{"as_of":"2026-08-07T21:50:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:3954f48788c599658807017c79ac28cf6d1bf25aa9ea0bf3d6a077dafc7f5a93","coverage":[{"denominator":31,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":31,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-06T21:32:23.852683Z","state":"measured"},{"denominator":33,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":33,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-07T06:34:17.273281+00:00","state":"measured"},{"denominator":2,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":2,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-06T15:06:48.889429Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-05-20T05:53:22.253104Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2506.23940","snapshot_observed_at":"2026-08-06T15:06:48.889429Z","title":"Graft: Integrating the domain knowledge via efficient parameter synergy for mllms","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2507.16731","last_updated":"2025-07-22T16:13:43Z","snapshot_observed_at":"2026-08-06T15:00:33.300567Z","submitted_at":"2025-07-22T16:13:43Z","title":"Collaborative Inference and Learning between Edge SLMs and Cloud LLMs: A Survey of Algorithms, Execution, and Open Challenges","version":1},"reference_index":277,"source":"pdf_text","source_observed_at":"2026-08-06T15:06:48.889429Z"},"links":{"cited_paper":"/paper/2506.23940","citing_paper":"/paper/2507.16731"},"observation_digest":"sha256:01263a847601a32b397c17d45833d285076e0e841c23b6bb640db084cab42a60","observation_id":"1dee66f4-3534-4c46-8d85-686f1c3457e2","resolution":{"observed_at":"2026-08-06T15:06:48.889429Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"cited_work":{"arxiv_id":"2506.23940","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2506.23940","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":null,"venue":null,"work_id":"e63f1368-f100-4f7a-bfb9-f4e8af0d109b","year":2025},"citing_paper":{"arxiv_id":"2605.19559","last_updated":"2026-05-19T09:02:20Z","snapshot_observed_at":"2026-07-06T23:30:16.094579Z","submitted_at":"2026-05-19T09:02:20Z","title":"EgoCoT-Bench: Benchmarking Grounded and Verifiable Operation-Centric Chain of Thought Reasoning for MLLMs","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-05-20T05:53:05.450946Z"},"links":{"cited_paper":"/paper/2506.23940","citing_paper":"/paper/2605.19559"},"observation_digest":"sha256:2f40553e3a20e0e63321f4517dd50db5d2ca0ed78f9f71fb0861eaa167377d43","observation_id":"25a26b1f-bdf5-4676-8aea-b04e09b847ba","resolution":{"observed_at":"2026-05-20T05:53:22.255505Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2506.23940/citation-record","integrity":"/paper/2506.23940/integrity","json":"/paper/2506.23940/citation-record.json","paper":"/paper/2506.23940"},"outbound":[{"citation":{"cited_paper":{"arxiv_id":"2409.12186","last_updated":"2024-11-12T13:24:25Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-09-18T17:57:57Z","title":"Qwen2.5-Coder Technical Report","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2409.12186","snapshot_observed_at":"2026-08-06T21:32:23.756344Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.756344Z"},"links":{"cited_paper":"/paper/2409.12186","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:e942f7fac191e6df35793091a31ec30be2d662624d63125455781b449a2bfad3","observation_id":"c2e55b09-be3f-4b28-b287-aa728940ea79","resolution":{"observed_at":"2026-08-06T21:32:23.756344Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.06470","last_updated":"2025-03-09T06:14:17Z","snapshot_observed_at":"2026-08-07T17:19:24.307173Z","submitted_at":"2025-03-09T06:14:17Z","title":"Think Twice, Click Once: Enhancing GUI Grounding via Fast and Slow Systems","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.06470","snapshot_observed_at":"2026-08-06T21:32:23.760575Z","title":"Think twice, click once: Enhancing gui grounding via fast and slow systems","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.760575Z"},"links":{"cited_paper":"/paper/2503.06470","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:80c357467f67d4ce1099897e81d0b71bf113a10d69a66de91d38dc0574bb5cb5","observation_id":"fefd5596-0887-4480-8620-f09adc20bb2f","resolution":{"observed_at":"2026-08-06T21:32:23.760575Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2408.07666","last_updated":"2025-12-31T04:06:49Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-08-14T16:58:48Z","title":"Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2408.07666","snapshot_observed_at":"2026-08-06T21:32:23.764650Z","title":"Model merging in llms, mllms, and beyond: Methods, theories, applications and opportunities","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.764650Z"},"links":{"cited_paper":"/paper/2408.07666","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:112436392c95640f73fe11031932888666714a1b8134b3b67d9def9e6d74ef16","observation_id":"d67f1f5e-9cb9-4bb6-807d-c69d893e9978","resolution":{"observed_at":"2026-08-06T21:32:23.764650Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2309.15698","last_updated":"2023-09-27T14:40:12Z","snapshot_observed_at":"2026-08-04T10:16:25.449869Z","submitted_at":"2023-09-27T14:40:12Z","title":"Deep Model Fusion: A Survey","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.15698","snapshot_observed_at":"2026-08-06T21:32:23.768843Z","title":"Deep model fusion: A survey","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.768843Z"},"links":{"cited_paper":"/paper/2309.15698","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:22a911bf5f819eb314ab99309c8b10c6b7f11093f1a6e04972c5a9f9467a06f8","observation_id":"441d0559-8928-43c9-a974-f4360e20a9da","resolution":{"observed_at":"2026-08-06T21:32:23.768843Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2501.07596","last_updated":"2025-03-03T13:27:01Z","snapshot_observed_at":"2026-07-06T20:20:33.068936Z","submitted_at":"2025-01-10T01:42:43Z","title":"Optimize Incompatible Parameters through Compatibility-aware Knowledge Integration","version":2},"cited_work":{"arxiv_id":"2501.07596","doi":null,"metadata_source":"pith","pith_arxiv_id":"2501.07596","snapshot_observed_at":"2026-08-06T21:32:23.994335Z","title":"Optimize Incompatible Parameters through Compatibility-aware Knowledge Integration","venue":"cs.LG","work_id":"20aa816e-ff12-4bfd-9836-e8ec69b48e13","year":2025},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.772900Z"},"links":{"cited_paper":"/paper/2501.07596","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:5eb072d27a1b250c47fac53769ca1e95bf73473feed9f8af4081ab3fa2d02ccd","observation_id":"b3f93aa0-927c-4011-8c47-14b2835ec6f8","resolution":{"observed_at":"2026-08-06T21:32:23.998053Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2212.04089","last_updated":"2023-03-31T15:27:01Z","snapshot_observed_at":"2026-08-03T22:12:27.993418Z","submitted_at":"2022-12-08T05:50:53Z","title":"Editing Models with Task Arithmetic","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2212.04089","snapshot_observed_at":"2026-08-06T21:32:23.776181Z","title":"Editing models with task arithmetic","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.776181Z"},"links":{"cited_paper":"/paper/2212.04089","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:677b251b0578d6b5a139eb763c8c1961edfc93a4d7597a239eb4c5642d6082e2","observation_id":"3aa019e0-e9c3-4517-82cc-2f3e43fb1aea","resolution":{"observed_at":"2026-08-06T21:32:23.776181Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T21:32:24.208610Z","title":"Duet: A tuning-free device-cloud collaborative parameters generation framework for efficient device model generalization","venue":null,"work_id":"40b647e6-35a2-4d43-b59b-e0b4e4ad9cbe","year":2023},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.780104Z"},"links":{"citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:141e71c6046e022279b625e212a4627c86d17a5e5fa2721a8e96edeb77c3a8f1","observation_id":"10fbc0c2-422a-4352-8ac0-6ce6b3ae4b89","resolution":{"observed_at":"2026-08-06T21:32:24.212625Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T21:32:24.197090Z","title":"Intelligent model update strategy for sequential recommendation","venue":null,"work_id":"fd92eee5-8f28-4130-a4c0-ff1972f9c51c","year":2024},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.783469Z"},"links":{"citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:64ac3e9a99cd01a1c1f2f7853b601bcdb97ec5ddf09088e500fc42ae14c59ed0","observation_id":"3cb8c70d-c601-46d4-8c88-b28b5a7708f1","resolution":{"observed_at":"2026-08-06T21:32:24.201369Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2406.12608","last_updated":"2024-10-14T06:44:14Z","snapshot_observed_at":"2026-07-06T18:32:57.295918Z","submitted_at":"2024-06-18T13:35:25Z","title":"Bridging Local Details and Global Context in Text-Attributed Graphs","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.12608","snapshot_observed_at":"2026-08-06T21:32:23.787390Z","title":"Bridging local details and global context in text-attributed graphs","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.787390Z"},"links":{"cited_paper":"/paper/2406.12608","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:7d3a5214d896f1e4bd1387ff15c4e7291df248be2b420e97f1fb35462bbc2a63","observation_id":"f033b2ce-8990-4934-b19d-3da318089dd4","resolution":{"observed_at":"2026-08-06T21:32:23.787390Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T21:32:23.791816Z","title":"Bert: Pre-training of deep bidirectional transformers for language understanding","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.791816Z"},"links":{"citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:6e15143acaca671e69ac3042da722da0603440a64b8be2b8233ee3cd876349f2","observation_id":"cd66d592-4414-4ccd-9d70-c434bfc77eee","resolution":{"observed_at":"2026-08-06T21:32:23.791816Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T21:32:24.182145Z","title":"Improving language understanding by generative pre-training.(2018),","venue":null,"work_id":"f3c0b8d2-a26e-411b-b69b-59aeb9b8d981","year":2018},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.794970Z"},"links":{"citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:9570d5cf68f5bc3d260d29fc65f816565d454b464faa1ba560635397d4e735c4","observation_id":"ef1d4479-9480-4626-adf0-f2f52c9b9949","resolution":{"observed_at":"2026-08-06T21:32:24.185780Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2104.08691","last_updated":"2021-09-02T17:34:41Z","snapshot_observed_at":"2026-08-06T15:24:34.790850Z","submitted_at":"2021-04-18T03:19:26Z","title":"The Power of Scale for Parameter-Efficient Prompt Tuning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2104.08691","snapshot_observed_at":"2026-08-06T21:32:23.799228Z","title":"The power of scale for parameter-efficient prompt tuning","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.799228Z"},"links":{"cited_paper":"/paper/2104.08691","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:fbe685f57b5cc369f16fa2e1d85be059959c5ef520bc1dea048958ed25eabf0a","observation_id":"7eeb99bc-9b95-4314-8b29-0d1ddf720893","resolution":{"observed_at":"2026-08-06T21:32:23.799228Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-03T22:14:53.516423Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-06T21:32:23.803598Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.803598Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:6042fa3ca4ca67ba716a775950853a72871b231c0451976dfce0ceda311eda1b","observation_id":"6fd8c473-3fda-48ae-81fc-ca8731ef683a","resolution":{"observed_at":"2026-08-06T21:32:23.803598Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2212.09849","last_updated":"2025-05-21T23:53:00Z","snapshot_observed_at":"2026-08-07T00:34:34.195906Z","submitted_at":"2022-12-19T20:46:43Z","title":"Dataless Knowledge Fusion by Merging Weights of Language Models","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2212.09849","snapshot_observed_at":"2026-08-06T21:32:23.806970Z","title":"Dataless knowledge fusion by merging weights of language models","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.806970Z"},"links":{"cited_paper":"/paper/2212.09849","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:7f2b4a602992a55e3e5a59621295a4eb42c435988e7b194d000c82530977a4ae","observation_id":"70ea454b-bf9e-44c1-90aa-86359ba138ae","resolution":{"observed_at":"2026-08-06T21:32:23.806970Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.02575","last_updated":"2024-05-28T06:42:31Z","snapshot_observed_at":"2026-08-05T09:56:02.040968Z","submitted_at":"2023-10-04T04:26:33Z","title":"AdaMerging: Adaptive Model Merging for Multi-Task Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.02575","snapshot_observed_at":"2026-08-06T21:32:23.810086Z","title":"Adamerging: Adaptive model merging for multi-task learning","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.810086Z"},"links":{"cited_paper":"/paper/2310.02575","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:4921d9055cb6572a41e7aa7c90ed002be0913e44397d2e097bf29278c4e816db","observation_id":"39e185ef-e9c3-4050-bc06-9cf17268d1e3","resolution":{"observed_at":"2026-08-06T21:32:23.810086Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T21:32:24.170512Z","title":"Mart: Learning hierarchical music audio representations with part-whole transformer","venue":null,"work_id":"30cc0dca-3586-4ea2-89e2-613e96a77e4f","year":2024},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.814293Z"},"links":{"citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:593c496f8dfce28d015489104c653275f5b394b18995c9169ed83c7261dadcc8","observation_id":"aac5b850-3d41-4e71-88fa-f19810a17c00","resolution":{"observed_at":"2026-08-06T21:32:24.175130Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2506.05287","last_updated":"2025-06-05T17:44:12Z","snapshot_observed_at":"2026-08-07T10:19:21.034875Z","submitted_at":"2025-06-05T17:44:12Z","title":"EOC-Bench: Can MLLMs Identify, Recall, and Forecast Objects in an Egocentric World?","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2506.05287","snapshot_observed_at":"2026-08-06T21:32:23.818358Z","title":"Eoc-bench: Can mllms identify, recall, and forecast objects in an egocentric world? arXiv preprint arXiv:2506.05287, 2025a","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.818358Z"},"links":{"cited_paper":"/paper/2506.05287","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:67d08f10637af2c9d2a1f87cf10fc0e56b684f88dbc30f49947fa6787dfaac26","observation_id":"e4308293-216f-4ced-bbec-7e0242a34670","resolution":{"observed_at":"2026-08-06T21:32:23.818358Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.19978","last_updated":"2024-12-28T02:36:51Z","snapshot_observed_at":"2026-07-06T20:13:58.813488Z","submitted_at":"2024-12-28T02:36:51Z","title":"MAKIMA: Tuning-free Multi-Attribute Open-domain Video Editing via Mask-Guided Attention Modulation","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.19978","snapshot_observed_at":"2026-08-06T21:32:23.821548Z","title":"Makima: Tuning-free multi-attribute open- domain video editing via mask-guided attention modulation","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.821548Z"},"links":{"cited_paper":"/paper/2412.19978","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:f653f8e14dcbefc8cc43d6bf01bc4c9d5fb578e53d177be479bb3829c4485ef0","observation_id":"6c7b1b8a-6327-4ecd-b6dc-77ded13a4995","resolution":{"observed_at":"2026-08-06T21:32:23.821548Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.19684","last_updated":"2025-02-17T15:32:50Z","snapshot_observed_at":"2026-07-06T20:13:45.123960Z","submitted_at":"2024-12-27T15:21:17Z","title":"Boosting Private Domain Understanding of Efficient MLLMs: A Tuning-free, Adaptive, Universal Prompt Optimization Framework","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.19684","snapshot_observed_at":"2026-08-06T21:32:23.825194Z","title":"Boosting private domain understanding of efficient mllms: A tuning-free, adaptive, universal prompt optimization framework","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.825194Z"},"links":{"cited_paper":"/paper/2412.19684","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:01510a92fab7a93da1d170f99c1dabd8d80cf68f51b475f6677f92c18885d129","observation_id":"29eb50a3-4367-4532-94f5-39646381daf0","resolution":{"observed_at":"2026-08-06T21:32:23.825194Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.01915","last_updated":"2023-05-03T06:14:31Z","snapshot_observed_at":"2026-08-07T15:32:31.672752Z","submitted_at":"2023-05-03T06:14:31Z","title":"Denoising Multi-modal Sequential Recommenders with Contrastive Learning","version":1},"cited_work":{"arxiv_id":"2305.01915","doi":null,"metadata_source":"pith","pith_arxiv_id":"2305.01915","snapshot_observed_at":"2026-08-06T21:32:23.911566Z","title":"Denoising Multi-modal Sequential Recommenders with Contrastive Learning","venue":"cs.IR","work_id":"e743131d-5d11-4f48-9d19-cc04411a9f5c","year":2023},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.829484Z"},"links":{"cited_paper":"/paper/2305.01915","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:83bd278f9970ed9006581be2b410fd28a0d95544443e8adf99b2a3d832b1ead0","observation_id":"06b4bad3-1ee3-441d-a3ea-6e65ad239507","resolution":{"observed_at":"2026-08-06T21:32:23.916950Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T21:32:24.159865Z","title":"Multimodal large language models: A survey","venue":null,"work_id":"00b2c8e9-321f-4410-add5-683bd4ff4577","year":2023},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.832964Z"},"links":{"citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:bb5390596c8793124d400a369d439bd7b78a7e87bc05cf2b3ebb89d8fbd81ead","observation_id":"9d0eeaf6-c37a-4a00-a772-42ebfd60eaee","resolution":{"observed_at":"2026-08-06T21:32:24.163319Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2401.13601","last_updated":"2024-05-28T05:36:23Z","snapshot_observed_at":"2026-07-06T17:20:00.101661Z","submitted_at":"2024-01-24T17:10:45Z","title":"MM-LLMs: Recent Advances in MultiModal Large Language Models","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2401.13601","snapshot_observed_at":"2026-08-06T21:32:23.836684Z","title":"Mm-llms: Recent advances in multimodal large language models","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.836684Z"},"links":{"cited_paper":"/paper/2401.13601","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:76bba3294cae8be96262b218b146c4ae162c6cd68b9ec15c13a5b22119cebdac","observation_id":"84c9c7fb-6905-4346-9de0-ef4d6511f7f9","resolution":{"observed_at":"2026-08-06T21:32:23.836684Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2309.17421","last_updated":"2023-10-11T05:07:37Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2023-09-29T17:34:51Z","title":"The Dawn of LMMs: Preliminary Explorations with GPT-4V(ision)","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.17421","snapshot_observed_at":"2026-08-06T21:32:23.840773Z","title":"The dawn of lmms: Preliminary explorations with gpt-4v (ision)","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.840773Z"},"links":{"cited_paper":"/paper/2309.17421","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:17a880a1e43353cbf6f6e9bbfc3c1e0831625d1f3abc5b61a94fc5d4109ea266","observation_id":"24963130-d5b9-47ec-8dd2-c8c0d851dafa","resolution":{"observed_at":"2026-08-06T21:32:23.840773Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.17294","last_updated":"2024-10-08T06:58:27Z","snapshot_observed_at":"2026-08-04T00:41:59.024387Z","submitted_at":"2024-06-25T05:43:21Z","title":"Math-LLaVA: Bootstrapping Mathematical Reasoning for Multimodal Large Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.17294","snapshot_observed_at":"2026-08-06T21:32:23.844440Z","title":"Math-llava: Bootstrapping mathematical reasoning for multimodal large language models","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.844440Z"},"links":{"cited_paper":"/paper/2406.17294","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:88c1c3e116e85008f72886bb3da7b140455ac17e7a02fae6867eb0e0ca504836","observation_id":"e92b9496-f065-4349-a926-08d18810bccf","resolution":{"observed_at":"2026-08-06T21:32:23.844440Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2409.12191","last_updated":"2024-10-03T15:54:49Z","snapshot_observed_at":"2026-08-06T05:35:29.109022Z","submitted_at":"2024-09-18T17:59:32Z","title":"Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2409.12191","snapshot_observed_at":"2026-08-06T21:32:23.852683Z","title":"Qwen2-vl: Enhancing vision-language model’s perception of the world at any resolution, 2024d","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.852683Z"},"links":{"cited_paper":"/paper/2409.12191","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:2ccffad69ef128b0cb9ee08be5638e4dcd35328cebe5e0896a03aff6ae56555c","observation_id":"4abe0bf8-b72c-42d2-8bff-02e4827374a1","resolution":{"observed_at":"2026-08-06T21:32:23.852683Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2203.08013","last_updated":"2022-05-23T11:46:10Z","snapshot_observed_at":"2026-08-03T15:06:12.409919Z","submitted_at":"2022-03-15T15:50:45Z","title":"End-to-End Modeling via Information Tree for One-Shot Natural Language Spatial Video Grounding","version":2},"cited_work":{"arxiv_id":"2203.08013","doi":null,"metadata_source":"pith","pith_arxiv_id":"2203.08013","snapshot_observed_at":"2026-08-06T21:32:24.148238Z","title":"End-to-End Modeling via Information Tree for One-Shot Natural Language Spatial Video Grounding","venue":"cs.CV","work_id":"4423a013-e1ff-465c-b03d-a0f0c9a05957","year":2022},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":2019,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.732034Z"},"links":{"cited_paper":"/paper/2203.08013","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:50115e1d69067c5586838044f984ba224c96b67c9e8ec418b3a9fd0ac95adedf","observation_id":"0055d29d-44b1-4e8c-8dfa-46b75906f816","resolution":{"observed_at":"2026-08-06T21:32:24.152856Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2003.10286","last_updated":"2020-03-07T17:55:41Z","snapshot_observed_at":"2026-07-06T09:06:42.071993Z","submitted_at":"2020-03-07T17:55:41Z","title":"PathVQA: 30000+ Questions for Medical Visual Question Answering","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2003.10286","snapshot_observed_at":"2026-08-06T21:32:23.848820Z","title":"Sujet AI","venue":null,"work_id":null,"year":2003},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":2020,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.848820Z"},"links":{"cited_paper":"/paper/2003.10286","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:45afe745663083f1664f573c2aaedb70c9f680bf818b00215f848568a1e309a1","observation_id":"a937b6f7-2bc8-4723-96c2-2aa5d0727d44","resolution":{"observed_at":"2026-08-06T21:32:23.848820Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2312.16831","last_updated":"2023-12-28T05:09:31Z","snapshot_observed_at":"2026-07-06T17:09:06.429946Z","submitted_at":"2023-12-28T05:09:31Z","title":"METER: A Dynamic Concept Adaptation Framework for Online Anomaly Detection","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2312.16831","snapshot_observed_at":"2026-08-06T21:32:23.736753Z","title":"Relational graph learning for grounded video description generation","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":2021,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.736753Z"},"links":{"cited_paper":"/paper/2312.16831","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:a45c1ecb0cdd90542cbdf1ca35002197a6b5803af0421841b78153d268416bc3","observation_id":"515a5ab5-4716-4b28-bc7f-247924f1171a","resolution":{"observed_at":"2026-08-06T21:32:23.736753Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2502.09838","last_updated":"2025-02-21T17:39:29Z","snapshot_observed_at":"2026-08-07T20:16:32.175759Z","submitted_at":"2025-02-14T00:42:36Z","title":"HealthGPT: A Medical Large Vision-Language Model for Unifying Comprehension and Generation via Heterogeneous Knowledge Adaptation","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2502.09838","snapshot_observed_at":"2026-08-06T21:32:23.749088Z","title":"Healthgpt: A medical large vision-language model for unifying comprehension and generation via heterogeneous knowledge adaptation","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":2022,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.749088Z"},"links":{"cited_paper":"/paper/2502.09838","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:411b66708798d80f4f8c117c99dd00399f6ff4dc5cbd0d3485f1e75c2df4a84f","observation_id":"bd69d292-3edb-4c3a-bc79-a689ea57871e","resolution":{"observed_at":"2026-08-06T21:32:23.749088Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T21:32:24.219835Z","title":"Minerva: Solving quantitative reasoning problems with language models","venue":null,"work_id":"894f0488-1c93-47af-a065-73f460142188","year":2022},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":2024,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.744747Z"},"links":{"citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:1ac88cde0d71905b5cbca3224aaed4f3da69bc95b9df8c024a4fe37884f1c31c","observation_id":"5f0da28e-8ecc-46ae-b530-63a178e8b090","resolution":{"observed_at":"2026-08-06T21:32:24.226426Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T21:32:23.752982Z","title":"Fin-r1: A large language model for financial reasoning through reinforcement learning","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":2025,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.752982Z"},"links":{"citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:77e2ba774cf31217c3a634af616e94128853bb527ed4b6efd7f0c9c28b28f0c2","observation_id":"9f2213b6-d442-46db-8de6-44a9160e38da","resolution":{"observed_at":"2026-08-06T21:32:23.752982Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","latest_version":2,"primary_category":"cs.CL","snapshot_observed_at":"2026-08-06T21:25:30.246335Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs"},"reference_resolution":{"displayed":31,"state_counts":{"malformed_identifier":0,"metadata_mismatch":2,"parse_uncertain":0,"unresolved":22,"verified_exact":1,"verified_fuzzy":6},"total_outbound_references":31},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"thesis":"As of 7 August 2026, this Paper Citation Record lists 31 of 31 outbound references and 2 inbound Pith citation observations for arXiv:2506.23940."}