{"as_of":"2026-08-13T10:11:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:f68bb791b0f13c0bf4f0bd5e843da8776fe60f9c957e02916f48352bb5e31e28","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":"2026-08-11T17:26:53.877030Z","state":"measured"},{"denominator":31,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":31,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-13T06:32:02.005865+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/2608.09444/citation-record","integrity":"/paper/2608.09444/integrity","json":"/paper/2608.09444/citation-record.json","paper":"/paper/2608.09444"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-11T17:26:53.789105Z","title":"The case for co-designing model architectures with hardware","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.789105Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:60fe4e31a3071bf2df9691bc28bc94fce6d2b8ed8ac68f70b405e28e864fe97f","observation_id":"07d25cd3-f1e5-4507-8412-848e36fd7d10","resolution":{"observed_at":"2026-08-11T17:26:53.789105Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.20672","last_updated":"2025-02-28T16:44:24Z","snapshot_observed_at":"2026-08-12T22:14:08.632340Z","submitted_at":"2024-10-28T02:15:45Z","title":"Relaxed Recursive Transformers: Effective Parameter Sharing with Layer-wise LoRA","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.20672","snapshot_observed_at":"2026-08-11T17:26:53.794727Z","title":"Relaxed recursive transformers: Effective parameter sharing with layer-wise lora, 2025 a","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":2,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.794727Z"},"links":{"cited_paper":"/paper/2410.20672","citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:116e9ddd6aab673a50fbcfe6959e9a0dc7c523b6c1db3c525e2fc8f696632a6e","observation_id":"088afc54-22b2-4fff-90dc-7b7c51bc9ccc","resolution":{"observed_at":"2026-08-11T17:26:53.794727Z","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":"2026-08-11T17:26:53.798373Z","title":"Mixture-of-recursions: Learning dynamic recursive depths for adaptive token-level computation","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.798373Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:c00642972baf3ac34c6c606a2a7a48de68c19943b08d09329725915706df6484","observation_id":"d2b8e30d-b719-4bef-90af-b6b593ac74b4","resolution":{"observed_at":"2026-08-11T17:26:53.798373Z","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":"2026-08-11T17:26:53.801140Z","title":"Pondernet: Learning to ponder","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.801140Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:37f04602c9a73337a1bbc9ade74ba0c1206522bcd045ad6a461f35f4fff10e58","observation_id":"bff012b2-d1d8-4338-8f49-b6fb8497de2f","resolution":{"observed_at":"2026-08-11T17:26:53.801140Z","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":"2026-08-11T17:26:54.420729Z","title":"Scaling laws meet model architecture: Toward inference-efficient LLM s","venue":null,"work_id":"0ef778a1-2b9a-45f6-8612-d24bdec080a0","year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.804088Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:79d3973e82cf3e81bb18d2dcceccfe30cf4748065ff11177f09a09d615a30932","observation_id":"753741f2-42f6-4fcb-bd56-86db4b8e597f","resolution":{"observed_at":"2026-08-11T17:26:54.423545Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2605.10661","last_updated":"2026-05-11T14:43:36Z","snapshot_observed_at":"2026-08-12T14:37:44.567991Z","submitted_at":"2026-05-11T14:43:36Z","title":"bViT: Investigating Single-Block Recurrence in Vision Transformers for Image Recognition","version":1},"cited_work":{"arxiv_id":"2605.10661","doi":null,"metadata_source":"pith","pith_arxiv_id":"2605.10661","snapshot_observed_at":"2026-08-11T17:26:54.276410Z","title":"bViT: Investigating Single-Block Recurrence in Vision Transformers for Image Recognition","venue":"cs.CV","work_id":"18f87887-8cf0-4b68-89ab-f6c7054435b8","year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.806969Z"},"links":{"cited_paper":"/paper/2605.10661","citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:54e49b396453251eaaafba28dd94a1e8d2ad51d114c15b2953887bf29cd80c7f","observation_id":"df33bc29-1644-4ad8-ae69-6dbb772d0563","resolution":{"observed_at":"2026-08-11T17:26:54.279659Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.14168","last_updated":"2021-11-18T00:23:45Z","snapshot_observed_at":"2026-08-07T01:45:38.840969Z","submitted_at":"2021-10-27T04:49:45Z","title":"Training Verifiers to Solve Math Word Problems","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.14168","snapshot_observed_at":"2026-08-11T17:26:53.810250Z","title":"Training verifiers to solve math word problems, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.810250Z"},"links":{"cited_paper":"/paper/2110.14168","citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:e5f03b3b0b395cc42526bc2874a4f19c88eff9ad426bdf29427bdd578a56e491","observation_id":"02c02286-f084-4583-805a-7d37db453be2","resolution":{"observed_at":"2026-08-11T17:26:53.810250Z","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":"2026-08-11T17:26:54.412814Z","title":"Adaptive loops and memory in transformers: Think harder or know more? In Workshop on Latent & Implicit Thinking Going Beyond CoT Reasoning , 2026","venue":null,"work_id":"b24ec646-f48f-48d8-8ae5-755b812bc468","year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.813239Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:3e102affe04dd184c375e013702c5d52254ce2d2d6a828b99037269e1f449463","observation_id":"2c420b19-2e83-49f5-80f8-32768cc50ddf","resolution":{"observed_at":"2026-08-11T17:26:54.415921Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2511.08577","last_updated":"2026-04-26T03:36:00Z","snapshot_observed_at":"2026-08-11T04:11:06.503601Z","submitted_at":"2025-11-11T18:57:02Z","title":"Think-at-Hard: Selective Latent Iterations to Improve Reasoning Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2511.08577","snapshot_observed_at":"2026-08-11T17:26:53.815625Z","title":"Think-at-hard: Selective latent iterations to improve reasoning language models, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.815625Z"},"links":{"cited_paper":"/paper/2511.08577","citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:e0730e54c01702c8b9c17f6f3790157030886a3f8dfca42636ff31821af9af98","observation_id":"ef472a37-498d-428b-8ac5-cb4bde71c066","resolution":{"observed_at":"2026-08-11T17:26:53.815625Z","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":"2026-08-11T17:26:54.405482Z","title":"Bartoldson, Bhavya Kailkhura, Abhinav Bhatele, and Tom Goldstein","venue":null,"work_id":"979e8618-bc14-480c-bf96-1211b98326fb","year":2025},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.818788Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:e21fccaa2c51fb1d8f24e8cc4fa8c917b896e9720c3c08ca7d62c842ffc50f43","observation_id":"d6853c67-7e6d-4dc3-a47e-feba7b49319c","resolution":{"observed_at":"2026-08-11T17:26:54.408249Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-11T17:26:53.821396Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.821396Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:cd18aaf15364f863a5dca06e3d2a857da3a59e54728f7ff800df506cefb306b9","observation_id":"92fafeda-e169-4a4a-b9e8-f36376cb7f6f","resolution":{"observed_at":"2026-08-11T17:26:53.821396Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2604.07822","last_updated":"2026-08-11T05:59:57Z","snapshot_observed_at":"2026-08-13T09:21:33.174297Z","submitted_at":"2026-04-09T05:24:32Z","title":"Loop, Think, & Generalize: Implicit Reasoning in Recurrent-Depth Transformers","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2604.07822","snapshot_observed_at":"2026-08-11T17:26:53.824303Z","title":"Loop, think, & generalize: Implicit reasoning in recurrent-depth transformers, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":12,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.824303Z"},"links":{"cited_paper":"/paper/2604.07822","citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:977312b276ed61e672f8cf18c13dc59259d40f23740a4cc2595929aebd5a421d","observation_id":"4eaf625f-7734-4716-b8cd-d91535f716f2","resolution":{"observed_at":"2026-08-11T17:26:53.824303Z","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":"2026-08-11T17:26:53.827094Z","title":"Efficient memory management for large language model serving with pagedattention","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.827094Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:901bc5717b70a29be708a58cad1fdfb8facb51fb8633b3f8d93a4ebecfb0d75c","observation_id":"450f3610-3738-4148-b99e-f309e2ccd511","resolution":{"observed_at":"2026-08-11T17:26:53.827094Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2605.09165","last_updated":"2026-06-30T06:15:20Z","snapshot_observed_at":"2026-08-11T15:12:03.236118Z","submitted_at":"2026-05-09T20:58:18Z","title":"Sparse Layers are Critical to Scaling Looped Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2605.09165","snapshot_observed_at":"2026-08-11T17:26:53.829565Z","title":"Hu, and Jonathan May","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.829565Z"},"links":{"cited_paper":"/paper/2605.09165","citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:9c3e9ccb1e56811e95954ff23c014019c6aa6c69089896cf3d3c27b67f0b8a0a","observation_id":"5d5002f9-3e51-4bb0-8e9d-c5940efec03f","resolution":{"observed_at":"2026-08-11T17:26:53.829565Z","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":"2603.02023","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-11T17:26:54.187402Z","title":"Ponderlm-3: Adaptive token-wise pondering with differentiable masking, 2026","venue":null,"work_id":"a04f1971-4353-4b1d-b89c-628597944668","year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.832304Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:152854c0ec643bde432418c5e7053a94b87a56fb20b225de02e2d8a7ee22990b","observation_id":"8eae1555-8ed8-4101-a38c-cf743b1ee22e","resolution":{"observed_at":"2026-08-11T17:26:54.192954Z","resolver_source":"raw_fallback","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-11T17:26:54.393238Z","title":"Co TF ormer: A chain of thought driven architecture with budget-adaptive computation cost at inference","venue":null,"work_id":"00a6e6ae-00b0-412d-a153-88f9f97ce11d","year":2025},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.834694Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:5dd7c0ba4fd31d9d2a94e3d722b5d42a603f8b4b9667b995841807b09b18bee2","observation_id":"75ff262c-039a-48c8-845d-0da0641bcc9b","resolution":{"observed_at":"2026-08-11T17:26:54.396219Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2404.02258","last_updated":"2024-04-02T19:28:11Z","snapshot_observed_at":"2026-07-06T17:54:47.689340Z","submitted_at":"2024-04-02T19:28:11Z","title":"Mixture-of-Depths: Dynamically allocating compute in transformer-based language models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.02258","snapshot_observed_at":"2026-08-11T17:26:53.837099Z","title":"Mixture-of-depths: Dynamically allocating compute in transformer-based language models, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.837099Z"},"links":{"cited_paper":"/paper/2404.02258","citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:9dcc1a94b39390de591e2da6b5fbd2ca098c01991120c54ea042591b2e01b71b","observation_id":"ad71a3a8-e3a2-41fd-8f74-0acaaad2c10f","resolution":{"observed_at":"2026-08-11T17:26:53.837099Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2604.21106","last_updated":"2026-05-07T08:18:20Z","snapshot_observed_at":"2026-07-06T23:07:47.244208Z","submitted_at":"2026-04-22T21:51:11Z","title":"How Much Is One Recurrence Worth? Iso-Depth Scaling Laws for Looped Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2604.21106","snapshot_observed_at":"2026-08-11T17:26:53.839937Z","title":"How much is one recurrence worth? iso-depth scaling laws for looped language models, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.839937Z"},"links":{"cited_paper":"/paper/2604.21106","citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:4ac78f880115c332ce61a3c2132eed9ad38c57c3bdb1a755b165cdbbd1b7665d","observation_id":"a74457df-b3f4-44f8-9647-9986ca738d81","resolution":{"observed_at":"2026-08-11T17:26:53.839937Z","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":"2026-08-11T17:26:53.842642Z","title":"Flashattention-3: Fast and accurate attention with asynchrony and low-precision","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":19,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.842642Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:e2d59323d0eabefa6db96b3e9739cb71c70b2b99024fcbe6f9b283e87f235dfe","observation_id":"a960c856-479f-4e8b-a928-f084c9dfbdeb","resolution":{"observed_at":"2026-08-11T17:26:53.842642Z","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":"2026-08-11T17:26:54.382172Z","title":"ShareGPT , 2023","venue":null,"work_id":"00e7b35e-ec41-418e-802f-3c20f782f9d2","year":2023},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.845224Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:d88f0c40f526a3decc1cfa3d6ddfa7f91910dd5b51ab0c2cde0ca7050f96fa32","observation_id":"45a25439-f503-460a-a12a-aba2e95c9d48","resolution":{"observed_at":"2026-08-11T17:26:54.384847Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-11T17:26:53.848072Z","title":"Loopvit: Scaling visual arc with looped transformers, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.848072Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:151382b775fa97f699a40e8a6e9ef3fa07816ba282f3acacddd47061ac074d01","observation_id":"9dbee458-2f5d-47d6-8dfd-6456401816e7","resolution":{"observed_at":"2026-08-11T17:26:53.848072Z","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":"2026-08-11T17:26:53.850843Z","title":"Adaponderlm: Gated pondering language models with token-wise adaptive depth, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.850843Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:63720e2832363e5616dec12a35a614d8511a2e78ded5da71fbc5db5f0793ad03","observation_id":"a716a677-3d23-405d-9393-7f77573d4d87","resolution":{"observed_at":"2026-08-11T17:26:53.850843Z","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":"10.5281/zenodo.20122284","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-11T17:26:53.896425Z","title":"Michaelov, Chris, Chessing234, Hanwool Albert Lee, Janna, Leonid Sinev, Khalid, Kiersten Stokes, and Zdeněk Kasner","venue":null,"work_id":"d0e9e482-0c63-4497-a02a-52a05ba873f8","year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.853674Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:9f10f322582a33315973b21d89de2c12ef0323b26b59578a1b68ffcfd77140f1","observation_id":"20caf359-8308-4ce8-bce2-b44e9d6b9365","resolution":{"observed_at":"2026-08-11T17:26:53.900773Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-11T17:26:53.856773Z","title":"Sparse universal transformer","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.856773Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:6c9895999936d2f75ec1d8999b30c7db0d7399236f43e9bd4d96dc031067b965","observation_id":"74c28abf-0699-4ca1-8df6-c6dd9b9748f8","resolution":{"observed_at":"2026-08-11T17:26:53.856773Z","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":"2026-08-11T17:26:54.371002Z","title":"Hashimoto","venue":null,"work_id":"4c809b8e-5484-469e-b40b-17dec4c9d24c","year":2023},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.859773Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:2b8411f27048aa4234eb53ad41395a9cd6ee8e74dc23d1bc6245624c4a7b9c26","observation_id":"0391ffa2-8574-4df9-82e1-02603bc55cc5","resolution":{"observed_at":"2026-08-11T17:26:54.373762Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+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":"2026-08-11T17:26:54.363399Z","title":"Recurrent-depth VLA : Implicit test-time compute scaling of vision-language-action models via latent iterative reasoning","venue":null,"work_id":"c4f87794-1486-4b2d-9f0c-be0e96ec07c4","year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.862780Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:8b8672a0e147b6365486fe8459cd505ceb7beb189bd40d4fcacd96ff835a2427","observation_id":"20baffb5-6d18-4dd5-8451-7f5a8354d8d2","resolution":{"observed_at":"2026-08-11T17:26:54.366259Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2605.07721","last_updated":"2026-05-19T08:08:32Z","snapshot_observed_at":"2026-08-02T23:35:32.635346Z","submitted_at":"2026-05-08T13:25:27Z","title":"Memory-Efficient Looped Transformer: Decoupling Compute from Memory in Looped Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2605.07721","snapshot_observed_at":"2026-08-11T17:26:53.865582Z","title":"Memory-efficient looped transformer: Decoupling compute from memory in looped language models, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.865582Z"},"links":{"cited_paper":"/paper/2605.07721","citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:0d2065e84bccb0154b828bf280461a4a36f21e48bc2223222a4c87140fb8bc98","observation_id":"a69a7bd4-7729-4b38-aba0-b2b9062f0def","resolution":{"observed_at":"2026-08-11T17:26:53.865582Z","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":"2026-08-11T17:26:53.868554Z","title":"Roofline: an insightful visual performance model for multicore architectures","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.868554Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:d86ee204983b2598ec0eb9f1c6b5c17c9a642ee5ade16feaada2f38bae7760fc","observation_id":"31f6872b-67a5-4257-87d8-92778a46b074","resolution":{"observed_at":"2026-08-11T17:26:53.868554Z","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":"2026-08-11T17:26:53.871322Z","title":"Transformers: State-of-the-art natural language processing","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.871322Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:c91316fa03514bf68ffdcc969356a369d4050c321eef3bfb0ad74ede034ae7bd","observation_id":"86448520-d82a-4a52-b769-286694514a76","resolution":{"observed_at":"2026-08-11T17:26:53.871322Z","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":"2026-08-11T17:26:54.350911Z","title":"Orca: A distributed serving system for Transformer-Based generative models","venue":null,"work_id":"0ecf65e5-6e07-4481-a23b-f6b3f231f6b4","year":2022},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.874188Z"},"links":{"citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:42df5b2276833b77863452e8dc90672a12827e1e64dff63b2bfd517d56b2c192","observation_id":"a57946e7-11f6-4ad6-8262-701b53fa4ca6","resolution":{"observed_at":"2026-08-11T17:26:54.353756Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2510.25741","last_updated":"2026-07-01T23:25:58Z","snapshot_observed_at":"2026-08-04T07:30:51.188041Z","submitted_at":"2025-10-29T17:45:42Z","title":"Scaling Latent Reasoning via Looped Language Models","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2510.25741","snapshot_observed_at":"2026-08-11T17:26:53.877030Z","title":"Scaling latent reasoning via looped language models, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching","version":1},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-08-11T17:26:53.877030Z"},"links":{"cited_paper":"/paper/2510.25741","citing_paper":"/paper/2608.09444"},"observation_digest":"sha256:026cab5b1fe767d204935fb3cd3b5d978c3c989363b2bcef5049301924058490","observation_id":"368f2810-e0ac-4aa5-9843-89cae2bbf036","resolution":{"observed_at":"2026-08-11T17:26:53.877030Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2608.09444","last_updated":"2026-08-10T11:20:14Z","latest_version":1,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-13T09:19:08.961697Z","submitted_at":"2026-08-10T11:20:14Z","title":"Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching"},"reference_resolution":{"displayed":31,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":20,"verified_exact":3,"verified_fuzzy":8},"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":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"thesis":"As of 13 August 2026, this Paper Citation Record lists 31 of 31 outbound references and 0 inbound Pith citation observations for arXiv:2608.09444."}