131216c39a5d","model_set":{"reader":"grok-4.3"},"one_line_summary":"ATLAS uses a single functional token to unify agentic and latent visual reasoning without image generation or external execution.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"One discrete functional token suffices for both agentic operations and latent visual reasoning.","strongest_claim":"ATLAS achieves superior performance on challenging benchmarks while maintaining clear interpretability by using a single discrete functional token that serves as both an agentic operation and a latent visual reasoning unit.","weakest_assumption":"That a single discrete functional token can effectively internalize visual operations without any visual supervision and still generalize across tasks when generated via standard next-token prediction."}},"verdict_id":"9f14be86-a865-4e9e-9c3a-131216c39a5d"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b1e69a5d45fb3bc5bcc8888ba5e6ab104818370ba8f83b8d5443fc564b0ed3b4","target":"record","created_at":"2026-05-17T21:18:32Z","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":"c1c2ced4f400f5c1fce21efc76027fb18be7750b51ced415b4e1c73bacb6dbbc","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T17:59:55Z","title_canon_sha256":"402a185441e65509516ec4e2243c4b86cb4d33c0e299ea49cffc3afaf9d513b1"},"schema_version":"1.0","source":{"id":"2605.15198","kind":"arxiv","version":1}},"canonical_sha256":"3f6af644fbcd899446a0da6914efcd951095b5d33afb0a260046634726373acb","receipt":{"builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3f6af644fbcd899446a0da6914efcd951095b5d33afb0a260046634726373acb","first_computed_at":"2026-05-17T21:40:24.984558Z","kind":"pith_receipt","last_reissued_at":"2026-05-17T21:57:18.377459Z","receipt_version":"0.2","signature_status":"unsigned_v0"},"source_id":"2605.15198","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b1e69a5d45fb3bc5bcc8888ba5e6ab104818370ba8f83b8d5443fc564b0ed3b4","sha256:3eb42b46da8ff91e22ce9751b401a69cf8da82aa516a30e82d3718cd3ab80706"],"state_sha256":"b8029cae22b36a5d6c8322a83980864772af06314ce498d31c57f496eb625635"}