{"as_of":"2026-08-08T04:49:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:c36dc34da3c70586ef4ae7a9d4a813e606392969a32c90622b1d87f518ce183e","coverage":[{"denominator":50,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":50,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-07T13:20:20.761016Z","state":"measured"},{"denominator":50,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":50,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-07T06:34:17.273281+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2505.22194/citation-record","integrity":"/paper/2505.22194/integrity","json":"/paper/2505.22194/citation-record.json","paper":"/paper/2505.22194"},"outbound":[{"citation":{"cited_paper":{"arxiv_id":"2401.03868","last_updated":"2024-01-09T06:47:46Z","snapshot_observed_at":"2026-07-06T17:12:47.723473Z","submitted_at":"2024-01-08T13:00:53Z","title":"FlightLLM: Efficient Large Language Model Inference with a Complete Mapping Flow on FPGAs","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2401.03868","snapshot_observed_at":"2026-08-07T13:20:14.616944Z","title":"Flightllm: Efficient large language model inference with a complete mapping flow on fpga,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:14.616944Z"},"links":{"cited_paper":"/paper/2401.03868","citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:8ddff3493332dc641c2a5e4ca442847a120afd02bab3e24c6cff47105fb44350","observation_id":"b1d80f47-8063-4e5b-b245-7cadb6bc7630","resolution":{"observed_at":"2026-08-07T13:20:14.616944Z","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":"2026-08-07T13:20:30.263262Z","title":"Heatvit: Hardware-efficient adaptive token pruning for vision transformers,","venue":null,"work_id":"1e00f40d-7028-4769-807d-ae879cf22b14","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:14.731586Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:6bc56f03afbe6258592d47e7270b81d009281743f1898321107296b346bccf37","observation_id":"93b91800-a221-433c-8f8a-9e5e5c9aaa6e","resolution":{"observed_at":"2026-08-07T13:20:30.314286Z","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":"2026-08-07T13:20:30.113342Z","title":"Auto-vit-acc: An fpga-aware automatic acceleration framework for vision transformer with mixed-scheme quantization,","venue":null,"work_id":"c1a22c85-340e-4d58-8b9a-0e66a2d5cb84","year":2022},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:14.874750Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:604c2fd5e2246209024f47375ebf6d0d9327b4a26c9664bcea03bccdf97b9fac","observation_id":"7aaed175-22b6-412c-8903-6606231d4add","resolution":{"observed_at":"2026-08-07T13:20:30.170844Z","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":"2010.11929","last_updated":"2021-06-03T13:08:56Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2020-10-22T17:55:59Z","title":"An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2010.11929","snapshot_observed_at":"2026-08-07T13:20:14.991420Z","title":"An image is worth 16x16 words: Transformers for image recognition at scale,","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:14.991420Z"},"links":{"cited_paper":"/paper/2010.11929","citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:f45772a8705a41505d24e5fc0bc3f8eb0a699a4ae2757f869ec072263218254b","observation_id":"d0500b1e-db1b-4f46-b68b-3db1a0a2fd1c","resolution":{"observed_at":"2026-08-07T13:20:14.991420Z","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":"2026-08-07T13:20:29.908885Z","title":"Sda: Low-bit stable diffusion acceleration on edge fpgas,","venue":null,"work_id":"528a0324-3d87-49fe-a1ac-a198cf447a4d","year":2024},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:15.104008Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:b4faa1ef9a63a89ce7bcd0821bf5a82c1b167923647a2244e014efcd3639bd1d","observation_id":"24584f6a-7a90-4a42-861b-0a2eeb3d912f","resolution":{"observed_at":"2026-08-07T13:20:29.989036Z","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":"2026-08-07T13:20:29.660530Z","title":"Pushing the limits of narrow precision inferencing at cloud scale with microsoft floating point,","venue":null,"work_id":"16d13858-8f15-4f13-8e6b-b71da70f8bc5","year":2020},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:15.208767Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:9ff1d297c78da5bac3773bd3a73d71c0c8ba26f3bb44661e1609b5519b1dbc3e","observation_id":"34bc6a8f-ad87-4d52-b725-5876218dc734","resolution":{"observed_at":"2026-08-07T13:20:29.797651Z","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":"2026-08-07T13:20:29.471149Z","title":"Revisiting block-based quantisation: What is important for sub-8-bit llm inference?,","venue":null,"work_id":"67d9b714-b4ce-4941-9d37-2da4c45123ec","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:15.334740Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:59e5a3d39465ab820cf71841f90c6f6154bd1d95fc01cb6d66be517a1101f7a7","observation_id":"f2c2aef5-7751-4aa4-9ee8-0c8182264bd5","resolution":{"observed_at":"2026-08-07T13:20:29.578285Z","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":"2026-08-07T13:20:29.233261Z","title":"Ex- ploring fpga designs for mx and beyond,","venue":null,"work_id":"60380730-efbb-4b79-8ddc-beec6a93f157","year":2024},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:15.411747Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:effc5971475c7571feb07d9aa832ccf57fd54c18e75b0b1dd5333cfdcc419bda","observation_id":"d83a95f4-5b83-4428-9d78-aceca25b0058","resolution":{"observed_at":"2026-08-07T13:20:29.316895Z","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":"2026-08-07T13:20:29.139413Z","title":"An integer- only and group-vector systolic accelerator for efficiently mapping vision transformer on edge,","venue":null,"work_id":"8a7ae081-fb4e-4f09-99a3-4396c1703a1f","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:15.542561Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:5fc9ce2ca9f9c814ea99b31d4e89cf0b8f22a8a1e4094eef7395ec061f37dd0f","observation_id":"92390164-b325-43d4-9eb4-ce62abb587b3","resolution":{"observed_at":"2026-08-07T13:20:29.180303Z","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":"2026-08-07T13:20:28.881265Z","title":"Deit iii: Revenge of the vit,","venue":null,"work_id":"47a74a35-9304-4dec-8d76-97c27ead8236","year":2022},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:15.647778Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:460b3a539673f3ff182928b9be027058d74eab1875f99639778021dfe004f7ad","observation_id":"810b85f5-3e70-4806-a317-fbdd153a705e","resolution":{"observed_at":"2026-08-07T13:20:28.975518Z","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":"2026-08-07T13:20:15.724754Z","title":"Imagenet: A large-scale hierarchical image database,","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:15.724754Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:f8a5abcd96a748cfd3fbc67aa08eb82e2c0de8d9d0f8d6d5f0351d501da97214","observation_id":"09fa846a-e671-4ea7-98cb-3080e7e18a93","resolution":{"observed_at":"2026-08-07T13:20:15.724754Z","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":"2026-08-07T13:20:28.661635Z","title":"Ieee standard 754 for binary floating-point arithmetic,","venue":null,"work_id":"87f97547-e479-4f01-9be7-2ed80b718e01","year":1996},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:15.852918Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:f52ed4a83bd7eaa7c0f79e3c84182cd1ddf35a39b37c1ed370d8f7de1f0d1f8e","observation_id":"3b193b75-27c2-428f-84dc-4203107c1238","resolution":{"observed_at":"2026-08-07T13:20:28.773838Z","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":"2026-08-07T13:20:28.473300Z","title":"Understanding the potential of fpga-based spatial accel- eration for large language model inference,","venue":null,"work_id":"54d472d0-fc93-49c6-a9d4-6bf26fefdb32","year":2024},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:15.970286Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:8eaca04addac50e724b2e8805e3024da5cdbe6c23c757cf6c34df87d4162a188","observation_id":"f4b21b32-3869-4dd1-9650-3dbcf1bd2d66","resolution":{"observed_at":"2026-08-07T13:20:28.522640Z","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":"2026-08-07T13:20:28.312469Z","title":"fpgaconvnet: A framework for map- ping convolutional neural networks on fpgas,","venue":null,"work_id":"f6354d2d-afe1-403a-a12e-266c32722123","year":2016},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:16.066098Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:76d529fb6226966d61bcfe6d63576bbeab6f12e49dbaedac004ed417fb8dae4e","observation_id":"713475d0-1cf2-4cd2-b937-fb329f6e340c","resolution":{"observed_at":"2026-08-07T13:20:28.379137Z","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":"2026-08-07T13:20:28.030615Z","title":"Finn: A framework for fast, scalable binarized neural network inference,","venue":null,"work_id":"6da5a7f5-21f6-4a94-9469-5562c3c92e53","year":2017},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:16.165595Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:782dac78bb4462a58cef9dde9a234aae2f3b202218817eff418340d497b391c3","observation_id":"546c268d-d9e4-49b9-bdb6-e8bffe9661f9","resolution":{"observed_at":"2026-08-07T13:20:28.145064Z","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":"2311.03379","last_updated":"2023-11-02T02:38:45Z","snapshot_observed_at":"2026-08-03T06:06:31.382272Z","submitted_at":"2023-11-02T02:38:45Z","title":"HIDA: A Hierarchical Dataflow Compiler for High-Level Synthesis","version":1},"cited_work":{"arxiv_id":"2311.03379","doi":null,"metadata_source":"pith","pith_arxiv_id":"2311.03379","snapshot_observed_at":"2026-08-07T13:20:21.247989Z","title":"HIDA: A Hierarchical Dataflow Compiler for High-Level Synthesis","venue":"cs.AR","work_id":"d83f93a7-2147-4e22-867e-7b0c23686fbd","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:16.362603Z"},"links":{"cited_paper":"/paper/2311.03379","citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:de8d497bfa307a106b561bfb9221537aaa0abebda1c3f1deafcf9be99a3d5008","observation_id":"3d543e4f-ace0-4ed0-98e5-538e57ba6ede","resolution":{"observed_at":"2026-08-07T13:20:21.362760Z","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":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-07T13:20:27.728675Z","title":"Automating constraint- aware datapath optimization using e-graphs,","venue":null,"work_id":"799d9811-de6d-485c-b788-f626d7e38e31","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:16.441786Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:15fd6790a08cb160a991b34eb1f89c1bdf25fdc50dda8ea8e2100becfdcfaa87","observation_id":"a15371d6-aec2-49f4-9ca4-423393c3170c","resolution":{"observed_at":"2026-08-07T13:20:27.892641Z","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":"2026-08-07T13:20:27.555107Z","title":"Online alignment and ad- dition in multiterm floating-point adders,","venue":null,"work_id":"ca7f86d2-0592-4636-8b6d-2e20382d79c2","year":2024},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:16.581136Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:6855922ca2b3a7ff0bf73c578ad8f5f6a0a3e586f6ccffd606f97f5cb9750923","observation_id":"408b7846-7e39-491a-a5f5-f82ddb2772f1","resolution":{"observed_at":"2026-08-07T13:20:27.628940Z","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":"2026-08-07T13:20:27.362409Z","title":null,"venue":null,"work_id":"d525fcbf-45fb-4975-b530-01676839d8ec","year":2025},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:16.734576Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:3837452431338ed8536ed1a552337d6671b7b2deffd03bd60fe83d7b4005f108","observation_id":"c88b84c8-2f4b-47d6-8ed8-3627c6249412","resolution":{"observed_at":"2026-08-07T13:20:27.451348Z","resolver_source":"raw_fallback","status":"parse_uncertain"},"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":"2026-08-07T13:20:27.112124Z","title":null,"venue":null,"work_id":"780a4c6f-bde0-4888-9efc-f38b003bae8e","year":2025},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:16.817200Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:5380a59a9fa5f35a78c2608a7aba1850d9ec54f3ef0b978c769974889f53ed5c","observation_id":"59ad3361-e666-4df0-b4ca-254189269fd1","resolution":{"observed_at":"2026-08-07T13:20:27.249098Z","resolver_source":"raw_fallback","status":"parse_uncertain"},"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":"2026-08-07T13:20:26.884949Z","title":"I-vit: Integer-only quantization for efficient vision transformer inference,","venue":null,"work_id":"322eb5f4-c016-4840-9657-053152f63b0c","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:16.906488Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:9412ef090cda1c3bd6cd059f85271aecee5d74c3b98afff4a68150cc14142a33","observation_id":"c2730087-e6d4-4500-afea-6caf382f9889","resolution":{"observed_at":"2026-08-07T13:20:26.990389Z","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":"2026-08-07T13:20:26.610378Z","title":"Packqvit: Faster sub-8-bit vision transformers via full and packed quantization on the mobile,","venue":null,"work_id":"0e281320-4c48-4d8e-ae81-04f1e47f29d6","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:17.124740Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:bcd410f7bc95659cc8eb1d45d317dc0699bc8a6ecd03f6ac193fd8592b0883da","observation_id":"1f0dd621-5734-4f90-97fa-340760f7ab64","resolution":{"observed_at":"2026-08-07T13:20:26.731802Z","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":"2026-08-07T13:20:26.366345Z","title":null,"venue":null,"work_id":"a385671c-65c1-4af3-a5f7-7834af3e707a","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:17.191838Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:76811397e4dd6b9b0ab2cfc29e69ab7d6649f257f7614b2df1f68e08322741ea","observation_id":"a934d7d2-87b8-4405-b442-a21b150bad02","resolution":{"observed_at":"2026-08-07T13:20:26.483776Z","resolver_source":"raw_fallback","status":"unresolved"},"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":"2026-08-07T13:20:26.141863Z","title":null,"venue":null,"work_id":"333ad293-53f0-45d8-b6d4-589fd7d5ae08","year":2025},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:17.325826Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:a0fe443446100e48ca00039e688da5afb5de6c4db08f0aa21e6f3719724a7d8b","observation_id":"4712e211-b82f-4160-ba88-97d83826e801","resolution":{"observed_at":"2026-08-07T13:20:26.224611Z","resolver_source":"raw_fallback","status":"unresolved"},"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":"2026-08-07T13:20:25.835739Z","title":"Towards accurate binary convolutional neural network,","venue":null,"work_id":"b9fa10b9-a374-4531-ac25-87fd45d54ff7","year":2017},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:17.431220Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:57ca8042b484c66f30b718e2205dcea0ecb30f167e9627378f91930722e5ea72","observation_id":"9032c207-d2cd-45b8-8ede-268ff8e141ca","resolution":{"observed_at":"2026-08-07T13:20:25.977907Z","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":"2026-08-07T13:20:25.671289Z","title":"Lq-nets: Learned quantization for highly accurate and compact deep neural networks,","venue":null,"work_id":"d219e7f4-2322-4777-9382-3165cd3da521","year":2018},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:17.476566Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:829982812ac843a3b75000d93080635d0d96bc3613edbfd3fc23801983a8edd9","observation_id":"ccb3feee-360d-4434-aa1e-0a2db0d53319","resolution":{"observed_at":"2026-08-07T13:20:25.740459Z","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":"1802.04680","last_updated":"2018-02-13T15:23:20Z","snapshot_observed_at":"2026-07-06T06:23:11.958112Z","submitted_at":"2018-02-13T15:23:20Z","title":"Training and Inference with Integers in Deep Neural Networks","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1802.04680","snapshot_observed_at":"2026-08-07T13:20:17.550384Z","title":"Training and inference with integers in deep neural networks,","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:17.550384Z"},"links":{"cited_paper":"/paper/1802.04680","citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:ba79b69fc6fc843385ab42b4b7d3dc245254af2e3ad7322e1963d1ca801567fe","observation_id":"37a05b00-9283-4024-9987-4eb4d9ec0293","resolution":{"observed_at":"2026-08-07T13:20:17.550384Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1806.08342","last_updated":"2018-06-21T17:32:46Z","snapshot_observed_at":"2026-07-06T06:46:08.396066Z","submitted_at":"2018-06-21T17:32:46Z","title":"Quantizing deep convolutional networks for efficient inference: A whitepaper","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1806.08342","snapshot_observed_at":"2026-08-07T13:20:17.699844Z","title":"Quantizing deep convolutional networks for effi- cient inference: A whitepaper,","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:17.699844Z"},"links":{"cited_paper":"/paper/1806.08342","citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:d8b7bb13462268943b6e35403860b9f10db87b4b721458fb6668933e9ed49b0e","observation_id":"407a4a88-9a1c-4882-962d-3d9049ce74ed","resolution":{"observed_at":"2026-08-07T13:20:17.699844Z","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":"2026-08-07T13:20:17.884330Z","title":"Vs-quant: Per-vector scaled quantization for accurate low-precision neural network inference,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:17.884330Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:3d67f61ea8ce9ec01096c2318abcb1431152546dbbf18e63fe1648c342f59840","observation_id":"422d7bdb-d9ca-4a2a-97cc-9896d9604b23","resolution":{"observed_at":"2026-08-07T13:20:17.884330Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2211.10737","last_updated":"2024-05-31T14:47:25Z","snapshot_observed_at":"2026-08-05T01:22:47.382499Z","submitted_at":"2022-11-19T16:17:11Z","title":"Accuracy Booster: Enabling 4-bit Fixed-point Arithmetic for DNN Training","version":4},"cited_work":{"arxiv_id":"2211.10737","doi":null,"metadata_source":"pith","pith_arxiv_id":"2211.10737","snapshot_observed_at":"2026-08-07T13:20:20.955747Z","title":"Accuracy Booster: Enabling 4-bit Fixed-point Arithmetic for DNN Training","venue":"cs.LG","work_id":"42e1d4ad-3867-45c6-92f9-f8680ac85104","year":2022},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:17.973388Z"},"links":{"cited_paper":"/paper/2211.10737","citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:da7845ed6ec04737e0604755e21aff32024ecdd40ef69a49da42d8799674712f","observation_id":"af4683c7-d53f-41a3-9408-7896a35855db","resolution":{"observed_at":"2026-08-07T13:20:21.035673Z","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":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-07T13:20:25.430108Z","title":"With shared microexponents, a little shifting goes a long way,","venue":null,"work_id":"4d4a0b3a-c72b-4f3c-90cc-b188f1beeff1","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:18.059549Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:67c0ba356a1730a986e532c093c8570fd0d8df5bd7632de2fb8049594e29ae72","observation_id":"94b9584b-dfb4-4ca5-bad2-bae5fce029ae","resolution":{"observed_at":"2026-08-07T13:20:25.557938Z","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":"2026-08-07T13:20:25.167829Z","title":"Going further with winograd convolutions: Tap-wise quantization for efficient inference on 4x4 tiles,","venue":null,"work_id":"37184c7c-90a4-4a63-a610-83f5d6588295","year":2022},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:18.177524Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:dd75fb373c73265a035740e9faabdc04d1cd95ccbb27eda0a4683b7a7c89526d","observation_id":"b6acd6af-d94f-4653-9e60-515c52b25e04","resolution":{"observed_at":"2026-08-07T13:20:25.293868Z","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":"2026-08-07T13:20:24.996315Z","title":"Drq: dynamic region-based quantization for deep neural network ac- celeration,","venue":null,"work_id":"10aa2f6f-4076-41f9-811f-344b0158d429","year":2020},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:18.341062Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:f4c2b9549c6098522bab7ceff32076fb01e1029469937b344d989133527a250f","observation_id":"ccb85a02-28b9-4d80-91c8-f5c55bc49ea0","resolution":{"observed_at":"2026-08-07T13:20:25.054370Z","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":"2026-08-07T13:20:24.829437Z","title":"Mokey: Enabling narrow fixed-point inference for out-of-the-box floating-point transformer models,","venue":null,"work_id":"0fdad97a-fad6-437f-8ae3-8bbf010fe484","year":2022},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:18.522360Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:3f595c720ffff52fb488a160ce24e05e12e34f034680f0278c103933f37eb6b5","observation_id":"ee0f18cc-6ed1-47db-b2b3-2cae6d51fc44","resolution":{"observed_at":"2026-08-07T13:20:24.880913Z","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":"2026-08-07T13:20:24.573457Z","title":"Cambricon-q: A hybrid architecture for efficient training,","venue":null,"work_id":"cb36e60b-bf4f-4621-af2e-1a6ab6eaf8e9","year":2021},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:18.684576Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:c103e388eaad4246c4e33c9b68a164bca495a71e5dbac23fa70400af397c1115","observation_id":"964d1f2e-b7b2-410d-8bd2-68143040524a","resolution":{"observed_at":"2026-08-07T13:20:24.709085Z","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":"2208.07339","last_updated":"2022-11-10T18:14:31Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2022-08-15T17:08:50Z","title":"LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2208.07339","snapshot_observed_at":"2026-08-07T13:20:18.868300Z","title":"Llm. int8 (): 8-bit matrix multiplication for transformers at scale,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:18.868300Z"},"links":{"cited_paper":"/paper/2208.07339","citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:869e58f5fea652a584036129098de14984030e4e5f05df6aa0ba5ac89b37e89e","observation_id":"a0abf53c-ee62-4502-9e30-561a023d4cab","resolution":{"observed_at":"2026-08-07T13:20:18.868300Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2210.17323","last_updated":"2023-03-22T13:10:47Z","snapshot_observed_at":"2026-08-07T08:38:54.025062Z","submitted_at":"2022-10-31T13:42:40Z","title":"GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2210.17323","snapshot_observed_at":"2026-08-07T13:20:19.087687Z","title":"Gptq: Accurate post-training quantization for generative pre-trained transformers,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:19.087687Z"},"links":{"cited_paper":"/paper/2210.17323","citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:e6254fa2b4e4d4075a14d98f15c6be4674083956973bec5ce3c0db3a5f774ad0","observation_id":"d31152a0-8288-4190-922b-694a8fb847c2","resolution":{"observed_at":"2026-08-07T13:20:19.087687Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2211.10438","last_updated":"2024-03-29T19:21:58Z","snapshot_observed_at":"2026-08-07T09:04:32.994880Z","submitted_at":"2022-11-18T18:59:33Z","title":"SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models","version":7},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2211.10438","snapshot_observed_at":"2026-08-07T13:20:19.242196Z","title":"Smoothquant: Accurate and efficient post-training quantization for large language models,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:19.242196Z"},"links":{"cited_paper":"/paper/2211.10438","citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:5a1cfa6369d6ea727f6fa87af61a92aabb4a7146342153f3c5e6b74f55c52553","observation_id":"d37599af-d40b-4cd4-8737-7db987586645","resolution":{"observed_at":"2026-08-07T13:20:19.242196Z","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":"2026-08-07T13:20:24.329408Z","title":"Zeroquant: Efficient and affordable post-training quantization for large- scale transformers,","venue":null,"work_id":"79651d74-27d7-4f40-b6f2-9f94ba4c189a","year":2022},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:19.373249Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:860bb9463c2c162542e2023a791a5f027cb2468bc734e8d40ebab750825dd043","observation_id":"4089cd28-7441-4add-a65f-bc84b86a7b12","resolution":{"observed_at":"2026-08-07T13:20:24.445910Z","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":"2026-08-07T13:20:24.070578Z","title":"Psq: An automatic search framework for data-free quantization on pim-based architecture,","venue":null,"work_id":"4703f2b0-2000-4164-86bf-f82ed894f428","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:19.494368Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:9224bc80f569ce38d84c3447c713f218c635b0497bec98dbcbf77a4fd8403ab5","observation_id":"e532194a-73d2-4e55-b5b3-0da2eba9ab79","resolution":{"observed_at":"2026-08-07T13:20:24.160943Z","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":"2026-08-07T13:20:23.876870Z","title":"Spark: Scalable and precision-aware acceleration of neural networks via efficient encoding,","venue":null,"work_id":"a6f9eee2-2711-4b07-9870-3bbfd7b8d953","year":2024},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:19.653899Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:509271766eab9de4c1ba27058eeea5d08a98884e760a4ccd5dfd4c1c0f4d89e4","observation_id":"a995ed6f-0779-4761-b0e0-df917764de87","resolution":{"observed_at":"2026-08-07T13:20:23.972753Z","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":"2026-08-07T13:20:23.173021Z","title":"Msd: Mixing signed digit representations for hardware-efficient dnn accelera- tion on fpga with heterogeneous resources,","venue":null,"work_id":"7844fd0f-73c0-40e3-a80d-87f300a2b5b5","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:19.798048Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:a592050dba035a10bba216fe2b7db10024802010711379f5d8020e1d50181af9","observation_id":"842aa567-9718-4b9e-9ec9-fb036c237450","resolution":{"observed_at":"2026-08-07T13:20:23.728158Z","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":"2026-08-07T13:20:22.803715Z","title":"Adaptable butterfly accelerator for attention-based nns via hardware and algorithm co-design,","venue":null,"work_id":"30b1854a-5f1e-4eb5-8f26-5899fb7cb389","year":2022},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:19.937800Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:c13f57d5ba768fdb1e45748b76a28175070a4d306e460d643cbb5a99e4dc0f59","observation_id":"3df7a2ae-3689-45ae-89df-52fd5ae9de63","resolution":{"observed_at":"2026-08-07T13:20:22.926207Z","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":"2026-08-07T13:20:22.658063Z","title":"Aˆ 3: Accelerating attention mechanisms in neural networks with approximation,","venue":null,"work_id":"0bfbd0fe-98e9-48b5-99c0-b641a1e00eaf","year":2020},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:20.050633Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:4e784fe91f9d8bfb01203e0b4053b77fcff8faecf1e56a85a241ff4be3090415","observation_id":"a4c4702f-6e33-447d-a5f4-de7bb2619b3b","resolution":{"observed_at":"2026-08-07T13:20:22.704664Z","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":"2026-08-07T13:20:22.517734Z","title":"Dfx: A low-latency multi-fpga appliance for accelerating transformer-based text generation,","venue":null,"work_id":"d2e27bac-67f4-48c2-a0aa-9987b60a08ab","year":2022},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:20.174741Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:304c54eaead256e2157e3178c5a2a42b289ed082fbcffe0632f7ae311d55f362","observation_id":"5b6e130a-1f18-44a6-b621-6f3ee4ead198","resolution":{"observed_at":"2026-08-07T13:20:22.593603Z","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":"2026-08-07T13:20:22.357421Z","title":"Flat: An optimized dataflow for mitigating attention bottlenecks,","venue":null,"work_id":"a8da81f0-5e75-404c-b229-125758502c2b","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:20.312317Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:6ce7495eb3439250d2a52d1bb48cba86b3d58020bc2ff2775cb9bba8e72e1191","observation_id":"b45650bb-52a9-40ae-8188-3cb9c7c13eae","resolution":{"observed_at":"2026-08-07T13:20:22.411976Z","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":"2026-08-07T13:20:22.221882Z","title":"Sanger: A co-design framework for enabling sparse attention using reconfigurable architecture,","venue":null,"work_id":"62a855ca-c356-48ac-b836-e0332ae5dbb4","year":2021},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:20.421232Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:9973d5ed47c760422fa63c9a33c12408ab2402cdd0b5b91c87848987763ec4ac","observation_id":"11413356-36d0-4d8e-8aea-4b57b13254fb","resolution":{"observed_at":"2026-08-07T13:20:22.290465Z","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":"2026-08-07T13:20:22.033749Z","title":"Gobo: Quan- tizing attention-based nlp models for low latency and energy efficient inference,","venue":null,"work_id":"fe3fc278-0a11-4ca4-962c-4de6bdbb79fc","year":2020},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:20.557466Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:8ce90ed33d0514d9b4ab1f40f6e1c55c0eca82d25c58125638b4ee708b4cde7d","observation_id":"5bd772ea-e322-411f-9197-bfed3fed9ce2","resolution":{"observed_at":"2026-08-07T13:20:22.099546Z","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":"2026-08-07T13:20:21.800568Z","title":"Edgebert: Sentence-level energy optimizations for latency-aware multi- task nlp inference,","venue":null,"work_id":"b8521744-c401-40ac-acbe-bba7b5aec69a","year":2021},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:20.686498Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:58d13c3b01e3caaec1ece53987b0f2dd39015a3493b73afd1082ee5a3e752fad","observation_id":"10fa9f0b-4741-4267-aabd-a4e4ce315d34","resolution":{"observed_at":"2026-08-07T13:20:21.918681Z","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":"2026-08-07T13:20:21.555290Z","title":"Fact: Ffn-attention co-optimized transformer architecture with eager correlation prediction,","venue":null,"work_id":"b87cbcb9-ae82-4dc2-bec2-bb53a838df06","year":2023},"citing_paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs","version":2},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-07T13:20:20.761016Z"},"links":{"citing_paper":"/paper/2505.22194"},"observation_digest":"sha256:40c579b2c85366c2bdcc23afa7fb55564d02d95a3843f63857d77ce0bfa23c76","observation_id":"9c750e4a-9a59-4d1d-b10f-6704424c44ad","resolution":{"observed_at":"2026-08-07T13:20:21.652869Z","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"}}],"paper":{"arxiv_id":"2505.22194","last_updated":"2025-06-15T10:28:09Z","latest_version":2,"primary_category":"cs.AR","snapshot_observed_at":"2026-08-07T13:10:34.703488Z","submitted_at":"2025-05-28T10:15:37Z","title":"Refining Datapath for Microscaling ViTs"},"reference_resolution":{"displayed":50,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":2,"unresolved":11,"verified_exact":2,"verified_fuzzy":35},"total_outbound_references":50},"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 8 August 2026, this Paper Citation Record lists 50 of 50 outbound references and 0 inbound Pith citation observations for arXiv:2505.22194."}